Animal Migration and InfectiousDisease RiskSonia Altizer,* Rebecca Bartel, Barbara A. Han

Animal migrations are often spectacular, and migratory species harbor zoonotic pathogens ofimportance to humans. Animal migrations are expected to enhance the global spread of pathogensand facilitate cross-species transmission. This does happen, but new research has also shown thatmigration allows hosts to escape from infected habitats, reduces disease levels when infectedanimals do not migrate successfully, and may lead to the evolution of less-virulent pathogens.Migratory demands can also reduce immune function, with consequences for host susceptibility andmortality. Studies of pathogen dynamics in migratory species and how these will respond to globalchange are urgently needed to predict future disease risks for wildlife and humans alike.

Billions of animals from groups as diverseasmammals, birds, fish, and insects under-take regular long-distance movements

each year to track seasonal changes in resourcesand habitats (1). The most dramatic migrations,such as those by monarch butterflies (Fig. 1),gray whales, and some shorebirds and dragon-flies (Fig. 2), span entire continents or hemi-spheres, can take several months to complete,and are accompanied by high energetic demandsand extreme physiological changes. The ultimatecause of these seasonal migrations remains de-bated; most hypotheses focus on avoidance offood scarcity, seeking physiologically optimalclimates, and avoiding predation during periodsof reproduction [e.g., (2)]. Contemporary studiesof migration have uncovered mechanisms of ani-mal navigation, energy budgets, resource use,and phenological responses to environmentalchange; migratory species have also been recog-nized for their potential to connect geographicallydistant habitats and transfer large amounts of bio-mass and nutrients between ecosystems [reviewedin (3)]. These studies illustrate the profound eco-logical and evolutionary consequences of mi-gratory journeys for animal populations on aglobal scale.

Owing to their long-distance movements andexposure to diverse habitats, migratory animalshave far-reaching implications for the emergenceand spread of infectious diseases. Importantly,most previous work on the role of host move-ment in infectious disease dynamics has focusedon spatially localized or random dispersal. Forexample, dispersal events give rise to travelingwaves of infection in pathogens such as raccoonrabies (4), influenza in humans (5), and nuclearpolyhedrosis viruses in insects (6). In the contextof metapopulations, limited amounts of hostmovement could actually prevent host extinction

in the face of a debilitating pathogen and allowhost resistance genes to spread (7, 8). From adifferent perspective, case studies of speciesinvasions demonstrate that one-time transfers ofeven a few individuals can transport pathogenslong distances and introduce them to novel hab-itats (9). Yet relatively few studies have examinedhow regular, directed mass movements that char-acterize seasonal migration affect the transmis-sion and evolution of pathogens within host

populations and the response of migratory spe-cies to infection risks.

In this article, we review the consequences oflong-distance migrations for the ecological dy-namics of host-pathogen interactions and out-line key challenges for future work. Ecologicalprocesses linked with migration can increase ordecrease the between-host transmission of patho-gens, depending on host migratory behavior andpathogen traits (Fig. 3).Moreover, newwork showsthat for some species, the energetic demands ofmigration compromise host immunity, possiblyincreasing susceptibility to infection and intensi-fying the impacts of disease. Importantly, manymigratory species are at risk of future declinesbecause of habitat loss and exploitation, and animalmigrations are shifting with ongoing anthropo-genic change (10). Thus, understandinghowhuman

activities that alter migratorypatterns influencewildlife-pathogendynamics is urgently needed tohelp guide conservation and man-agement of migratory species andmitigate future risks from infec-tious disease.

What Goes Around ComesAround: Pathogen Exposureand Spatial Spread

Anoft-cited but poorly supportedassumption is that long-distancemovements of migrating animalscan enhance the geographic spreadof pathogens, including zoonot-ic pathogens important for hu-man health such as Ebola virusin bats, avian influenza virusesin waterfowl and shorebirds, andLyme disease andWest Nile virus(WNV) in songbirds. For exam-ple,WNVinitially spread inNorthAmerica along a major corridorfor migrating birds and rapidlyexpanded from its point of origininNewYorkCityalong theAtlanticseaboard from 1999 to 2000 (11).Although experimental workconcluded that passerine birds inmigratory condition were com-petent hosts for WNVand couldserve as effective dispersal agents(12), evidence to show that thisexpansion resulted from move-ments of migratory birds remains

equivocal. For the zoonotic pathogen Ebola virus,a recent study points to the coincident timing of anannual influx of migratory fruit bats in theDemocratic Republic of Congo and the start ofhuman Ebola outbreaks in local villages during2007 (13). In central Kazakhstan, saiga antelopes(Saiga tatarica) become infected with gastro-intestinal nematodes (Marshallagia) during thecourse of seasonal migration by grazing onpastures used by domesticated sheep earlier in

REVIEW

Odum School of Ecology, University of Georgia, Athens, GA30602, USA.

*To whom correspondence should be addressed. E-mail:saltizer@uga.edu

Fig. 1. Monarch butterflies (Danaus plexippus), shown here at awintering site in central Mexico, undertake one of the longest dis-tance two-way migrations of any insect species worldwide. Mon-archs are commonly infected by a debilitating protozoan parasitethat can lower the flight ability of migrating butterflies.

21 JANUARY 2011 VOL 331 SCIENCE www.sciencemag.org296

on

Janu

ary

20, 2

011

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

Fig. 2. Representative migratory species, including migration distancesand routes, known parasites and pathogens, and major threats to speciespersistence. Infectious diseases have been examined in the context of

migration for some, but not all, of these species. Supporting refer-ences and photo credits are provided in the supporting online material(SOM) text.

www.sciencemag.org SCIENCE VOL 331 21 JANUARY 2011 297

REVIEW

on

Janu

ary

20, 2

011

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

the season. As migration continues, saiga carryand transmit Marshallagia to northern sheeppopulations, leading to pulses of infection thatcoincide with annual saiga migrations (14).

The potential for serious disease risks forhuman and livestock health has raised alarmabout the role of migratory species in movinginfectious agents to distant locations. Yet surpris-ingly few examples of long-distance pathogendispersal by migrating animals have been clearlydocumented in the published literature, and somestudies indicate that migrants might be unfairlyblamed for transporting pathogens. As a case inpoint, wild waterfowl (Anseriformes) and shore-birds (Charadriiformes) represent the major nat-ural reservoirs for diverse strains of avianinfluenza virus (AIV) worldwide, including thehighly pathogenic (HP) H5N1 subtype that canlethally infect humans (15). Although many ofthese migratory birds can become infected by HPH5N1, recent work incorporating what is knownabout viral shedding period, host migration phe-nology, and the geographical distribution of viralsubtypes suggests that most wild birds are un-likely to spread HP H5N1 long distances (e.g.,between Asia and the Americas) as previouslysuspected [e.g., (16, 17)]. Central to the questionof how far any host species can transport apathogen are the concepts of pathogen virulenceand host tolerance to infection. Specifically, vir-ulence refers to the damage that parasites inflicton their hosts, and tolerance refers to the host’sability to withstand infection without sufferingmajor fitness costs. Thus, host-parasite species orgenotype combinations associated with very lowvirulence or high tolerance should be the mostpromising candidates for long-distance move-ment of pathogen strains, a simple prediction thatcould be explored within migratory species orusing cross-species comparisons.

Beyond their potential role in pathogen spa-tial spread, a handful of studies suggest that mi-gratory species themselves encounter a broaderrange of pathogens from diverse environmentsthroughout their annual cycle compared withspecies residing in the same area year-round(Fig. 3). One field study showed that songbirdspecies migrating from Europe became infectedby strains of vector-borne blood parasites origi-nating from tropical bird species at overwinteringsites in Africa (18), in addition to the suite ofparasite strains transmitted at their summer breed-ing grounds. The authors posited that winter ex-posure to parasites in tropical locations is asignificant cost of migration, because residentspecies wintering in northern latitudes encounterfewer parasite strains and do not experience year-round transmission. Similarly, the number ofparasite species per host was positively related todistances flown by migratory waterfowl (19),indicating that migrating animals could becomeexposed to parasites through encounters with dif-ferent host species and habitat types.

Although some animals undertake nonstopmigrations, most migratory species use stop-

over points along the migration route to restand feed. These stopover points usually occurfrequently along a journey, although some spe-cies like shorebirds fly thousands of kilome-ters between only a handful of staging areas(1). Refueling locations are often shared bymultiple species, and the high local densitiesand high species diversity can increase bothwithin- and between-species transmission ofpathogens. In one of the most striking examplesof this phenomenon, shorebirds such as sander-lings (Calidris alba), ruddy turnstones (Arenariainterpres; Fig. 2), and red knots (Calidris canutus),which migrate annually between Arctic breed-ing grounds and South American winteringsites, congregate to feed in massive numbers

in the Delaware Bay and the Bay of Fundy torebuild fat reserves, leading to upwards of 1.5million birds intermingling, at densities ofover 200 birds per square meter (20). This phe-nomenon creates an ecological hotspot at Dela-ware Bay, where the prevalence of AIV is 17times greater than at any other surveillance siteworldwide (20).

Leaving Parasites Behind: Migrationas a Way of Lowering Infection RiskAlthough greater exposure to parasites and path-ogens can pose a significant cost of long-distancemigration, for some animal species, long-distancemigration will reduce infection risk by at leasttwo nonexclusive processes (Fig. 3). First, if pro-

Breeding grounds

Stopover sites

Stopover sites

Wintering grounds

Springmigration

Fallmigration

Build-up of susceptibles

Aggregation at higher densityMulti-species contacts

Exposure to novel strains

Spa

tial s

prea

d

Spatial spread

Host disseminationonto breeding

grounds

Migratoryculling

Migratoryculling

Migratoryescape

Immunomodulation

Immunomodulation

Secondaryinfections

AAggS

patia

l spr

ead

Stopover sites

hher d de insittyMulti-species contacts

greg tiation tat hihighh

Breeding grounds

B iBuildld-up fof suscep ibtiblles

Wintering grounds

Exposure to novel strains

Fig. 3. Points along a general annual migratory cycle where key processes can increase (red text) ordecrease (blue text) pathogen exposure or transmission. Behavioral mechanisms such as migratory escapeand migratory culling could reduce overall pathogen prevalence. As animals travel to distant geographiclocations, the use of multiple habitat types including stopover sites, breeding areas, and wintering groundscan increase transmission as a result of host aggregations and exposure to multihost pathogens. This mightbe especially true for migratory staging areas where animals stop to rest and refuel. High energeticdemands of spring and fall migration can also result in immunomodulation, possibly leading to immunesuppression and secondary infections. [Photo credits (clockwise): J. Goldstein, B. McCord, A. Friedlaender,and R. Hall]

21 JANUARY 2011 VOL 331 SCIENCE www.sciencemag.org298

REVIEW

on

Janu

ary

20, 2

011

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

longed use of habitats allows parasites with envi-ronmental transmission modes to accumulate(i.e., those parasites with infectious stages thatcan persist outside of hosts, such as many hel-minths, ectoparasites, and microbial pathogenswith fecal-oral transmission), migration will al-low animals to escape from contaminated hab-itats [i.e., “migratory escape” (21)]. Betweenintervals of habitat use, unfavorable conditions(such as harsh winters and a lack of hosts) couldeliminate most parasites, resulting in hostsreturning to these habitats after a long absenceto encounter largely disease-free conditions (21).Empirical support for migratory escape comesfrom a few well-studied host-parasite interac-tions, including research on reindeer (Rangifertarandus), which showed that the abundance ofwarble flies (Hypoderma tarandi) was negative-ly correlated with the distance migrated to sum-mer pastures from reindeer calving grounds (themain larval shedding area in early spring) (22).This observation prompted researchers to suggestthat the reindeers’ annual postcalving migrationreduces warble fly transmission by allowing ani-mals to leave behind areas where large numbersof larvae have been shed (and where adult flieswill later emerge). It is worth noting that escapewill be less successful from pathogens with long-lived infectious stages that persist between pe-riods of host absence or pathogens that causechronic or life-long infections.

Long-distance migration can also lower path-ogen prevalence by removing infected animalsfrom the population [i.e., “migratory culling” (23)].In this scenario, diseased animals suffering fromthe negative consequences of infection are lesslikely to migrate long distances owing to thecombined physiological demands of migrationand infection. Work on the migratory fall army-wormmoth (Spodoptera frugiperda) suggestedthat insects infected by an ectoparasitic nema-tode (Noctuidonema guyanense) had reducedmigratory ability because few to no parasiteswere detected in moths recolonizing sites asthey returned north (24). More recent work onBewick’s swans (Cygnus columbianus bewickii)showed that infection by low-pathogenic avianinfluenza (LPAI) viruses delayed migration overa month and reduced the travel distances ofinfected birds compared with those of healthyindividuals (25). However, a study of AIV inwhite fronted geese did not find any differencein distances migrated between infected and unin-fected birds (26), suggesting that, not surprisingly,some species are better able to tolerate infec-tions during long journeys and raising the pos-sibility that migration could select for greatertolerance to infections in some hosts due to thehigh fitness costs of attempting migration with adebilitating pathogen.

Whether the net effects of migration willincrease or decrease prevalence depends in largepart on the mode of parasite transmission and thelevel of host specificity, both of which will affectopportunities for cross-species transmission at

staging and stopover sites. Parasites that de-cline in response to host migration may includespecialist pathogens, as well as those with trans-mission stages that can build up in the envi-ronment, pathogens transmitted by biting vectorsor intermediate hosts, or for which transmis-

sion occurs mainly from adults to juveniles dur-ing the breeding season (e.g., Box 1). Conversely,migrating hosts could experience higher pres-sure from generalist parasites if opportunitiesfor cross-species transmission are high at stop-over areas or wintering grounds or from special-

0.80

0.60

0.40

0.20

0.00

Pre

vale

nce

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

Breeding

Overwintering

Pre

vale

nce

0.30

0.25

0.20

0.15

0.10

0.05

0.00

2006/20072007/20082008/2009

A

B

C

Box 1. Lessons from a model system: Monarch migration drives large-scale variation inparasite prevalence.

During the past 10 years, we studied monarch butterflies (Danaus plexippus) and a protozoan parasite(Ophryocystis elektroscirrha) (top right images) for the effects of seasonal migration on host-pathogendynamics. Monarchs in eastern North America (A) migrate up to 2500 km each fall from as far north asCanada to wintering sites in Central Mexico (60). Monarchs in western North America (B) migrateshorter distances to winter along the coast of California (61). Monarchs also form nonmigratorypopulations that breed year-round in southern Florida (C), Hawai’i, the Caribbean Islands, and Centraland South America (62). Because monarchs are abundant and widespread and can be studied easilyboth in the wild and in captivity, field and experimental studies can explore effects of annualmigrations on host-pathogen ecology and evolution. A recent continent-scale analysis showed thatparasite prevalence increased throughout the monarchs’ breeding season, with highest prevalenceamong adults associated with more intense habitat use and longer residency in eastern North America,consistent with the idea of migratory escape (bottom right graph) (63). Experiments showed thatmonarchs infected with O. elektroscirrha flew shorter distances and with reduced flight speeds, andfield studies showed parasite prevalence decreased as monarchs moved southward during their fallmigrations (23, 63), consistent with the idea of migratory culling. Parasite prevalence was also highestamong butterflies sampled at the end of the breeding season than for those that reached theiroverwintering sites in Mexico (bottom right graph). Collectively, these processes have likely generatedthe striking differences in parasite prevalence reported among wild monarch populations with differentmigratory behaviors (bottom left graph) (64). Laboratory studies also showed that parasite isolates fromthe longest-distance migratory population in eastern North America (A) were less virulent than isolatesfrom short-distance (B) and nonmigratory (C) populations (55, 65), suggesting that longer migrationdistances cull monarchs carrying virulent parasite genotypes. Work on this model system illustrates howmultiple mechanisms can operate at different points along a migratory cycle and highlights the role thatmigration plays in keeping populations healthy. Monarch migrations are now considered an endangeredphenomenon (60) due to deforestation of overwintering grounds, loss of critical breeding habitats, andclimate-related shifts inmigration phenology. If climate warming extendsmonarch breeding seasons intofall and winter months, migrations may eventually cease altogether. Evidence to date indicates that theloss of migration in response to mild winters and year-round resources could prolong exposure toparasites, elevate infection prevalence, and favor more virulent parasite genotypes. Images reproducedfrom (63, 64). [Photos by S. Altizer]

www.sciencemag.org SCIENCE VOL 331 21 JANUARY 2011 299

REVIEW

on

Janu

ary

20, 2

011

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

ist pathogens if transmission increases withdense host aggregations that accompany massmigrations. Importantly, effects of migrationon pathogen dynamics within host populationsshould translate to large differences in preva-lence across host populations with differentmigratory strategies. Over the past few years,we have focused on monarch butterflies (Dan-aus plexippus) as a model system to study theeffects of migration on host-pathogen inter-actions (Box 1) and found that both migratoryculling and migratory escape can cause spatio-temporal variation in prevalence within popu-lations and extreme differences in prevalenceamong populations with different migratorystrategies. However, we are not aware of intra-specific comparisons of prevalence between mi-gratory and nonmigratory populations for otheranimal species.

Immune Defense Balanced Against theDemands of MigrationIn addition to ecological mechanisms affectingbetween-host transmission, the physiological stressand energetic demands of migration can alter theoutcome of infection within individuals throughinteractions with the host’s immune system (Fig.2). More generally, because several immune path-ways in both vertebrates and invertebrates areknown to be costly (27, 28), seasonal demandssuch as premigratory fattening or strenuous ac-tivity will likely lower the resource pool availablefor mounting an immune response (29). In antic-ipation of migration, for example, some animalsaccrue up to 50% of their lean body mass in fat,increase muscle mass, and atrophy organs that arenot essential during migration (1). Thus, beforemigration, animals might adjust components oftheir immune response to a desired level (i.e.,immunomodulation), or the energetic demandsof migration could reduce the efficacy of someimmune pathways (i.e., immunosuppression).

To date, the effects of long-distance migrationon immune defenses have been best studied inbirds. In a rare study of immune changes in wildindividuals during migration, field observationsof three species of thrushes showed that migrat-ing birds had lower baseline measures for severalcomponents of innate immunity (including leu-kocyte and lymphocyte counts), and exhibitedlower fat reserves and higher energetic stress,relative to individuals measured outside of themigratory season (30). Captive experiments withSwainson’s thrushes (Catharus ustulatus; Fig. 2)later demonstrated that cell-mediated immunitywas suppressed with the onset of migratory rest-lessness (the agitated behavior of birds thatwould normally precede their migratory depar-ture) (31), suggesting that predictable changes inimmunity occur in preparation for long-distanceflight. In this species, the energetic costs of mi-gration can intensify seasonal immune changes:Migrating thrushes that arrived at stopover sitesin poorest condition had the lowest counts ofimmune cells (32).

The extent of altered immunity before andduring migration is likely to be both species- andresource-dependent and will further depend onthe specific immune pathwaymeasured. Red knots,for example, exhibited no change in either anti-body production or cell-mediated immunity afterlong flights in a wind tunnel, a result that arguesagainst migration-mediated immunosuppression(33). Another study of captive red knots revealedno declines in costly immune defenses during theannual periods of mass gain (34); however, ani-mals in this study had constant access to high-quality food, which might have negated energetictrade-offs between immune investment and weightgain. Interestingly, barn swallows (Hirundo rustica)in better physical condition showed greater mea-sures of cellular immunity duringmigration, clearedectoparasites and blood parasites more effectively,and arrived earlier at breeding grounds than birdswith poor energy reserves (35). These studies sug-gest that animals in robust condition or with accessto resources might tolerate long journeys withoutsignificant immunocompromise. Studies of migra-tory species to date also emphasize the need for amore detailed understanding of the mechanismslinking nutrient intake and metabolic activity toinnate and adaptive immunemeasures, a step that isessential to predicting how different immune path-ways will respond to physiological changes thatoccur before and during long-distance migrations.

Perhaps most importantly, immune changesthat accompany long-distance migration couldlead to a relapse of prior infections and moresevere disease following exposure to new path-ogens, increasing the likelihood of migratoryculling and lowering the probability of spatialspread. This possibility was investigated forLyme disease in redwings (Turdus iliacus) (36).Consistent with results showing negative effectsof migratory status on immunity, migratoryrestlessness alone was sufficient to reactivatelatent Borrelia infections in captive birds. Thus,the demands of migration could ultimately leadto more severe infections and greater removal ofinfected hosts. Together, these results point to arole for migration-mediated immune changes inthe dynamics of other wildlife pathogens, includ-ing zoonotic agents such as WNV (12) and bat-transmitted corona and rabies viruses (37, 38).

Effects of Anthropogenic Change and ClimateChanges to the ecology of migratory species inthe past century (Fig. 2) could have enormousimpacts on pathogen spread in wildlife and live-stock, as well as altering human exposure tozoonotic infections. As one example, habitat losscaused by urbanization or agricultural expansioncan eliminate stopover sites and result in higherdensities of animals that use fewer remainingsites along the migration route (10). Although theresulting impacts on infectious diseases remainspeculative, dense aggregations of animals atdwindling stopover sites might create ecologicalhot spots for pathogen transmission among wild-life species, as illustrated in the case of AIV in

migrating shorebirds at Delaware Bay (20). More-over, continuing human encroachment on stop-over habitats increases the likelihood of contactand spillover infection from wildlife reservoirhosts to humans and domesticated species.

For some animal species, physical barrierssuch as fences (terrestrial species) or hydro-electric dams (aquatic species) impede migration(39), leaving animals to choose between navigat-ing a narrow migratory corridor or forming non-migratory populations. Consequently, pathogenprevalence could increase when animals stopmigrating and become confined to smaller hab-itats, if parasite infectious stages build up withmore intense use of a given habitat. Attempts tocontrol cattle exposure to brucellosis from bison(Bison bison) and elk (Cervus elaphus) in theGreater Yellowstone Ecosystem illustrate theserisks. Due to the potential threat of Brucella trans-mission from bison to cattle, bison are routinelyculled if they leave the confines of YellowstoneNational Park (40). Elk migration is less re-stricted, but there is evidence that supplementalfeeding areas encourage the formation of densenonmigratory populations that support higherprevalence of brucellosis, with 10 to 30% sero-prevalence in animals at the feeding groundscompared with 2 to 3% seroprevalence in un-fed elk ranging the park (41). High populationdensities in elk also correlate with higher gastro-intestinal parasite loads at feeding grounds (42),suggesting that high densities of nonmigratinghosts lead to increasing intraspecific transmissionof multiple parasites.

More generally, human activities that dis-courage long-distance animal movements andencourage the formation of local year-roundpopulations can cause the emergence of zoonoticpathogens in humans. For example, human-meditated environmental changes facilitated out-breaks of two zoonotic paramyxoviruses carriedby flying foxes (Pteropus fruit bats; Fig. 2): Theseanimals are highly mobile and seasonally no-madic in response to local food availability (43).Anthropogenic changes such as deforestationand agricultural production likely influenced theemergence of lethal Nipah and Hendra virusoutbreaks in humans in Australia andMalaysia intwo key ways: by resource supplementation andhabitat alteration limiting migratory behaviors offruit bats and by facilitating close contact withdomesticated virus-amplifying hosts (pigs andhorses). In Malaysia, resident flying foxes forag-ing on fruit trees on or near pig farms transmittedNipah virus to pigs, probably via urine or partiallyconsumed fruit with subsequent spread from pigsto humans [(43) and references therein]. Humanactivities are also thought to increase the risk ofHendra virus outbreaks in Australia by drivingflying foxes from formerly forested areas intourban and suburban areas (44), where they formdense nonmigratory colonies that aggregate inpublic gardens containing abundant food sources.

In marine systems, aquaculture increases ex-posure to parasites in wild fish species, partic-

21 JANUARY 2011 VOL 331 SCIENCE www.sciencemag.org300

REVIEW

on

Janu

ary

20, 2

011

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

ularly in salmonids. Migration normally protectswild juvenile salmon from marine parasites incoastal waters by spatially separating them frominfected wild adults offshore (45), but denselypopulated salmon farms place farmed fish enclo-sures adjacent to wild salmon migratory corridors,increasing the transmission of parasitic sea lice(Lepeophtheirus salmonis) to wild juveniles re-turning to sea (45).

Finally, climate change will alter infectiousdisease dynamics in some migratory species (46).To survive, many migratory species must re-spond to climate changes by shifting migratoryroutes and phenology in response totemperature and the availability ofkey resources (i.e., flowering plants,insects) [e.g., (47)]. It is possible thatchanges in the timing of migrationcould disrupt the synchronicity ofhost and parasite life cycles, muchin the way that ecological mismatchin migration timing or altered migra-tory routes could impact whethersuitable food and habitat are avail-able when migrants arrive. For ex-ample, the spawning periodicity ofwhale barnacles in calving lagoonsof gray whales is a classic exampleof a parasite synchronizing itsreproduction to overlap with a host’smigratory cycle (48). If the timing ofwhale migrations and barnaclereproduction shift in response todifferent environmental cues, thiscould result in reduced infectionsover time. On the other hand, alteredmigration routes might facilitatecontact between otherwise geo-graphically separated host species,leading to novel pathogen introduc-tions and increasing disease risks forsome wildlife species (46). One ex-ample of this phenomenon involvesoutbreaks of phocine distemper vi-rus in harbor seals (Phoca vitulina)in the North Sea, which was likelyintroduced by harp seals (Pagophilusgroenlandicus) migrating beyondtheir normal range and contactingharbor seal populations (49). More-over, if climatewarming extends hosts’breeding seasons, migrations maycease altogether, with year-round resident pop-ulations replacing migratory ones (Box 1), leadingto greater pathogen prevalence through a loss ofmigratory culling and escape.

Outlook and Future ChallengesUnderstanding the mechanisms by which long-distance movements affect host-pathogen sys-tems offers exciting challenges for future work,especially in the context of global change andevolutionary dynamics. In terms of basic research,there remains a great need to identify conditionsunder whichmigration will increase host exposure

to infectious agents versus cases where migra-tion can reduce transmission, with the ultimategoal of predicting the net outcomes for host spe-cies where multiple mechanisms operate on thesame or different pathogens (e.g., Box 1). To thatend, mechanistic models are needed to examinehow migration affects infectious disease dynam-ics and to explore the relevance of possible mech-anisms. Suchmodelsmust combine within-seasonprocesses (including host reproduction, over-wintering survival, and pathogen transmission)with between-season migration (Fig. 4). For ex-ample, to examine the importance of environ-

mental transmission for the dynamics of LPAI inNorth American birds, Breban et al. (50) mod-eled a waterfowl population migrating betweentwo geographically distant sites, with transmis-sion dynamics occurring at both breeding andwintering grounds. Similarly, models describinginterconnected networks ofmetapopulations couldbe useful in investigating disease dynamics be-tween habitats linked through seasonal migra-tions (51). Although currently uncommon inthe literature, epidemiological models can alsobe extended to capture mechanisms such as mi-gratory culling andmigratory escape and to include

multiple infectious agents to explore questionsof coinfection and multihost transmission dy-namics (Fig. 4).

One outstanding question is whether parasitescan increase the migratory propensity of theirhosts by favoring the evolution of migratory be-haviors. Long-distance migration has previouslybeen hypothesized to reduce predation risks forungulates and birds, with the general rationalebeing that the survival costs of migration shouldbe outweighed by fitness benefits associated withreproduction. In support of this idea, field studiesof wolf predation on North American elk at their

summer breeding grounds (52) andnest predation on migrating song-birds (2) showed that animals travel-ing farthest experienced the lowestpredation risk. Similar observationalstudies could ask how the preva-lence, intensity, virulence, and diver-sity of key parasites change withmigratory distances traveled. To thatend, comparing infection dynam-ics between migratory and nonmi-gratory populations of the samespecies offers a powerful test of bothpattern and process (e.g., Box 1),although researchers will need tokeep inmind that climate differences(e.g, milder climates for habitatsused by nonmigratory populations)could confound some compari-sons. Modeling approaches are alsoneeded to explore how seasonalmigration might respond evolution-arily to parasite-driven pressures,similar to other studies that exam-ined effects of within-site competi-tion, costs of dispersal, and variationin habitat quality on randomdispersalstrategies (53).

Another question related to hostevolution is whether the combineddemands of migration and diseaserisk could select for greater or lowerinvestment in resistance or immu-nity. Field and laboratory studieshave already documented between-season changes in immune invest-ment, suggesting that somemigratoryspecies suppress specific immuneresponses before or during migra-

tion (30). The reduction in investment in im-mune defense could be an adaptive response tolower risks from certain parasites in migratoryspecies (beyond issues related to energetic trade-offs) and might affect adaptive immunity (shownto be costly for many vertebrate species) morestrongly than innate defenses. Over longer timescales, long-distance migration could select forgreater levels of innate immunity in migratoryspecies or populations, especially if migratinganimals encounter more diverse parasite assem-blages (54). With this in mind, comparisons ofadaptive and innate immune defense and re-

Fig. 4. A compartmental model illustrating infectious disease dynamics (S-Imodel) in a migratory host population moving between geographically distinctbreeding and overwintering habitats. Susceptible hosts (S) in the breeding groundsare born (v), die (mb) because of background mortality, and become infected at arate, b. Infected hosts (I) suffer disease-induced mortality (ab). Different fractionsof susceptible (xb) and infected hosts (yb) survive migration from the breedinghabitat and arrive successfully at an overwintering habitat at some rate (db). Here,natural (mw) and disease-induced mortality (aw) are both influenced by a differentset of environmental conditions that characterize wintering grounds. The fractionof hosts surviving the spring migration the following year (xwdw, ywdw) will returnto the breeding grounds to reproduce. A simple model like this can be readilymodified to accommodate different parasite species and their transmissionmodes,host recovery, host age structure, and cross-species transmission.

www.sciencemag.org SCIENCE VOL 331 21 JANUARY 2011 301

REVIEW

on

Janu

ary

20, 2

011

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

sistance to specific pathogens between migratoryand nonmigratory populations represent a chal-lenge for future work that could be especiallytractable with invertebrate systems (55).

Pathogens might also respond to migration-mediated selection, with ecological pressuresarising from migration leading to divergence invirulence. There is some evidence to show thatless-virulent strains circulate in migratory pop-ulations than in resident populations. The nega-tive correlation between virulence and hostmigration distance, illustrated in the monarchsystem (Box 1), highlights the troubling possi-bility that pathogens infecting other migratoryspecies could become more virulent if migrationsdecline. Moreover, dwindling migrations mightaffect host life history by altering pathogenvirulence in once-migratory hosts. For example,a theoretical study showed that even moderateincreases in virulence can change host breedingphenology to stimulate hosts to develop morequickly and breed earlier before they have achance to become heavily infected (56). Therecent facial tumor disease devastating Tasmaniandevil populations provides a striking empiricalexample of high disease-induced mortality shift-ing host reproductive strategy from an iteroparousto a semelparous pattern through precocious sex-ual maturity in young devils (57). Although thehosts in this example are nonmigratory, theyillustrate how virulent pathogens can generatelonger-term fecundity costs beyond their directimpacts on host survival.

Studying the migratory process in any wild-life species poses exceptional logistical chal-lenges, in part because distances separatingmultiple habitats can sometimes span thousandsof kilometers, making sampling for infection orimmunity intractable for field researchers. Oneproblem is that historically, large numbers of ani-mals have been sampled and marked at migratorystaging areas, but for many species their subse-quent whereabouts remain unknown (58). Track-ing animals over long time periods and acrossvast distances has become more feasible withtechnological innovations such as radar and satel-lite telemetry for larger animals and ultra-lightgeolocators, stable isotopes, and radio tags torecord or infer the movements of smaller animals(59). Furthermore, physiological measurementssuch as heart rate, wing beat frequency, and bloodmetabolites can be obtained remotely for somespecies, enabling scientists to examine how in-fection status influences movement rates and theenergetic costs of migration (59).

Interdisciplinary studies to connect the fieldsof migration biology and infectious disease ecol-ogy are still in the early stages, and there aremany exciting research opportunities to examinehow infection dynamics relate to animal physi-ology, evolution, behavior, and environmentalvariation across the annual migratory cycle. Most

evidence comes from studies of avian-pathogensystems, especially viruses. Although this is notsurprising given the relevance of pathogens suchas avian influenza and WNV to human health,there remains a great need to explore other sys-tems. Good places to start would be to makeconnections between disease and migration forspecies such as sea turtles, wildebeest, bats,dragonflies, and whales (Fig. 2). Parasite infec-tions andmovement ecology in species in each ofthese groups have been well studied separatelybut not yet bridged. Taking a broad view of di-verse host life histories and parasite transmissionmodes will allow future studies to identifyecological generalities and system-specific com-plexities that govern the mechanistic relation-ships between host movement behavior andinfectious disease dynamics.

References and Notes1. H. Dingle, Migration: The Biology of Life on the Move

(Oxford Univ. Press, Oxford, 1996).2. L. McKinnon et al., Science 327, 326 (2010).3. M. S. Bowlin et al., Integr. Comp. Biol. 50, 261 (2010).4. C. A. Russell, D. L. Smith, J. E. Childs, L. A. Real, PLoS Biol.

3, e88 (2005).5. C. Viboud et al., Science 312, 447 (2006); 10.1126/

science.1125237.6. A. White, A. D. Watt, R. S. Hails, S. E. Hartley, Oikos 89,

137 (2000).7. U. Carlsson-Graner, P. H. Thrall, Oikos 97, 97 (2002).8. J. Gog, R. Woodroffe, J. Swinton, Proc. R. Soc. London

Ser. B 269, 671 (2002).9. P. Daszak, A. A. Cunningham, A. D. Hyatt, Science 287,

443 (2000).10. D. S. Wilcove,NoWay Home: The Decline of the World's Great

Animal Migrations (Island Press, Washington, DC, 2008).11. J. H. Rappole, S. R. Derrickson, Z. Hubálek, Emerg. Infect.

Dis. 6, 319 (2000).12. J. Owen et al., EcoHealth 3, 79 (2006).13. E. M. Leroy et al., Vector Borne Zoonotic Dis. 9, 723

(2009).14. E. R. Morgan, G. F. Medley, P. R. Torgerson,

B. S. Shaikenov, E. J. Milner-Gulland, Ecol. Modell.200, 511 (2007).

15. B. Olsen et al., Science 312, 384 (2006).16. S. Krauss et al., PLoS Pathog. 3, e167 (2007).17. J. Y. Takekawa et al., Avian Dis. 54 (suppl.), 466

(2010).18. J. Waldenström, S. Bensch, S. Kiboi, D. Hasselquist,

U. Ottosson, Mol. Ecol. 11, 1545 (2002).19. J. Figuerola, A. J. Green, Evol. Ecol. 14, 143 (2000).20. S. Krauss et al., Proc. R. Soc. London Ser. B 277, 3373

(2010).21. C. Loehle, Ecology 76, 326 (1995).22. I. Folstad, A. C. Nilssen, O. Halvorsen, J. Andersen,

Can. J. Zool. 69, 2423 (1991).23. C. Bradley, S. Altizer, Ecol. Lett. 8, 290 (2005).24. A. M. Simmons, C. E. Rogers, J. Entomol. Sci. 26, 136

(1991).25. J. A. van Gils et al., PLoS One 2, e184 (2007).26. D. Kleijn et al., Proc. R. Soc. London Ser. B 277, 2041

(2010).27. P. Schmid-Hempel, Annu. Rev. Entomol. 50, 529 (2005).28. C. Eraud, O. Duriez, O. Chastel, B. Faivre, Funct. Ecol.

19, 110 (2005).29. T. P. Weber, N. I. Stilianakis, Emerg. Infect. Dis. 13, 1139

(2007).30. J. C. Owen, F. R. Moore, Condor 108, 389 (2006).31. J. C. Owen, F. R. Moore, J. Ethol. 26, 383 (2008).32. J. C. Owen, F. R. Moore, Can. J. Zool. 86, 638 (2008).

33. D. Hasselquist, A. Lindström, S. Jenni-Eiermann,A. Koolhaas, T. Piersma, J. Exp. Biol. 210, 1123 (2007).

34. D. M. Buehler, T. Piersma, K. Matson, B. I. Tieleman,Am. Nat. 172, 783 (2008).

35. A. P. Møller, F. de Lope, N. Saino, Ecology 85, 206 (2004).36. A. Gylfe, S. Bergström, J. Lundström, B. Olsen, Nature

403, 724 (2000).37. S. L. Messenger, J. S. Smith, C. E. Rupprecht, Clin. Infect. Dis.

35, 738 (2002).38. W. Li et al., Science 310, 676 (2005); 10.1126/

science.1118391.39. J. Berger, J. K. Young, K. M. Berger, PLoS Biol. 6, e165

(2008).40. L. Bienen, G. Tabor, Front. Ecol. Environ 4, 319 (2006).41. P. C. Cross et al., Ecol. Appl. 20, 278 (2010).42. A. M. Hines, V. O. Ezenwa, P. Cross, J. D. Rogerson,

Vet. Parasitol. 148, 350 (2007).43. P. Daszak et al., in Disease Ecology: Community Structure

and Pathogen Dynamics, S. K. Collinge, C. Ray, Eds.(Oxford Univ. Press, Oxford, 2006), pp. 186–201.

44. R. K. Plowright et al., Proc. R. Soc. London Ser. B 275,861 (2008).

45. M. Krkošek et al., Proc. R. Soc. London Ser. B 274, 3141(2007).

46. D. Harvell, S. Altizer, I. M. Cattadori, L. Harrington,E. Weil, Ecology 90, 912 (2009).

47. N. Saino et al., Proc. R. Soc. London Ser. B; publishedonline 22 September 2010 (10.1098/rspb.2010.1778).

48. D. W. Rice, A. A. Wolman, The Life History and Ecology ofthe Gray Whale (Eschrichtius robustus) (American Society ofMammalogists Special Publication 3, Stillwater, OK, 1971).

49. T. Jensen et al., Science 297, 209 (2002).50. R. Breban, J. M. Drake, D. E. Stallknecht, P. Rohani,

PLoS Comput. Biol. 5, e1000346 (2009).51. M. J. Keeling, L. Danon, M. C. Vernon, T. A. House,

Proc. Natl. Acad. Sci. U.S.A. 107, 8866 (2010).52. M. Hebblewhite, E. H. Merrill, Oecologia 152, 377

(2007).53. M. A. McPeek, R. D. Holt, Am. Nat. 140, 1010 (1992).54. A. P. Møller, J. Erritzøe, Evol. Ecol. 12, 945 (1998).55. S. M. Altizer, Evol. Ecol. Res. 3, 1 (2001).56. O. Restif, M. E. Hochberg, J. C. Koella, J. Evol. Biol. 14,

967 (2004).57. M. E. Jones et al., Proc. Natl. Acad. Sci. U.S.A. 105,

10023 (2008).58. M. S. Webster et al., Trends Ecol. Evol. 17, 76 (2002).59. W. D. Robinson et al., Front. Ecol. Environ 8, 354

(2010).60. L. P. Brower, S. B. Malcolm, Am. Zool. 31, 265 (1991).61. C. D. Nagano et al., in Biology and Conservation of the

Monarch Butterfly, S. B. Malcolm, M. P. Zalucki, Eds.(Natural History Museum of Los Angeles County, LosAngeles, CA, 1993), pp. 217–232.

62. P. R. Ackery, R. I. Vane-Wright, Milkweed Butterflies: TheirCladistics and Biology (Cornell Univ. Press, Ithaca, NY, 1984).

63. R. A. Bartel et al., Ecology; published online 19 July2010 (10.1890/10-0489.1).

64. S. M. Altizer, K. Oberhauser, L. P. Brower, Ecol. Entomol.25, 125 (2000).

65. J. C. de Roode, S. Altizer, Evolution 64, 502 (2010).66. For helpful discussion and comments, we thank J. Antonovics,

A. Davis, A. Dobson, V. Ezenwa, R. Hall, C. Lebarbenchon,A. Park, L. Ries, P. Rohani, P. R. Stephens, D. Streicker, andthe Altizer/Ezenwa lab groups at the University of Georgia.This work was supported by an NSF grant (DEB-0643831) toS.A., a Ruth L. Kirschstein National Research Service Awardthrough the NIH to R.B., an NSF Bioinformatics PostdoctoralFellowship to B.A.H., and a National Center for EcologicalAnalysis and Synthesis working group on Migration Dynamicsorganized by S.A., L. Ries, and K. Oberhauser.

Supporting Online Materialwww.sciencemag.org/cgi/content/full/331/6015/296/DC1SOM TextReferences

10.1126/science.1194694

21 JANUARY 2011 VOL 331 SCIENCE www.sciencemag.org302

REVIEW

on

Janu

ary

20, 2

011

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om