Sea urchin barrens as alternative stable states of collapsed … · Filbee-Dexter & Scheibling: Urchin barrens as alternative stable state termed a discontinuous phase shift (Fig

MARINE ECOLOGY PROGRESS SERIESMar Ecol Prog Ser

Vol 495 1ndash25 2014doi 103354meps10573

Published January 9

INTRODUCTION

Sea urchin barrens are benthic communities thatare dominated by herbivorous sea urchins andcoralline red algae on rocky reefs devoid of seaweed(Pearse et al 1970) (Fig 1) Barrens generally occurin regions that can support kelp beds (or forests)which are highly productive and provide habitat andfood for many ecologically and commercially impor-tant fish and invertebrate species (Konar amp Estes2003 Ling 2008 Bonaviri et al 2012) Over the last4 decades transitions between kelp beds and seaurchin barrens have been widely reported alongtemperate coastlines globally (Sala et al 1998 Pin-negar et al 2000 Steneck et al 2002) These transi-tions termed phase shifts generally occur when achange in sea urchin grazing intensity moves the sys-

copy Inter-Research 2014 middot wwwint-rescomCorresponding author kfilbeedextergmailcom

FEATURE ARTICLE REVIEW

Sea urchin barrens as alternative stable states ofcollapsed kelp ecosystems

Karen Filbee-Dexter Robert E Scheibling

Department of Biology Dalhousie University Halifax Nova Scotia B3H 4J1 Canada

ABSTRACT Sea urchin barrens are benthic commu-nities on rocky subtidal reefs that are dominated byurchins and coralline algae in the absence of intenseherbivory by urchins these barrens support luxuri-ant seaweed communities such as kelp beds (orforests) Barrens can extend over 1000s of km ofcoastline or occur in small patches (10s to 100s of m)within a kelp bed They are characterized by low primary productivity and low food-web complexityrelative to kelp communities and are generally con-sidered a collapsed state of the kelp ecosystem Toassess the stability of sea urchin barrens and poten-tial for return to a kelp-dominated state we docu-ment temporal and spatial patterns of occurrence ofbarrens along temperate and polar coasts We exam-ine the various drivers of phase (or regime) shifts inthese areas the threshold levels of urchin abundancethat trigger abrupt changes in ecosystem state andthe feedback mechanisms that stabilize each stateAlthough longitudinal (decadal) studies are limitedwe find evidence in several regions that transitionsbetween barrens and kelp beds are characterized bydiscontinuous phase shifts with different thresholdsfor forward (to barrens) and reverse (to kelp beds)shifts in accordance with alternative stable-statedynamics In other areas barrens may reflect regimeshifts associated with large-scale oceanographicchanges Accelerating climate change and increas-ing anthropogenic impacts play important roles inaltering alternative stable-state dynamics and trig-gering phase shifts Recovery of the kelp state maybe possible through management or remediationmeasures but this necessitates a clear understandingof the thresholds and stabilizing factors for a givensystem

KEY WORDS Sea urchin barrens middot Kelp beds middot Alternative stable states middot Phase shifts

Resale or republication not permitted without written consent of the publisher

Sea urchins Strongylocentrotus droebachiensis graze alongthe edge of a kelp bed creating a barren

Photo Robert Scheibling

FREEREE ACCESSCCESS

Mar Ecol Prog Ser 495 1ndash25 2014

tem from one stable (ie robust to relatively smallperturbations) community state to another (Lawrence1975 Steneck et al 2002) Sea urchin barrens havemuch lower primary productivity and habitat struc-tural complexity than kelp beds and consequentlycan be considered a collapse of the kelp state(Simenstad et al 1978 Chapman amp Johnson 1990Sivertsen 1996 Graham 2004 Christie et al 2009)Since kelp beds are key components of coastal eco-systems that provide important services to residentcommunities (Mann 1973 Levin 1994 Krumhansl

amp Scheibling 2012) understanding the factors thatcause phase shifts to urchin barrens and that enablekelp beds to recover is crucial for the proper man-agement of these ecosystems

Of particular concern to managers is the possibilitythat sea urchin barrens are a stable state of the sub-tidal ecosystem maintained by various feedbackmechanisms that prevent recovery of the kelp-dominated state after the initial driver of the phaseshift has been relaxed or reversed (Lauzon-Guay etal 2009 Ling et al 2009) This type of transition is

Fig 1 (A) Destructive grazing front of sea urchins Strongylocentrotus droebachiensis advancing into a kelp bed near HalifaxNova Scotia Canada Photo credit R E Scheibling (B) Extensive urchin (S polyacanthus) barrens in the Aleutian IslandsUSA Photo credit B Konar (C) Urchins S droebachiensis on scoured coralline algae in barrens in Norway Photo credit C WFagerli (D) Range-expanding urchin Centrostephanus rodgersii forming patchy barrens in a kelp bed in southeast Tasmania

Photo credit S D Ling (E) S nudus grazing a kelp bed in Japan Photo credit D Fujita

Filbee-Dexter amp Scheibling Urchin barrens as alternative stable state

termed a discontinuous phase shift (Fig 2A) andcharacterizes an alternative stable-state system(Lewontin 1969 Scheffer et al 2001 Collie et al2004 Mumby et al 2007 Fung et al 2011) It is dis-continuous because the threshold for the forwardshift to the barrens state is at a different level thanthe threshold for the reverse shift back to the kelpstate In contrast the forward and reverse transitionsof a continuous phase shift (Fig 2B) occur around the

same threshold level (Petraitis amp Dudgeon 2004)There is mounting evidence from marine systems(such as kelp beds seagrass beds and coral reefs)that collapse to less productive or structurally com-plex states occurs at a critical threshold of a forcingvariable (Sutherland 1974 Scheffer et al 2001Petraitis amp Dudgeon 2004 Casini et al 2009) How-ever few studies have conclusively documentedalternative stable-state dynamics (Knowlton 2004)and these have focused mainly on tropical coral reefs(Jackson 1997 Mumby et al 2007 Dudgeon et al2010 Fung et al 2011)

Despite compelling evidence of discontinuousphase shifts to sea urchin barrens for several regionsincluding Alaska USA (Estes et al 1998) Nova Sco-tia Canada (Lauzon-Guay et al 2009) and Tas -mania Australia (Ling et al 2009) the existence ofbarrens as a true alternative stable state of kelp eco-systems remains controversial Petraitis amp Dudgeon(2004) argue that inadequate information on themechanisms that create and stabilize kelp beds andurchin barrens precludes their classification as alter-native stable-state systems but that they remainstrong candidates for this designation Other expla-nations for large-scale shifts between kelp beds andbarrens are that they represent continuous phaseshifts between states most likely caused by ongoinganthropogenic impact (Connell amp Sousa 1983 Pe -traitis amp Dudgeon 2004) or that they are part of alarger oceanic regime shift to coralline-dominatedbarrens (Dayton et al 1998 Lees et al 2006 Wern-berg et al 2011) If shifts to sea urchin barrens arepart of a regime shift these transitions will likelyinvolve an abrupt long-term (decadal) change inoceanographic conditions occurring at large spatialscales and impacting multiple trophic levels (DeYoung et al 2004 Lees et al 2006)

In a comprehensive review of sea urchin grazingbehaviour on kelps and other macroalgae Lawrence(1975) summarized existing records of the distribu-tion of sea urchin-dominated barren grounds Ste-neck et al (2002) reviewed the literature on kelp eco-system collapses in temperate and boreal regionsworldwide including transitions to sea urchin bar-rens and possible forcing variables of phase shiftsEcosystem-specific reviews of alternations betweenkelp and barrens states also exist for Chile (Vaacutesquezamp Buschmann 1997) Maine USA (Steneck et al2013) Nova Scotia (Scheibling et al in press) and theNortheast Atlantic (Norderhaug amp Christie 2009)Here we document the extent and history of occur-rence of sea urchin barrens amid kelp-bed eco -systems worldwide to compile evidence on the nature

Fig 2 (A) Discontinuous phase shift As a kelp ecosystem(upper green path) approaches the threshold sea urchindensity F1 a small increase in density will forward-shift thekelp-bed state to a barrens state Once barrens have formeda reverse shift (lower pink path) back to the kelp-bed stateoccurs when sea urchin density is reduced below the F2threshold The difference between F1 and F2 thresholds in-dicates the strength of hysteresis in the system The dashedgray line represents the region of instability between the 2alternative stable states (B) Continuous phase shift The for-ward shift threshold F1 and reverse shift threshold F2 occurat the same sea urchin density The barren state only per-sists with high urchin densities and the kelp state immedi-ately recovers when densities are reduced Redrawn from

Scheffer et al (2001)

of phase shifts and potential for alternative stablestates We include several canopy-forming brownalgal communities (of the genera Sargassum andCystoseira) in the Mediterranean in our survey asthese macroalgae are functionally and taxonomicallysimilar to kelps (Round 1967) and offer furtherinsights into the formation of sea urchin barrens Webegin by briefly reviewing the theoretical frameworkof alternative stable-state dynamics and the associ-ated terminology which has been used inconsis-tently and often inaccurately in the large and growing body of literature on the subject We thenexamine the drivers of phase shifts between kelpbeds and barrens and the feedback mechanisms thatstabilize each community state Lastly we examineshifts to sea urchin barrens in the context of changingmarine environments and investigate the implica-tions of a collapse in kelp ecosystems for marinemanagement and conservation

ALTERNATIVE STABLE STATES

The concept of alternative stable states had its the-oretical underpinnings in the models of Lewontin(1969) Sutherland (1974) and May (1977) Peterson(1984) provided evidence of state shifts among natu-rally occurring communities and identified the con-cept of stability as a critical aspect of alternative sta-ble-state theory He proposed a simple criterion asevidence of alternative stable states different self-replacing communities can potentially dominate agiven site Connell amp Sousa (1983) presented strictercriteria that required each state to exist at a long-term stable equilibrium (longer than 1 complete turn-over of the dominant species) and the system toreturn to this point following a relatively small per-turbation or disturbance such as a fluctuation in aspeciesrsquo density or a storm event They suggestedthat long-term (decadal) studies are required to dis-tinguish alternative stable states Additional condi-tions for stability are that each state must persistin the absence of the perturbation(s) that triggeredthe transition and be maintained by feedbacks thatstrengthen a current state (Petraitis amp Latham 1999)

When this theoretical framework is applied to nat-ural systems this definition of stability becomes criti-cal (Grimm amp Wissel 1997) The requirement that thestate must exist for the lifespan of the dominant spe-cies under similar environmental conditions can bedifficult to assess because (1) it requires long-termresearch (eg over 100 yr for the red sea urchinStrongylocentrotus franciscanus Ebert amp Southon

2003) (2) it does not allow for natural variation inenvironmental conditions and (3) selection of domi-nants can be subjective in systems with many abun-dant species For this reason our review focuses onthe stabilizing mechanisms and feedback loops thatcreate domains of stability instead of defining stabil-ity as the elapsed time in a state Here we define astable state as a distinct community assemblage withfeedback mechanisms that under normal environ-mental conditions confer resistance or resilience ofthe community to relatively small perturbations (seeTable 1 for a glossary of ecological terms)

An important property of alternative stable states ishysteresis (Scheffer et al 2001) Hysteresis occurswhen an alternative state persists after the driverof the transition is relaxed or reversed Hysteresisis created by various stabilizing mechanisms thatinhibit return to the previous state Therefore fora kelp bed to re-establish in the barrens state seaurchin density (a proxy for herbivory) would haveto decrease well below the threshold density thatcaused the initial shift to barrens (Breen amp Mann1976 Ling et al 2009) (Fig 2A) The differencebetween thresholds for shifts in either directiondetermines the degree of hysteresis and the range ofsea urchin densities that can occur in either a kelp ora barrens state Transitions between 2 states withouthysteresis are continuous phase shifts and are readilyreversed by relaxing the forcing variable to thethreshold level that caused the shift (Petraitis ampDudgeon 2004) (Fig 2B) For example phase shiftstriggered by anthropogenic drivers may result in anecosystem state that is only stabilized by the pres-ence of continuing anthropogenic perturbation suchthat the original state is recovered when humanimpact ceases (Knowlton 2004)

From a modelling perspective a system can under -go a phase shift to a new state when a change ineither state variables or system parameters passes athreshold where stabilizing mechanisms maintainingthe original state are overcome (Beisner et al 2003)(Fig 3) State variables are system quantities (egkelp biomass abundances of urchins or their preda-tors larval supply) that change quickly in response tofeedback mechanisms within the ecosystem Systemparameters are measures that describe the behaviourof state variables and their interactions (eg grazingrate per capita predation settlement rate) Parame-ters can either change independently of state vari-ables or be subject to slow feedback mechanismsoriginating within the system state (Table 1) A phaseshift due to a strong perturbation or gradual changein state variables can shift the community from one

state to another without affecting the stability land-scape or parameters of the system In this type of transition the system can exist in 2 or more commu-nity states under the same set of environmental con-ditions Conversely a large change in system para -meters will alter the behaviour of the state variableswhich could destabilize a community and shift itto another domain of stability Some examples ofchanges in parameters that have caused shifts be -tween kelp beds and coralline barrens are the in -

creased mortality rate of sea urchins due to diseaseoutbreaks associated with warming ocean tempera-tures and storm severity in Nova Scotia (Scheibling ampLauzon-Guay 2010) the increased survival rate of seaurchins due to changes in ocean currents in Tasmania(Ling 2008) and the change in crab predation ratesdue to large-scale overfishing of groundfish in Maine(Steneck et al 2004) It is difficult to conceive of amarine system existing under a relatively constant setof parameters for decades particularly when seasonal

Term Definition

Alternative stable- An ecosystem that experiences discontinuous state ecosystem phase shifts meaning it can exist in two stable

states under the same environmental conditions

Continuous phase Transition from one ecosystem state to another shift where the threshold for the forward shift is at the

same level as the threshold for the reverse shiftback to the previous state

Discontinuous Transition from one ecosystem state to another phase shift where the threshold for the forward shift is at a

different level than the threshold for the reverseshift back to the previous state

Driver A forcing agent that causes a change in statevariable(s) or parameter(s) that results in a phaseshift

Ecosystem state The arrangement of species or populationswithin an ecosystem and their interactions withthe physi cal environment

Regime shift Changes in oceanographic processes and marinesystem functioning that are persistent occur at alarge spatial scale and over multiple trophic levelsand are related to climate oscillations or change

Resilience The ability of a community to return to an equi -librium state after a disturbance or perturbation

Stability The result of various feedback mechanisms thatunder normal environmental conditions enablea community to persist in a given state andresist or be resilient to small perturbations

State parameter Measure that governs the behaviour of statevariables and how they interact in an ecosystem

State variable Property of an ecosystem that responds tochanges in parameters

Table 1 Glossary of ecological terms with examples from kelp bed and urchin barrens community states

Examples

Sea urchin barrens and kelp beds

A shift to barrens where the kelp bed can re-establish when urchin grazing intensity decreasesto the threshold density triggering the initial shift

A shift to barrens where the kelp bed does not re-establish until urchin grazing intensity decreaseswell below the threshold density triggering theinitial shift

Overfishing or recovery of urchin predatorsurchin recruitment pulse disease outbreak stormevent or loss of kelp that results in an increase ordecrease in sea urchin grazing intensity

Abundances of macroalgal species coralline algaeand sea urchins as well as the 3-dimensionalstructure of the kelp bed and its associatedproperties

Shifts to barrens caused by El Nintildeo-SouthernOscillation events in California and Chile andsouthern intrusion of the East Australian Currentoff Tasmania

Regeneration of a kelp bed after a defoliationevent Return to a barrens state after a partial die-off of sea urchins or a temporary cessation in theirforaging activity due to strong wave action

A kelp bed that stays essentially unchangedunder constant environmental conditions isresistant to increases in urchin density and isresilient to small perturbations such as canopyloss temperature change or predator decline

Urchin grazing rate kelp growth rate recruitmentrates per capita predation rates These measurescan vary with changes in ocean currents oceantemperature and large-scale overfishing

Kelp biomass sea urchin density predator abun-dance larval abundance

cycles natural variability and anthropogenic impactsare continually changing the community land scapeTherefore domains of stability represent dynamiccommunity assemblages that are constantly beingmodified as system parameters change over time

KELP DISTRIBUTION AND ECOLOGY

Kelps are large brown seaweeds (class Phaeo-phyceae order Laminariales) that inhabit temperateor polar coastal regions throughout the world (Ste-neck amp Dethier 1994 Dayton et al 1999 Steneck etal 2002) (Fig 4 Appendix) They exhibit 3 basic mor-phologies that characterize kelp stands as forests(stipitate and canopy-forming kelps with fronds sus-pended above the seabed) or beds (prostrate formswith fronds lying on or near the seabed) (Steneck etal 2002) Canopy kelps (eg Macrocystis pyriferaNereocystis leutkeana Ecklonia maxima and Alariafistulosa) can extend to the ocean surface formingextensive forests along the western coasts of Northand South America They also are scattered through-out South Africa Southern Australia and New Zea -

land Stipitate kelps (eg Laminaria japonica Lesso-nia trabeculata and Ecklonia radiata) form midwaterstands extending from the Japan Sea across theNorth Pacific to California USA Prostrate kelps (egSaccharina latissima and Laminaria digitata) formlow-lying kelp beds through out much of the NorthAtlantic and are the dominant forms in GreenlandNorway and along the east coast of Canada to Maine(For simplicity here we generally designate kelpcommunities as beds unless the distinction as forestis important)

Kelps typically live a maximum of 25 yr (Steneck ampDethier 1994) and grow best in high-nutrient cold-water areas (Tegner et al 1996) They have highrates of primary production (Dayton 1985) and sup-port a variety of herbivorous and detritivorous spe-cies that graze attached or drift kelp (Duggins et al1989 Krumhansl amp Scheibling 2012) Kelps also arehost to various suspension feeders and micropreda-tors (Ling 2008) and serve as important nursery habi-tats for many fish (Bodkin 1988 Levin 1994) Periodsof high recruitment and primary productivity enablekelp beds to rapidly increase in biomass while peri-ods of severe storm activity (Filbee-Dexter amp Scheib-ling 2012) intensive grazing (Vadas amp Steneck 1988)low light or nutrient conditions (Dayton 1985 Tegneramp Dayton 1991 Tegner et al 1996) and warm water(Dayton et al 1999) erode or defoliate kelp beds

GLOBAL DISTRIBUTION OF SEA URCHIN BARRENS

Open clearings that are denuded of seaweeds andhave high densities of sea urchins have been ob -served in shallow rocky habitats worldwide (Table 2Fig 4) The spatial extent of these barrens can rangefrom 1000s of km of coastline to small patches (100s ofm in extent) within a kelp bed (Table 2) Urchin bar-rens are dominated by invertebrate species mainlysea urchins but also sea stars mussels and brittlestars They are devoid of fleshy and filamentous algaeand are primarily covered by encrusting coralline al-gae of low nutritional value Coastal areas dominatedby sea urchin barrens typically retain some localizedor spatially limited stands of kelp and other seaweedsFor example kelps have a refuge from urchin graz-ing in wave-swept shallow waters without sea ice (Lauzon-Guay amp Scheibling 2007a) and form smallpatches throughout urchin barrens in some areas(Vaacutesquez amp Buschmann 1997 Konar 2000)

Depending largely on the time span and intensityof research in different regions sea urchin barrens

Change in parameters Change in variables

Fig 3 Ball-in-cup diagram of alternative stable states A ballrepresents a particular community state that exists on a land-scape representing all possible states (2 states are consideredfor simplicity) Cups represent domains of attraction withinthat landscape Each ball is continually lsquovibratingrsquo withinthese domains in response to seasonal cycles and naturalvariability in the ecosystem The depth of a basin approxi-mates resilience to these natural variations and small pertur-bations in the environment Domains of attraction are alsomodified as system parameters change over time The eco-system can shift from one state to another (as represented bydisplacement of the ball) by either a change in state variablesthat moves the ball to a new domain of attraction or a changein state parameters that alters the landscape Top diagraminitial condition with a community in 1 of 2 possible statesred vertical arrow change in domains of attraction dashedblack arrows shifts from one domain of attraction to another

Redrawn from Beisner et al (2003)

Filbee-Dexter amp Scheibling Urchin barrens as alternative stable state 7

sl amp

Cminus

minus19

85minus

)Brarr

minus19

70minus

es amp

minus19

70minus

ABrarr

nminus

minus79

87minus

aBrarr

minus20

AKrarr

79minus

73minus

sBrarr

minus77

minus19

59minus

intildea

minus20

aacutesq

minus96

Vaacutes

ez amp

minus19

AKrarr

sminus20

minus98

Bminus

minus87

)Brarr

seminus

00minus

re amp

minus19

78minus

(Krarr

r Brarr

ndash n

Kminus

minusA

66minus

minus10

ayKrarr

minus93

seKrarr

70sminus

)Brarr

minus18

minus91

dBrarr

minus20

minus93

yminus

minus08

sKrarr

01minus

aacutend

gminus

02minus

aacutend

aBrarr

minus87

minus09

minusminus

78minus

seKrarr

5minus20

minus20

minus01

dKrarr

0)Krarr

78minus

minus79

minus19

30sminus

alKrarr

minus60

sminus90

Kminus

minus07

iaBrarr

minus05

iaBrarr

5minus15

minus86

have been documented under 3 types of conditions(1) through multiple phase shifts between kelp bedsand urchin barrens (2) following a single phase shiftfrom a kelp to a barrens state or vice versa and (3) inareas that might otherwise support kelp although aphase shift has not been observed In what followswe survey the occurrence of sea urchin barrensthroughout the global range of kelps (and some othercanopy-forming brown algae) and consider the driv-ers of phase shifts that have led to barrens

Barrens in regions with documented multiplephase shifts

Ecosystems where multiple shifts between kelpand barrens states have been documented provideimportant information on the drivers of these tran -sitions and the stability of each state Drivers ofchanges in urchin grazing intensity vary betweenthese systems but grazing typically increases afterperiods of high urchin recruitment and drift kelpshortage and decreases with predation overfishingand disease (Fig 5) The first evidence of kelp bedsalternating with sea urchin barrens comes from theAleutian Islands in the Northwest Pacific where seaotters are major predators of the sea urchin Strongy-locentrotus polyacanthus (Fig 5A) Early Europeanexplorers described subtidal areas in the Aleutians asa lush kelp forest with abundant sea otter popula-tions (Simenstad et al 1978) By the 1800s extensivehunting for the fur trade had decimated sea otterpopulations and caused the sea urchin population toincrease and destructively graze kelp forests (Simen-stad et al 1978) This shifted the system to stablecoralline barrens Legal protection of sea otters in1911 enabled sea otter populations to recover andreduce sea urchin densities to a level where kelpforests could re-establish (Estes amp Palmisano 1974)The recovered kelp forests (Alaria fistulos and Lami-naria spp) were maintained for decades until otterpopulations began to sharply decline due to preda-tion by killer whales in the 1990s (Estes et al 1998)This enabled sea urchin populations to increase onceagain and destructively graze kelp leading to theformation of barrens across most of the Aleutianarchipelago (Doroff et al 2003) The Aleutians alsoprovide a unique historical record of the state of thecoastal ecosystem based on the contents of aboriginalmiddens (Simenstad et al 1978) High abundances offish and sea otter remains in middens from 580 BCsuggest a kelp forest state whereas high abundancesof sea urchins and limpets in middens from 80 BC

suggest a barrens state providing evidence of local-ized transitions from kelp forests to coralline algalbarrens over 2000 yr ago possibly associated withaboriginal overharvest of sea urchin predators (Simen -stad et al 1978)

In California there is similar archeological evi-dence of short-lived localized shifts from giant kelpMacrocystis pyrifera forests to barrens thousands ofyears ago (Erlandson et al 1996) Phase shifts fromkelp forests to sea urchin barrens were recorded inCalifornia in the 1950s (Dayton et al 1984) and wereattributed mainly to the fishery-induced collapse ofspiny lobster and sheepshead fish populations (Day-ton et al 1998) predators of sea urchins that filled thefunctional role of sea otters after the fur trade hadeliminated them in the 1800s (Fig 5B) These seaurchin barrens persisted in California until the 1960swhen the reintroduction of sea otters led to reinstate-ment of kelp forests in some areas (McLean 1962Ebert 1968) However widespread kelp forest recov-ery did not occur until the mid-1970s when a fisheryopened for red sea urchins Strongylocentrotus fran-ciscanus (Dayton et al 1998) In 1988 localized phaseshifts to urchin barrens were documented followinga winter storm event and they persisted until seaurchin disease outbreaks in 1991 enabled kelp forestrecovery (Dayton et al 1992 Tegner et al 1997)Presently kelp forests dominate much of the Cali-fornian coast although patchy ur chin barrens occuramid these forests and kelp only occupies a third ofthe range measured in 1911 (Tegner et al 1996)

Shifts between kelp forests and barrens also havebeen associated with changing oceanographic con -ditions due to the El Nintildeo-Southern Oscillation InCalifornia El Nintildeo events in 1957minus1959 1982minus1984and 1992minus1993 disrupted upwelling and broughtwarm nutrient-depleted waters to coastal regions(Tegner amp Dayton 1991 Dayton et al 1998 Dayton etal 1999) This reduced kelp biomass and in someregions created temporary barrens that were recolo-nized by kelps during La Nintildea conditions (Tegner ampDayton 1987 Tegner et al 1997) Conversely in an8 yr study of a kelp forest (Macrocystis integrifoliaand Lessonia trabeculata) in northern Chile Vaacutesquezet al (2006) documented a 3-fold increase in re -cruitment of sea urchins Tetrapygus niger and asharp decline of kelp cover during a La Nintildea eventin 1999 This created a barrens state that was stablefor 4 yr until the kelp forest re-established in 2003

In the Northwest Atlantic kelp beds (Saccharinalatissima) in Maine have exhibited 3 distinct phasesin the last century (Steneck et al 2004) (Fig 5C) Thehistorical state was dominated by large predatory

fish such as cod haddock and wolffish which con-trolled sea urchin Strongylocentrotus droebachi ensispopulations and maintained the kelp-bed state (Ste-neck 1997) In the mid-1960s the functional loss ofpredatory fish due to fishing enabled sea urchin pop-ulations to increase driving the transition to urchin

barrens (Lamb amp Zimmerman 1964 Steneck et al2004) The barrens state persisted until 1987 whenan urchin fishery opened and decreased densitiesto the point at which kelp beds could re-establish(McNaught 1999) Currently the kelp-bed state ismaintained by crab predation which limits sea urchin

Urchin fishery

Otter extirpation

Low lobster fish abundance

A Aleutian islands B Central California C Maine

Otter mortality Storm defoliation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Low predation high recruitmenta

Urchin disease

Storm eventsa

Present

Present Present Present

Urchin fishery

Present

Higher temperaturea

Urchin introduction (current change)

High predation

(marine reserves)

Low predation (lobster fishery)

Groundfishcollapse

D Nova Scotia E Norway F Tasmania

Otter extirpation (fur trade)

Higher crab predation

Otter recovery

Urchin increase

Fig 5 Ball-in-cup diagrams representing phase shifts between kelp beds and sea urchin barrens in 6 regions (A) Aleutian Is-lands USA (B) Central California USA (C) Maine USA (D) Nova Scotia Canada (E) Norway and (F) Tasmania AustraliaFor each region the top diagram (in chronological order) represents the earliest known community assemblage (determinedby archeological evidence for the Aleutian Islands California and Maine) this is followed by documented phase shifts and as-sociated drivers leading to the present community state See lsquoBarrens in regions with documented multiple phase shiftsrsquo for de-tailed explanation of drivers and dynamics for each region Green balls represent kelp states pink balls represent barrenstates and light green or light pink balls indicate a transitional stage (eg kelp bed with active urchin grazing patches or bar-rens with kelp regrowth) Balls with dashed lines represent patchy kelp or barrens balls with solid lines represent extensivekelp or barrens Red vertical arrows represent changes in domains of attraction (resilience) dashed black arrows represent

shifts from one domain of attraction to another aA statistical association not a mechanistic driver

recruitment and keeps urchin populations at lowdensities (Steneck et al 2004 2013)

In Eastern Canada a shift from kelp beds (Lami-naria digitata and Saccharina latissima) to barrenswas first recorded in the late 1960s to the early 1970swhen dense aggregations of sea urchins overgrazedkelp in a large embayment near Halifax Nova Scotia(Breen amp Mann 1976 Wharton amp Mann 1981) (Fig 5D)By the late 1970s barrens dominated the entire At-lantic coast of Nova Scotia until recurrent outbreaksof amoebic disease in 1980minus1983 caused mass mor -talities of sea urchins that enabled kelp beds to re-establish (Scheibling 1984 1986 Jones 1985 Miller1985) Initial increases in sea urchin density withinkelp beds were attributed to declines of predatoryfishes crabs and lobsters (Wharton amp Mann 1981Bernstein amp Mann 1982) or possible recruitmentevents (Hart amp Scheibling 1988) The kelp beds transi-tioned to barrens again in the early 1990s after seaurchin density increased along the deep margins ofkelp beds and recovered in the late 1990s following awidespread recurrence of disease in 1995 and 1999(Scheibling et al 1999 Brady amp Scheibling 2005 Kellyet al 2011) Disease outbreaks in Nova Scotia havebeen linked to storm activity and warm water temper-atures and are becoming increasingly more frequent(Scheibling amp Lauzon-Guay 2010 Scheibling et al inpress) Con sequently kelp beds currently dominatemuch of the Nova Scotian coast although barrens ex-ist locally along headlands off central Nova Scotia andthe southwestern shore (Feehan et al 2013)

In the Northeast Atlantic luxuriant beds of Lami-naria hyperborea historically dominated the westerncoast of Norway (Skadsheim et al 1995) In 1975record high densities of the sea urchin Strongylocen-trotus droebachiensis destructively grazed kelp beds(Skadsheim et al 1995) (Fig 5E) and extensive urchinbarrens were created although sea urchins were un -able to remove mature kelp beds under low temper-ature and high light conditions (Leinaas amp Christie1996) Sea urchin die-offs in the early 1990s due toeither a macroparasitic infection (Sivertsen 1996) oran unidentified waterborne pathogen (Skadsheim etal 1995) returned parts of the coast to the kelp-bedstate Currently northern Norway is dominated bydecades-old sea urchin barrens although kelp bedsare re-establishing in southern and central Norwaymost likely as a result of low sea urchin recruitmentassociated with warmer ocean temperatures andincreased larval mortality (Sivertsen 2006 Fagerli etal 2013)

In the Southwest Pacific a long-term change in theEast Australian Current introduced the sea urchin

Centrostephanus rodgersii to coastal Tasmania inthe late 1970s (Edgar et al 2004) This influx ofsea urchins caused areas with particularly high seaurchin densities along the northeastern coast of Tasmania to shift from a kelp-dominated (Eckloniaradiata and Phyllospora comosa) state to barrens(Johnson et al 2005 Ling 2008 Johnson et al 2011)(Fig 5F) The resilience of kelp beds to these shiftshas likely been reduced by the spiny lobster fisherywhich removes an urchin predator from the system(Ling et al 2009 Ling amp Johnson 2012)

Barrens in regions with one or no documentedphase shift

Isolated phase shifts between kelp forest and bar-rens states do not offer direct evidence for alternativestable states but can provide information about driv-ers of transitions and the potential stability of seaurchin barrens Likewise long-term reports of seaurchin barrens that occur within the range of kelpdistribution but without documented phase shiftscan provide information about the global prevalenceand stability of the barrens state

In the Northeast Pacific along the coast of BritishColumbia Canada phase shifts from sea urchin(Stron gylocentrotus franciscanus) barrens to kelpforests (Nereocystic luetkeana) were documentedfollowing the reintroduction of sea otters in the late1960s (Breen et al 1982) and their subsequent rangeexpansion in the 1980s and 1990s (Watson amp Estes2011) Coastal surveys from 1987 to 2009 showed thatkelp forests occurred in areas with continuously highabundances of sea otters whereas urchin barrenswere found in areas where otters were absent (Wat-son amp Estes 2011) According to local fishermen partsof the coast were urchin barrens for decades prior tosea otter re-introduction (Breen et al 1982) A local-ized phase shift from kelp forests to urchin barrensalso was documented off British Columbia Canadawhen destructive grazing by sea urchins S droe-bachiensis removed a kelp forest in 1973 (Foreman1977)

In the Northwest Pacific a similar transition frombarrens to a kelp state occurred when sea otters re-colonized the Commander Islands in Russia reduc-ing sea urchin Strongylocentrotus polyacanthus den-sities and enabling the reestablishment of kelp beds(Saccharina dentigera) (Oshurkov et al 1988) On theeast coast of Japan sea urchins Strongylocentrotusnudus caused a phase shift from kelp forest (Undariapinnatifida and Laminaria japonica) to barrens in

Ogatsu Bay in the 1990s and prevented kelp re -growth for over 11 yr (Tamaki et al 2005 2009)Along the west coast of Hok kaido Japan sea urchinswere documented overgrazing the kelp forest in the1930s and formed extensive coralline barrens by the1960s (Matsunaga et al 1999 Fujita 2010 Graham2010) These barrens are most common in areas withlow water movement and have been reduced in someareas by harvesting and remediation (sea urchinremoval) efforts (Fujita 2010)

In the Southeast Pacific stable sea urchin (Tetra -pygus niger) barrens interspersed with patches ofkelp (Macrocystis integrifolia and Lessonia trabecu-lata) extend along much of the 2000 km of coastline ofnorthern Chile (Vaacutesquez amp Buschmann 1997) The re-silience of these barrens has likely been increased byunregulated macroalgal harvesting that targets kelp(Vaacutesquez 2008) In southern Chile only a few local-ized barrens maintained by sea urchin Loxechinus albus grazing have been reported within large tractsof kelp forest (Dayton 1985) Throughout most of thisregion sea urchins passively consume drift kelp anddo not actively graze kelp stands (Vaacutesquez et al 1984)

In the Northwest Atlantic extensive Strongylocen-trotus droebachiensis barrens exist along the west-ern eastern and southern coasts of Newfoundlandand southern coast of Labrador Canada (Keats 1991)Observations of barrens in these regions span peri-ods of 40 yr among the longest on record Althoughphase shifts to kelp have not been documented inthese areas (Keats 1991) kelp beds (Saccharina latis-sima) occur in some protected bays adjacent to bar-rens (Hooper 1975 Keats et al 1991) Sea urchinremoval experiments in Newfoundland showed thatmacroalgae colonized barrens but low-lying bedsof the brown alga Desmarestia aculeata often domi-nated the assemblage instead of kelp (Keats et al1990) In contrast the majority of the Greenlandcoast appears to be largely kelp-dominated (Saccha-rina spp) with dense patches of S droebachiensis ob -served in some regions (Krause-Jensen et al 2012)Blicher (2010) described a sea urchin barrens span-ning 200 m along the east coast of Greenland withina protected fjord in the Godtharingbsfjord system

In the Northeast Atlantic Hjorleifsson et al (1995)documented sea urchin fronts that emerged fromdeeper water to graze a kelp bed (Laminaria hyper-borea) and form barrens in Iceland Along the north-ern coasts of Norway and western Russia approxi-mately 2000 km of kelp beds (L hyperborea) weredestructively grazed in the early 1970s and sea urchin(Strongylocentrotus droebachiensis) barrens have per -sisted for almost 40 yr (Propp 1977 Sivertsen 1997

Norderhaug amp Christie 2009) Long-term monitoringof a localized sea urchin (Paracentrotus lividus) bar-rens in Lough Hyne Ireland captured a transition tolarge brown algae (Cystoseira foeniculacea and Sar-gassum muticum) in the 1990s that persisted for atleast 10 yr (Kitching 1987 Trowbridge et al 2011)Declines in sea urchin populations within the loughmay have been due to disease or predation (Trow-bridge et al 2011)

In the Mediterranean along the Albegraveres coast inSouthern France 20 km of macroalgal beds (Cysto-seira spp) collapsed to barrens in the 1970s and havenot recovered (Thibaut et al 2005) The possiblecauses included overfishing of sea urchin Paracen-trotus lividus predators and the recent prohibitionon sea urchin collection (Thibaut et al 2005) Inthe Adriatic Sea transitions from macroalgal beds(Cystoseira amentacea) to stable (9 yr) sea urchin(P lividus) barrens have occurred along 200 km ofcoastline (Fanelli et al 1994 Guidetti et al 2002Guidetti et al 2003) These shifts are attributed to thedestructive date mussel fishery which breaks apartreefs increasing the availability free space and smallspatial refugia This enhances settlement and re -cruit ment rates of sea urchins resulting in higherurchin densities on impacted reefs (Guidetti et al2003) Management of the fishery enabled macro-algae to recolonize some areas but urchin barrenspersist along most of the coast Paracentrotus lividusbarrens also have been described amid macroalgal-dominated (Cystoseira spp) reefs in the Aegean Seaoff the coast of Greece (Giakoumi et al 2012)

In the Southwest Pacific urchin barrens have beendocumented in Eastern Australia and New ZealandIn New South Wales Australia about 50 of 2000 kmof rocky coastline exists in a stable sea urchin (Centrostephanus rodgersii) barrens state (Andrewand Underwood 1989 Andrew amp Underwood 1993Andrew amp Byrne 2007 Connell amp Irving 2008 Glad-stone amp Masens 2009) Small-scale experimentalremovals of sea urchins from these barrens causeda shift to macroalgal-dominated habitats (Eckloniaradiata) (Fletcher 1987) Along the coasts of NewZealand urchin barrens have been documentedthrough out kelp forests (E radiata) (Shears amp Bab-cock 2007) In northern New Zealand the benthos at6minus8 m depth is dominated by sea urchins Evechinuschloroticus on coralline algal crusts and has persistedfor at least 10 yr (Schiel 1990) Establishment of amarine reserve in this region resulted in phase shiftsfrom urchin barrens to kelp forests that were attrib-uted to an increase in fish and invertebrate predatorsof sea urchins (Leleu et al 2012)

Notwithstanding the numerous examples of urchinbarrens worldwide the extent of phase shifts to bar-rens has in some cases been overstated or exagger-ated in the literature Past reviews have describedentire coastlines that alternate between kelp bedsand sea urchin barrens or coastal regions that haveremained in a kelp-bed state for thousands of yearsprior to overfishing (eg Steneck et al 2002) Whatthe evidence actually shows is that 10s to 100s of kmof temperate coastline in regions around the world atdepth ranges between wave-swept shallows andlight-limited deeper waters can transition betweena stable barrens state and a kelp- or macroalgal-dominated state (Table 2) The only data on kelp sys-tems older than 200 yr come from a handful of mid-den sites in Alaska California and Maine (Erlandsonet al 1996 2008 Bourque et al 2008) Althoughthese findings contribute greatly to our understand-ing of the dynamics of kelp ecosystems (Steneck etal 2002) they cannot be used to make broad conclu-sions about the historical state of kelp or barrens eco-systems throughout the world Furthermore much ofthe research on kelp beds and sea urchin barrenscomes from well-studied areas where attention wasinitially directed to high urchin densities or dramaticecosystem shifts As researchers continue to return toregions where barrens have previously been docu-mented we may be left with a lopsided view of thescale and importance of transitions Fig 4 showslarge spans of coastal kelp regions where sea urchin

barrens have not been documented mainly due tothe lack of research This indicates the need for abroader perspective to accurately assess the world-wide extent of sea urchin barrens

THRESHOLDS FOR PHASE SHIFTS AND STATESTABILITY

Field observations or sea urchin removal and trans-plantation experiments in Alaska California ChileNova Scotia Norway and Tasmania provide esti-mates of thresholds of urchin density or biomass forphase shifts These studies consistently show that thethreshold required to initiate destructive grazing ismuch greater than that which enables kelp recovery(Table 3) This difference between thresholds for for-ward and reverse shifts indicates hysteresis in thesedynamics and provides strong evidence of discontin-uous phase shifts between alternative stable statesThe percentage decrease in the threshold biomassof sea urchins between forward and reverse shiftsranged from 77 to 91 in these regions Thresholddensities varied markedly among regions reflectingdifferences in body size of the dominant sea urchinspecies while biomass thresholds were relativelyconsistent with order of magnitude differences be -tween forward shifts to barrens (1minus3 kg mminus2) andreverse shifts to kelp beds (01minus06 kg mminus2) Thresh-olds for phase shifts can vary locally with changes in

Region Method Threshold density Threshold biomass Biomass Mean Kelp Source(ind mndash2) (kg mndash2) decrease urchin biomass

KrarrB BrarrK KrarrB BrarrK () mass (g) (kg mndash2)

Alaska USA Exp 72 16 181a 041a 77 25 minus Konar amp Estes (2003)

California USA Obs 14 2minus3 281a 04minus061a 82b 200 02minus06 Dean et al (1984) Dayton et al (1992)

Chile Obs 36 20minus28 minus minus minus minus minus Vaacutesquez et al (2006)

Nova Scotia Obs Exp 31minus65 minus 15minus32c 015minus025 91b 49 20minus50 Breen amp Mann (1976) Chapman (1981) Canada Lauzon-Guay amp Scheibling (2007a)

Scheibling et al (1999)

Norway Obs Exp 45minus75 10 10minus17a 022a 84b 22 10 Hagen (1995) Leinaas amp Christie (1996) Sjoslashtun et al (1998)

Tasmania Obs Exp 4minus10 02minus12 09minus23d 005minus028d 90b 230 08 Ling (2008) Ling et al (2009) PecorinoAustralia et al (2012) Marzloff et al (2013)aUrchin biomass calculated as mean individual mass multiplied by mean density bmid-point in biomass range used in calculationcbiomass measured during destructive grazing dbiomass range estimated from TRITON model of alternative stable states (Marzloff etal 2013)

Table 3 Threshold sea urchin density and biomass required to trigger forward shifts from kelp to barrens states (KrarrB) and reverse shiftsfrom barrens to kelp states (BrarrK) in Alaska California Chile Nova Scotia Norway and Tasmania Thresholds were measured using fieldobservations (Obs) during phase shifts or experimental transplantation or removal of sea urchins (Exp) Biomass decrease indicates the percentage decrease in threshold biomass between forward and reverse shifts Measures were obtained from specific study sites and may

not reflect thresholds for entire regions

hydrodynamic conditions Strong wave action caninhibit aggregative feeding behaviour of sea urchinsby limiting their ability to climb kelp stipes andanchor blades (Lauzon-Guay amp Scheibling 2007a)Experimental transplantation of sea urchins in kelpbeds in Alaska and Nova Scotia showed that the den-sity threshold for destructive grazing was lowerwithin kelp beds than along the kelpminusbarrens inter-face at the edge of beds where wave action isgreater (Konar amp Estes 2003 Feehan et al 2012) Thebiomass of kelp also can directly influence thethreshold urchin biomass for destructive grazing anda shift to barrens (Lauzon-Guay amp Scheibling 2007bLauzon-Guay et al 2009)

Once threshold urchin densities are attained phaseshifts between kelp beds and barrens are relativelyabrupt Destructive grazing creates positive feedbackmechanisms that accelerate the shift to barrens Actively grazing sea urchins have unlimited highquality food which enables them to grow rapidly andallocate a large amount of energy to reproduction(Meidel amp Scheibling 1998) Because highly fecundsea urchins are aggregating in high densities fertil-ization rates are maximal (Meidel amp Scheibling 2001Lauzon-Guay amp Scheibling 2007c) which likely in-creases larval supply and recruitment to barrens onregional scales Similarly when sea urchin densitiesin barrens drop significantly the release from grazingtriggers an immediate response filamentous algaeand diatoms appear within days of urchin removaland kelps recruit and grow into cano -pies within 1 to 3 yr (Duggins 1980Harrold amp Reed 1985 Johnson amp Mann1988 Tegner et al 1997 McNaught1999 Konar amp Estes 2003 Ling 2008Ford amp Meux 2010 Watanuki et al2010 Watson amp Estes 2011)

There are 2 types of feedback mech-anisms that stabilize the communityassemblage in the barrens state pro-cesses that reduce kelp recruitmenton barrens and processes that allowsea urchins to maintain high densitieson barrens (Fig 6) Sea urchins inbarrens prevent kelp recruitment bycontinually scraping coralline algalcrusts con suming the surficial layersalong with any microalgal films andmacroalgal recruits (Chapman 1981)This reduces the survival of kelpsporophytes in barrens (Jones amp Kain1967) Sea urchin exclusion experi-ments in the Gulf of St Lawrence

Canada found that kelp recruitment was 100 timeshigher on barrens without urchins than on barrenswith urchins (Gagnon et al 2004) In widespreadbarrens the urchin-dominated state may be furtherstabilized by a lack of reproductive source popula-tions of kelp that provide spores for recruitment(Keats 1991) Kelp spores are short-lived and typi-cally settle within 5 to 10 m of the parent plant (Nor-ton 1992 Gaylord et al 2012) although maximumdispersal distances of up to 5 km have been meas-ured for some species (eg Laminaria hyperboreaNorton 1992 Macrocystis pyrifera Gaylord et al2006) In Nova Scotia barrens adjacent to shallowstands of reproductive kelp sporophytes re-estab-lished kelp beds within 18 mo following urchinmass mortality whereas it took 4 yr for kelp beds torecover on barrens that were 3 km away fromthe nearest reproductive kelps (Johnson amp Mann1988) Likewise sea urchin removal experimentsconducted within ex tensive barrens off Newfound-land Canada did not result in colonization by kelpafter 3 yr because the nearest reproductive kelpswere several kilometers away from the removal plots(Keats 1991 Keats et al 1990) This effect may bemitigated in barrens where a few remaining sporo-phytes are ex posed to elevated light nutrients andcurrents which can result in greater fecundity Inthe Aleutian Islands individual sporophytes of Eu -alaria fistulosa in barrens produced 3 times morespores than individual sporophytes in adjacent kelp

Kelp state

Barren state

Urchin recruitment

Urchin fertilization

Urchin settlement facilitation

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

Predator abundance Urchin

mortality

Kelp recruitment

Whiplash Urchin grazing

Spore production

Urchin migration

Fig 6 Stabilizing feedback mechanisms for the kelp-bed and sea urchin bar-rens states Blue (solid line) is positive and red (dashed line) is negative feed-back 2 sequential negative feedbacks indicate an indirect positive feedback

forests (Edwards amp Konar 2012) In California somekelp species form free-floating rafts that can dis-perse spores over great distances and may mitigatethe loss of reproductive sporophytes in widespreadbarrens (Hobday 2000)

Despite the lack of kelp and other fleshy macro-algae as food sources sea urchins can maintainhigh densities on barrens by allocating fewer re -sources to reproduction and growth undergoingmorphological changes in their body wall (Edwardsamp Ebert 1991) and reabsorbing parts of their bodywall or gut (Pearse et al 1970) High densities of seaurchins in barrens can offset decreased individualreproductive output enabling populations to sustainmoderately high fertilization rates and contribute tothe larval pool (Lauzon-Guay amp Scheibling 2007c)However since sea urchins have a planktonic larvalstage of 2minus3 mo and can disperse distances of up to100 to 1000 km (Huggett et al 2005) any positiveimpact of a larger larval pool on sea urchin settle-ment would likely be limited to large-scale barrens(100s of km)

Settlement of sea urchins in barrens is enhancedby a chemical cue associated with coralline algaethat induces settlement and metamorphosis of seaurchin larvae (Pearse amp Scheibling 1990) Thereforeby preventing kelps and other fleshy or filamentousmacroalgae from overgrowing and outcompetingcorallines sea urchin grazing in barrens facilitatesthe supply of new individuals to the population(Miner et al 2006 Hernaacutendez et al 2010) Baskett ampSalomon (2010) generated discontinuous phase shiftsbetween barrens and kelp beds in a model that in -corporated sea urchin grazing on kelp competitionbetween kelp and coralline algae and facilitationof sea urchin recruitment by coralline algae Seaurchins in barrens likely experience lower post-settlement mortality due to predation compared withkelp beds which also acts to increase recruitmentand stabilize a barrens state The low structural com-plexity of barrens compared with the 3-dimensionalstructure of kelp beds limits available habitat forpredators of sea urchins such as decapod crusta -ceans and fish (Levin 1994 Konar amp Estes 2003Gianguzza et al 2010) including those that prey onthe early juvenile stages (Hacker amp Steneck 1990Bonaviri et al 2012)

The decrease in kelp cover during a shift to the bar-rens state reduces the supply of kelp detritus both toshallow kelp beds (Ebeling et al 1985) and to adja-cent habitats in deeper regions (Vanderklift amp Ken -drick 2005 Krumhansl amp Scheibling 2011) Residentsea urchins in kelp beds like those in deeper regions

generally feed passively on drift kelp (Harrold ampReed 1985 Filbee-Dexter amp Scheibling 2012) Whenthis subsidy declines urchins emerge from sheltersto actively graze attached kelp and augment popula-tions in barrens providing another form of feedbackthat can stabilize the kelp-bed state

A healthy kelp bed is maintained by various feed-back mechanisms that prevent the increases in seaurchin density that lead to destructive grazing andthe formation of barrens (Fig 6) Algal films andunderstory algae inhibit settlement of sea urchin larvae by reducing the availability of open space onrocky substrata (Trowbridge et al 2011) High levelsof predation on juvenile urchins in kelp habitatscompared with barrens limits recruitment (Tegneramp Dayton 1981 Leinaas amp Christie 1996 Scheibling1996) Sea urchins within kelp beds often are cryptic sheltering in spatial refuges from predators(eg crevices undersides of boulders and kelphodlfasts) few reach a size refuge from all butthe largest predators (Scheibling amp Hamm 1991Clemente et al 2007)

The physical structure of kelp beds also can pre-vent sea urchin grazing The wave-driven whiplashand sweeping motion of large kelps impedes urchinsfrom moving into kelp beds (Vaacutesquez 1992 Konar2000 Tamaki et al 2009) In experimental kelpremovals in Alaska Konar amp Estes (2003) showedthat sea urchins advanced beyond the deep marginsof kelp forests (at 8minus13 m depth) when kelp wasremoved but not when kelp was replaced with phys-ical mimics indicating that the sweeping motion ofkelp arrested the onshore advance of grazing aggre-gations Dislodgment of sea urchins may also behigher in kelp beds compared with barrens In a lab-oratory study using a flume Kawamata (2010) showedthat sea urchins attached to turf algae stoppedactively moving and were dislodged at lower watervelocities than when attached to bare rock

The high production of detrital material within kelpbeds (Krumhansl amp Scheibling 2012) provides animportant subsidy for resident sea urchins (Harrold ampReed 1985) and offshore populations (Britton-Sim-mons et al 2009 Filbee-Dexter amp Scheibling 2012)This reliance on passive detrivory lowers grazingintensity on attached kelp (Day amp Branch 2000) andlikely reduces adult migration into kelp beds In con-trast detrital subsidy from a highly productive kelpstate can also enhance reproductive output of off-shore sea urchin populations (Britton-Simmons et al2009 K Filbee-Dexter unpubl data) which couldincrease the larval pool and consequently settle-ment of sea urchins in the kelp bed

ARE SEA URCHIN BARRENS AN ALTERNATIVESTABLE STATE

With some exceptions sea urchin barrens gener-ally result from discontinuous phase shifts and there-fore are considered an alternative stable state of kelpecosystems (Table 2) Phase shifts between kelp bedsand sea urchin barrens show evidence of hysteresisafter a transition (Table 3) and both the kelp and barrens states are stabilized by numerous feedbackmechanisms and are resistant to small perturbationsor fluctuations in sea urchin densities Sea urchinbarrens can persist for decades and exist under envi-ronmental conditions similar to those of kelp bedsMost shifts to barrens are driven by localized changesin state variables and parameters and as such are nota part of a larger oceanic regime shift linked to cli-mate change or climate oscillations Exceptions arethe phase shifts observed in Tasmania and Californiathat are caused by changing ocean currents The rapidly changing ocean temperature in Tasmaniadue to the increased southern penetration of theEast Australian Current may constitute an oceanicregime shift (Johnson et al 2011) Likewise periodicchanges in coastal upwelling in California due to theEl Nintildeo-Southern Oscillation also may represent anoceanic regime shift (Tegner amp Dayton 1987 Daytonamp Tegner 1990)

We find little evidence supporting prior argumentsthat human-induced shifts between kelp beds andbarrens constitute continuous phase shifts that aremaintained by ongoing anthropogenic impacts (Con-nell amp Sousa 1983 Petraitis amp Dudgeon 2004) In ecosystems where human perturbations cause phaseshifts hysteresis still occurs and the alternative statepersists after the human control is relaxed (Tables 2amp 3) However human activities such as moratoria onotter hunting expanding sea urchin fisheries or con-tinued depletion of groundfish are likely increasingthe occurrence of phase shifts in kelp ecosystems(Scheffer et al 2001 Knowlton 2004) The dramaticchanges in sea urchin densities that are requiredto trigger phase shifts may be difficult to achievethrough natural causes but could readily occur throughstrong anthropogenic perturbations (Knowlton 2004)Given that humans are increasingly impacting oceanecosystems globally the implications of human per-turbations in triggering phase shifts in kelp ecosys-tems are of growing concern

In kelp ecosystems that exhibit alternative statedynamics the recovered community state often dif-fers from the state that existed prior to a collapse Forexample in Maine the groundfish associated with

kelp beds in the 1930s and subsequently depleted bycoastal fisheries were not re-established with thereturn to the kelp state in the 1990s (Steneck et al2004) In California the sheepshead fish and lobsterpopulations that controlled sea urchins in kelp forestsin the 1930s did not recover in kelp forests in the1970s (Dayton et al 1998) In the Aleutian Islandssea otter populations in re-established kelp forestsare encountering a new agent of mortality in the formof killer whale predation (Estes et al 1998 Tegner ampDayton 2000) In the last 3 decades climate changehas been implicated in lowering recruitment of seaurchins in Norway (Fagerli et al in 2013) increasingthe frequency of disease outbreaks that cause massmortality of sea urchins in Nova Scotia (Scheiblinget al in press) and modifying currents that haveexpanded the range of sea urchins into Tasmania(Johnson et al 2011) The escalating influences ofhumans in each of these regions may be causingphase shifts to new more deteriorated ecosystemstates with fewer species less biomass and increasedlevels of human impact rather than alternationsbetween 2 persistent community configurations Al -though human perturbations may be altering theresilience of these communities they still exhibitbroad domains of attraction in both the kelp-domi-nated and urchin barrens state which allows theirclassification as alternative stable-state systems

IMPLICATIONS FOR MANAGEMENT OF KELP-BASED ECOSYSTEMS

Given that phase shifts to barrens often are consid-ered as manifestations of the collapse of a kelp-basedecosystem various strategies have been attempted torecover the productive kelp state By definition sys-tem recovery can be challenging after a discontinuousphase shift because of hysteresis making it difficultto reverse a collapse (Scheffer et al 2001) Even sosome forms of management particularly those focusedon controlling populations of urchin predators havebeen effective in restoring kelp forests Actions to re-establish populations of the sea otter Enhydra lutrisconsidered a keystone species in the North Pacific forits cascading effects on kelp abundance (Paine 1969)provide an early example of this strategy Historicalmoratoria on sea otter hunting effectively restoredpopulations in eastern Russia western Alaska andCalifornia and together with sea otter translocationsacross the eastern Pacific led to the recovery ofkelp forests in many regions (Estes amp Palmisano 1974Breen et al 1982) Currently sea otter populations are

declining because of oil spills (Bodkin et al 2002) dis-ease (Kannan et al 2006) and killer whale predation(Doroff et al 2003) Wilmers et al (2012) proposedthat proper otter conservation strategies would maxi-mize kelp forest abundance in the northeast Pacificand create an important carbon sink

The establishment of marine reserves also canrestore predator populations and recover the kelpstate In New Zealand increased lobster and preda-tory fish populations and substantial re-growth ofkelp was documented in marine protected areascompared with unregulated areas (Babcock et al1999 Shears amp Babcock 2003 Shears et al 2006Leleu et al 2012) In the Adriatic Sea the percentagecover of barrens was lower in marine reserves wherefishing prohibitions are strictly enforced than inunmonitored areas where poaching occurs (Guidettiet al 2003) In Tasmania marine reserves increasedspiny lobster populations and maintained the kelpstate by limiting the potential for destructive grazingby sea urchins through higher predation rates (Lingamp Johnson 2012) However the effectiveness of suchprotection strategies can be limited In Tasmania alarge-scale experimental introduction of thousandsof spiny lobsters into both widespread barrens andpatchy barrens amid kelp beds resulted in no in -crease in kelp cover in widespread barrens and onlya small increase in kelp cover in patchy barrens (S DLing pers comm) This indicates that the barrensstate is extremely resilient to kelp recovery (Marzloffet al 2013) and only preventative management toincrease the resilience of the kelp-bed state may beeffective in halting phase shifts to barrens

Judicious management of fisheries may recoverkelp assemblages Sea urchin fisheries in Maine andCalifornia have reduced sea urchin densities belowthresholds that maintain barrens enabling a reverseshift to a kelp-dominated state (Tegner amp Dayton1991 Steneck et al 2004) In Nova Scotia the seaurchin fishery manages the stock by targeting thegrazing front at the deep edge of a kelp bed (Miller ampNolan 2000) This halts or slows the advance of frontsuntil trailing sea urchins in the barrens encounter thekelp and re-establish aggregations allowing for asustainable harvest (Miller amp Nolan 2000) In Cali -fornia commercial kelp harvesters prevented seaurchin grazing fronts from advancing into kelp forestsusing quicklime before the establishment of anurchin fishery (North 1971) In Japan artificial reefshave been suspended above the substratum onbuoyed arrays to exclude sea urchins resulting in therecovery of kelps for commercial harvest (Tamaki etal 2009)

An unexplored strategy for conserving the kelpstate could involve managing human impacts thataffect feedback mechanisms in kelp systems Forexample minimizing kelp harvesting or coastal sedi-mentation due to runoff would increase kelp bio-mass This would increase the supply of drift kelpwhich could prevent behavioural switches to activegrazing in resident sea urchins Similarly seedingbarren areas with reproductive kelp fronds could en -hance kelp settlement in regions with limited sporesA better understanding of the feedback mechanismsthat stabilize the barrens state may help inform management strategies

Two major challenges face effective managementof kelp-bed and barrens ecosystems First manage-ment strategies require a clear understanding of individual ecosystems as the relative importance ofstabilizing mechanisms and drivers of state shifts canvary with species composition trophic interactionsfunctional redundancy and environmental conditionsthat are unique to each system Successful manage-ment of barrens has mainly been limited to well-studied systems where the drivers of transitions arewell understood Further research is needed in otherregions of the kelp range such as South AmericaAfrica Asia and the Arctic Second it is not possibleto manage phase shifts re sulting from environmentalchanges such as warming oceans increased stormseverity and altered currents (Ebeling et al 1985Ling 2008 Harley et al 2012 Scheibling et al in press)These changes may be mitigated to some extent bymaintaining high biodiversity and species richnesswithin kelp beds (Folke et al 2004) as phase shifts tobarrens tend to be more common in systems with lowtrophic complexity and low functional redundancy(Steneck et al 2002) However future impacts of cli-mate change on these ecosystems greatly exceed themanagement capacities of coastal areas and wouldrequire a larger global initiative that prevents furtherenvironmental change in ocean ecosystems

PERSPECTIVES FOR FUTURE RESEACH

Kelp forests or beds are complex ecosystems thathave the potential to drastically change in terms ofboth structure and function through phase shifts tosea urchin barrens To fully understand whether barrens represent an alternative stable state ofkelp ecosystems further longitudinal studies of kelpand barrens communities are required Manipula-tive field experiments typically provide the strongestevidence of alternative stable states and can be

used to elucidate thresholds for state shifts as wellas system-specific feedback mechanisms that canstabilize both kelp and barrens states For examplesea urchin removal experiments not only indicatethe potential macroalgal community that can de -velop within barren grounds but also can be used toquantify thresholds for recovery of the kelp stateLong-term monitoring programs and statis tical mod-els are also useful in evaluating the stability andorganization of different ecosystem states (Johnsonet al 2013 Marzloff et al 2013) A major source ofuncertainty in kelp and barrens ecosystems is theperiod between sea urchin larval release and settle-ment In the majority of these ecosystems the fate oflarvae produced by resident populations in barrenskelp beds or nearby deep areas is largely unex-plored and likely plays an important role in bothdriving density-induced phase shifts and stabilizingthe barrens or kelp state

There are several trends in the global occurrenceof sea urchin barrens that may warrant furtherresearch It is unclear why barrens dominatethrough out eastern Canada western Russia andnorthern Norway but are rarely documented alongthe coasts of Greenland and Iceland regions withsimilar species composition and environmental con-ditions Sea urchin barrens also tend to be morewidespread and phase shifts occur more frequentlyalong temperate coasts in the northern hemispherethan along tropical and south temperate coastsThe trend of increasing marine species richness andecosystem complexity from the Arctic to the tropics(Gray 2001) may explain this discrepancy becausethe more simplified food webs in temperate ecosys-tems may collapse more readily Southern kelp bedsalso occur in upwelling zones which may havealtered feedbacks and dynamics compared with temperate ecosystems

Considerable attention has been directed towardsestablishing criteria for defining an alternative stablestate (Connell amp Sousa 1983 Beisner et al 2003Petraitis amp Dudgeon 2004) However in a practicalsense regardless of whether phase shifts betweenkelp beds and barrens reflect an actual alternativestable-state system the barrens state typically ex -hibits multiple feedback mechanisms that can inhibitkelp recovery for decades As Knowlton (2004) aptlyobserved in the context of marine conservation itprobably makes little difference in human timescales if sea urchin barrens persist indefinitely whatmatters is that the system can undergo a long-termdeparture from prevailing conditions that is difficultto reverse

Acknowledgements We thank A Metaxas S Ling C Fee-han K Krumhansl S Walde and H Whitehead for helpfulcomments on earlier drafts of the manuscript This researchwas funded by a Discovery Grant to RES from the NaturalSciences and Engineering Research Council (NSERC) ofCanada KF-D was supported by a Dalhousie Killam Schol-arship and an NSERC Canada Graduate Scholarship

LITERATURE CITED

Alves F Chiacutecharo LM Serrao E Abreu AD (2001) Algalcover and sea urchin spatial distribution at MadeiraIsland (NE Atlantic) Sci Mar 65383minus392

Andrew N Byrne M (2007) Ecology of Centrostephanus InLawrence JM (ed) Edible sea urchins biology and eco -logy Elsevier Amsterdam p 191minus204

Andrew NL Underwood AJ (1989) Patterns of abundance ofthe sea urchin Centrostephanus rodgersii (Agassiz) onthe central coast of New South Wales Australia J ExpMar Biol Ecol 13161minus80

Andrew NL Underwood AJ (1993) Density-dependant for-aging in the sea urchin Centrostephanus rodgersii onshallow subtidal reefs in New South Wales AustraliaMar Ecol Prog Ser 9989minus98

Babcock RC Kelly S Shears NT Walker JW Willis TJ(1999) Changes in community structure in temperatemarine reserves Mar Ecol Prog Ser 189125minus134

Baskett ML Salomon AK (2010) Recruitment facilitationcan drive alternative states on temperate reefs Ecology911763minus1773

Beisner BE Haydon DT Cuddington K (2003) Alternativestable states in ecology Front Ecol Environ 1376minus382

Bernstein BB Mann KH (1982) Changes in the nearshoreecosystem of the Atlantic coast of Nova Scotia 1968minus81NAFO Sci Counc Stud 5101minus105

Bischoff B Wiencke C (1993) Temperature requirements forgrowth and survival of macroalgae from Disko Island(Greenland) Helgol Meeresunters 47167minus191

Blicher ME (2010) Structure and dynamics of marine macro-zoobenthos in Greenland and the link to environmentaldrivers PhD dissertation Greenland Climate ResearchCentre amp Department of Biology Greenland Institute ofNatural Resources University of Copenhagen

Bodkin JL (1988) Effects of kelp forest removal on associatedfish assemblages in central California J Exp Mar BiolEcol 117227minus238

Bodkin JL Ballachey BE Dean TA Fukuyama AK and oth-ers (2002) Sea otter population status and the process ofrecovery from the 1989 lsquoExxon Valdezrsquo oil spill Mar EcolProg Ser 241237minus253

Bonaviri C Gianguzza P Pipitone C Hereu B (2012) Micro-predation on sea urchins as a potential stabilizing pro-cess for rocky reefs J Sea Res 7318minus23

Bourque BJ Johnson BJ Steneck RS (2008) Possible prehis-toric fishing effects on coastal marine food webs in theGulf of Maine In Erlandson J Rick T (eds) Humanimpacts on ancient marine ecosystems a global per -spective University of California Press Berkeley CAp 165minus185

Brady SM Scheibling RE (2005) Repopulation of the shallowsubtidal zone by green sea urchins (Strongylocentrotusdroebachiensis) following mass mortality in N CanadaJ Mar Biol 851511minus1517

Breen PA Mann KH (1976) Destructive grazing of kelp by

sea urchins in eastern Canada J Fish Res Board Can331278minus1283

Breen PA Carson TA Foster JB Stewart EA (1982) Changesin subtidal community structure associated with BritishColumbia sea otter transplants Mar Ecol Prog Ser 713minus20

Britton-Simmons KH Foley G Okamoto D (2009) Spatialsubsidy in the subtidal zone utilization of drift algae bya deep subtidal sea urchin Aquat Biol 5233minus243

Casini M Hjelm J Molinero JC Loumlvgren J and others (2009)Trophic cascades promote threshold-like shifts in pelagicmarine ecosystems Proc Natl Acad Sci USA 106197minus202

Chapman ARO (1981) Stability of sea urchin dominated barren grounds following destructive grazing of kelp inSt Margarets Bay eastern Canada Mar Biol 62307minus311

Chapman ARO Johnson CR (1990) Disturbance and organ-ization of macroalgal assemblages in the northwestAtlantic Hydrobiologia 19277minus121

Chapman ARO Lindley JE (1981) Productivity of Laminariasolidungula in the Canadian high Arctic a year-roundstudy In Levring T (ed) 10th International SeaweedSymposium Proceedings Walter de Gruyter Berlinp 247minus252

Choat JH Schiel DR (1982) Patterns of distribution andabundance of large brown algae and invertebrate herbi-vores in subtidal regions of northern New Zealand J ExpMar Biol Ecol 60129minus162

Christie H Norderhaug KM Fredriksen S (2009) Macro-phytes as habitat for fauna Mar Ecol Prog Ser 396221minus233

Clemente S Joseacute CH Toledo K Brito A (2007) Predationupon Diadema antillarum in barren grounds in theCanary Islands Sci Mar 71745minus754

Collie JS Richardson K Steele JH (2004) Regime shiftsCan ecological theory illuminate the mechanisms ProgOceanogr 60281minus302

Connell JH Sousa WP (1983) On evidence needed to judgeecological stability or persistence Am Nat 121789minus824

Connell SD Irving AD (2008) Integrating ecology with bio-geography using landscape characteristics a case studyof subtidal habitat across continental Australia J Bio-geogr 351608minus1621

Cross WE Wilce RT Fabijan MF (1987) Effects of experi-mental releases of oil and dispersed oil on Arctic near -shore macrobenthos III Macroalgae Arctic 40211minus219

Day EG Branch GM (2000) Evidence for a positive relation-ship between juvenile abalone Haliotis midae and thesea urchin Parechinus angulosus in the south-westernCape South Africa S Afr J Mar Sci 22145minus156

Dayton PK (1985) Ecology of kelp communities Annu RevEcol Syst 16215minus245

Dayton PK Tegner MJ (1990) Bottoms beneath troubledwaters benthic impacts of the 1982minus1984 El Nintildeo in thetemperate zone In Glynn PW (ed) Global consequencesof the 1982minus1983 El Nintildeo-Southern Oscillation ElsevierOceanography Series Amsterdam p 433minus466

Dayton PK Currie V Gerrodette T Keller BD Rosenthal RTresca DV (1984) Patch dynamics and stability of someCalifornia kelp communities Ecol Monogr 54253minus289

Dayton PK Tegner MJ Parnell PE Edwards PB (1992) Tem-poral and spatial patterns of disturbance and recovery ina kelp forest community Ecol Monogr 62421minus445

Dayton PK Tegner MJ Edwards PB Riser KL (1998) Slidingbaselines ghosts and reduced expectations in kelp forest communities Ecol Appl 8309minus322

Dayton PK Tegner MJ Edwards PB Riser KL (1999) Tempo-ral and spatial scales of kelp demography the role of theoceanographic climate Ecol Monogr 69219minus250

Dean TA Schroeter SC Dixon JD (1984) Effects of grazingby two species of sea urchins (Strongylocentrotus fran-ciscanus and Lytechinus anamesus) on recruitment andsurvival of two species of kelp (Macrocystis pyrifera andPterygophora californica) Mar Biol 78301minus313

DeYoung B Harris R Alheit J Beaugrand G Mantua NShannon L (2004) Detecting regime shifts in the oceandata considerations Prog Oceanogr 60143minus164

Doroff AM Estes JA Tinker MT Burn DM Evans TJ (2003)Sea otter population declines in the Aleutian archipel-ago J Mammal 8455minus64

Dudgeon SR Aronson RB Bruno JF Precht WF (2010) Phaseshifts and stable states on coral reefs Mar Ecol Prog Ser413201minus216

Duggins DO (1980) Kelp beds and sea otters an experi -mental approach Ecology 61447minus453

Duggins DO Simenstad CA Estes JA (1989) Magnificationof secondary production by kelp detritus in coastal mar-ine ecosystems Science 245170minus173

Dumont C Himmelman JH Russell MP (2004) Size-specificmovement of green sea urchins Strongylocentrotus droe-bachiensis on urchin barrens in eastern Canada MarEcol Prog Ser 27693minus101

Dunton KH Reimnitz E Schonberg S (1982) An arctic kelpcommunity in the Alaskan Beaufort Sea Arctic 35465minus484

Ebeling AW Laur DR Rowley RJ (1985) Severe storm distur-bances and reversal of community structure in a south-ern California kelp forest Mar Biol 84287minus294

Ebert EE (1968) California sea ottermdashcensus and habitat survey Underwat Nat 520minus23

Ebert TA Southon JR (2003) Red sea urchins (Strongylocen-trotus franciscanus) can live over 100 years confirmationwith A-bomb 14carbon Fish Bull 101915minus922

Edgar GJ Barrett NS Morton AJ Samson CR (2004) Effectsof algal canopy clearance on plant fish and macroinver-tebrate communities on eastern Tasmanian reefs J ExpMar Biol Ecol 31267minus87

Edwards PB Ebert TA (1991) Plastic responses to limitedfood availability and spine damage in the sea urchinStrongylocentrotus purpuratus J Exp Mar Biol Ecol 145205minus220

Edwards MS Konar B (2012) A comparison of dragon kelpEualaria fistulosa (Phaeophyceae) fecundity in urchinbarrens and nearby kelp beds throughout the Aleutianarchipelago J Phycol 48897minus901

Erlandson JM Kennett DJ Ingram BL Guthrie DA and others (1996) An archaeological and palaeontologicalchronology for Daisy Cove (CA-SMI-261) San MiguelIsland California Radiocarbon 38355minus373

Erlandson JM Rick TC Estes JA Graham MH Braje TJVellanoweth RL (2008) Sea otters shellfish and humansa 10000 year record from San Miguel Island CaliforniaJ Archaeol Sci 352144minus2152

Estes JA Palmisano JF (1974) Sea otters their role in struc-turing nearshore communities Science 1851058minus1060

Estes JA Smith NS Palmisano JF (1978) Sea otter predationand community organization in the western Aleutianislands Alaska Ecology 59822minus833

Estes JA Tinker MT Williams TM (1998) Killer whale predation on sea otters linking oceanic and nearshoreecosystems Science 282473minus476

Fagerli CW Norderhaug KM Christie HC (2013) Lack ofsea urchin settlement may explain kelp forest recoveryin overgrazed areas in Norway Mar Ecol Prog Ser 488119minus132

Fanelli G Piraino S Belmonte G Geraci S Boero F (1994)Human predation along Apulian rocky coasts (SE Italy)desertification caused by Lithophaga lithophaga (Mol-lusca) fisheries Mar Ecol Prog Ser 1101minus8

Feehan C Scheibling RE Lauzon-Guay JS (2012) Aggrega-tive feeding behavior in sea urchins leads to destructivegrazing in a Nova Scotian kelp bed Mar Ecol Prog Ser44469minus83

Feehan CJ Johnson-Mackinnon J Scheibling RE Lauzon-Guay JS Simpson AGB (2013) Validating the identity ofParamoeba invadens the causative agent of recurrentmass mortality of sea urchins in Nova Scotia CanadaDis Aquat Org 103209minus227

Filbee-Dexter K Scheibling RE (2012) Hurricane-mediateddefoliation of kelp beds and pulsed delivery of kelp detri-tus to offshore sedimentary habitats Mar Ecol Prog Ser45551minus64

Fletcher WJ (1987) Interactions among subtidal Australiansea urchins gastropods and algae effects of experimen-tal removals Ecol Monogr 5789minus109

Folke C Carpenter S Walker B Scheffer M Elmqvist TGunderson L Holling CS (2004) Regime shifts resili-ence and biodiversity in ecosystem management AnnuRev Ecol Syst 35557minus581

Ford T Meux B (2010) Giant kelp community restoration inSanta Monica Bay Urban Coast 243minus46

Foreman RE (1977) Benthic community modification andrecovery following intensive grazing by Strongylocen-trotus droebachiensis Helgol Meeresunters 30468minus484

Fujita D (1998) Strongylocentrotid sea urchin-dominatedbarren grounds on the Sea of Japan coast of northernJapan In Mooi R Telford M (eds) Echinoderms BalkemaRotterdam p 659minus664

Fujita D (2010) Current status and problems of isoyake inJapan Bull Fish Res Agency 3233minus42

Fung T Seymour RM Johnson CR (2011) Alternative stablestates and phase shifts in coral reefs under anthro-pogenic stress Ecology 92967minus982

Gagnon P Himmelman JH Johnson LE (2004) Temporalvariation in community interfaces kelp-bed boundarydynamics adjacent to persistent urchin barrens Mar Biol1441191minus1203

Gaylord B Reed DC Raimondi PT Washborn L (2006)Macroalgae spore dispersal in coastal environmentsmechanistic insights revealed by theory and experimentEcol Monogr 76481minus502

Gaylord B Nickols KJ Jurgens L (2012) Roles of transportand mixing processes in kelp forest ecology J Exp Biol215997minus1007

Giakoumi S Cebrian E Kokkoris GD Ballesteros E Sala E(2012) Relationships between fish sea urchins andmacroalgae the structure of shallow rocky sublittoralcommunities in the Cyclades Eastern MediterraneanEstuar Coast Shelf Sci 1091minus10

Gianguzza P Bonaviri C Milisenda G Barcellona AAgnetta D Vega Fernaacutendez T Badalamenti F (2010)Macroalgal assemblage type affects predation pressureon sea urchins by altering adhesion strength Mar Envi-ron Res 7082minus86

Gladstone W Masens O (2009) Monitoring and assessmentof urchin barrens as a marine condition indicator Hunter-

Central Rivers Catchment Management Authority Uni-versity of Newcastle

Glud RN Woelfel J Karsten U Kuumlhl M Rysgaard S (2009)Benthic microalgal production in the Arctic appliedmethods and status of the current database Bot Mar52559minus572

Graham MH (2004) Effects of local deforestation on thediversity and structure of Southern California giant kelpforest food webs Ecosystems 7341minus357

Graham MH (2010) Comparisions between East-Asian iso -yake and deforestation in global kelp ecoystems BullFish Res Agen 3247ndash50

Gray JS (2001) Marine diversity the paradigms in patternsof species richness examined Sci Mar 6541minus56

Grimm V Wissel C (1997) Babel or the ecological stabilitydiscussions an inventory and analysis of terminologyand a guide for avoiding confusion Oecologia 109323minus334

Guidetti P Fanelli G Fraschetti S Terlizzi A Boero F (2002)Coastal fish indicate human-induced changes in theMediterranean littoral Mar Environ Res 5377minus94

Guidetti P Fraschetti S Terlizzi A Boero F (2003) Distribu-tion patterns of sea urchins and barrens in shallow Medi-terranean rocky reefs impacted by the illegal fishery ofthe rock-boring mollusc Lithophaga lithophaga Mar Biol1431135minus1142

Hacker SD Steneck RS (1990) Habitat architecture and theabundance and body-size-dependent habitat selection ofa phytal amphipod Ecology 712269minus2285

Hagen NT (1995) Recurrent destructive grazing of succes-sionally immature kelp forests by green sea urchins inVestfjorden Northern Norway Mar Ecol Prog Ser 12395minus106

Harley CD Anderson KM Demes KW Jorve JP Kordas RLCoyle TA Graham MH (2012) Effects of climate changeon global seaweed communities J Phycol 481064minus1078

Harrold C Reed DC (1985) Food availability sea urchingrazing and kelp forest community structure Ecology661160minus1169

Hart MW Scheibling RE (1988) Heat waves baby boomsand the destruction of kelp beds by sea urchins Mar Biol99167minus176

Hernaacutendez JC Clemente S Sangil C Brito A (2008) Actualstatus of the sea urchin Diadema antillarum populationsand macroalgal cover in marine protected areas com-pared to a highly fished area (Canary IslandsmdasheasternAtlantic Ocean) Aquat Conserv 181091minus1108

Hernaacutendez JC Clemente S Girard D Peacuterez-Ruzafa Aacute BritoA (2010) Effect of temperature on settlement and post -settlement survival in a barrens-forming sea urchin MarEcol Prog Ser 41369minus80

Himmelman JH (1986) Population biology of sea urchins onrocky barrens Mar Ecol Prog Ser 33295minus306

Hjorleifsson E Kassa O Gunnarsson K (1995) Grazing ofkelp by green sea urchins in Eyyjafjordu North IcelandIn Skjoldal HR Hopkins C Erikstad KK Leinass HP(eds) Ecology of fjords and coastal waters Elsevier Ams-terdam p 593minus597

Hobday AJ (2000) Persistence and transport of fauna ondrifting kelp (Macrocystis pyrifera (L) C Agardh) rafts inthe Southern California Bight J Exp Mar Biol Ecol 25375minus96

Hooper RG (1975) Bonne Bay marine resources An ecologi-cal and biological assessment Parks Canada Halifax

Hop H Pearson T Hegseth EN Kovacs KM Wiencke C

Kwasniewski S Gerland S (2002) The marine ecosystemof Kongsfjorden Svalbard Polar Res 21167minus208

Huggett M King C Williamson JE Steinberg P (2005) Lar-val development and metamorphosis of the Australiandiadematid sea urchin Centrostephanus rodgersii Inver-tebr Reprod Dev 47197minus204

Jackson JB (1997) Reefs since Columbus Coral Reefs 16S23minusS32

Jamieson GS Campbell A (1995) Red sea urchins and kelpin northern British Columbia In Skjoldal HR HopkinsC Erikstad KE Leinaas HP (eds) Ecology of fjords andcoastal waters Elsevier Science Amsterdam p 537minus547

Johnson CR Mann KH (1988) Diversity patterns of adap -tation and stability of Nova Scotian kelp beds EcolMonogr 58129minus154

Johnson CR Ling SD Ross DJ Shepherd S Miller KJ (2005)Establishment of the long-spined sea urchin (Cen-trostephanus rodgersii) in Tasmania first assessment ofpotential threats to fisheries Project Report School ofZoology and Tasmanian Aquaculture and Fisheries Institute Hobart

Johnson CR Banks SC Barrett NS Cazassus F and others(2011) Climate change cascades shifts in oceanographyspeciesrsquo ranges and subtidal marine community dynam-ics in eastern Tasmania J Exp Mar Biol Ecol 40017minus32

Johnson TR Wilson JA Cleaver C Morehead G Vadas R(2013) Modeling fine scale urchin and kelp dynamicsImplications for management of the Maine sea urchinfishery Fish Res 141107minus117

Jones GM (1985) Paramoeba invadens n sp (AmoebidaParamoebidae) a pathogenic amoeba from the sea urchinStrongylocentrotus droebachiensis in Eastern CanadaJ Protozool 32564minus569

Jones NS Kain JM (1967) Subtidal algal colonization follow-ing the removal of Echinus Helgol Meeresunters 15460minus466

Kannan K Agusa T Perrotta E Thomas NJ Tanabe S (2006)Comparison of trace element concentrations in livers ofdiseased emaciated and non-diseased southern sea ottersfrom the California coast Chemosphere 652160minus2167

Kawamata S (2010) Inhibitory effects of wave action ondestructive grazing by sea urchins a review Bull FishRes Agency 3295minus102

Keats DW (1991) Refugial Laminaria abundance and reduc-tion in urchin grazing in communities in the NorthwestAtlantic J Mar Biol 71867minus876

Keats DW South GR Steele DH (1990) The effects of anexperimental reduction in grazing by green sea urchinson a benthic macroalgal community in eastern New-foundland Mar Ecol Prog Ser 68181minus193

Kelly JR Scheibling RE Balch T (2011) Invasion-mediatedshifts in the macrobenthic assemblage of a rocky subtidalecosystem Mar Ecol Prog Ser 43769minus78

Kitching JA (1987) Ecological studies at Lough Hyne AdvEcol Res 17115minus189

Knowlton N (2004) Multiple lsquostablersquo states and the conserva-tion of marine ecosystems Prog Oceanogr 60387minus396

Konar B (2000) Seasonal inhibitory effects of marine plantson sea urchins structuring communities the algal wayOecologia 125208minus217

Konar B Estes JA (2003) The stability of boundary regionsbetween kelp beds and deforested areas Ecology 84174minus185

Krause Jensen D Marbagrave N Olesen B Sejr MK and others(2012) Seasonal sea ice cover as principal driver of

spatial and temporal variation in depth extension andannual production of kelp in Greenland Glob ChangeBiol 182981minus2994

Krumhansl KA Scheibling RE (2011) Detrital production inNova Scotian kelp beds patterns and processes MarEcol Prog Ser 42167minus82

Krumhansl KA Scheibling RE (2012) Production and fate ofkelp detritus Mar Ecol Prog Ser 467281minus302

Lamb I Zimmerman M (1964) Marine vegetation on CapeAnne Essex County Massachusetts Rhodora 66217minus254

Lauzon-Guay JS Scheibling RE (2007a) Seasonal variationin movement aggregation and destructive grazing of thegreen sea urchin (Strongylocentrotus droebachiensis) inrelation to wave action and sea temperature Mar Biol1512109minus2118

Lauzon-Guay JS Scheibling RE (2007b) Behaviour of seaurchin Strongylocentrotus droebachiensis grazing frontsfood-mediated aggregation and density-dependent facil-itation Mar Ecol Prog Ser 329191minus204

Lauzon-Guay JS Scheibling RE (2007c) Importance of spa-tial population characteristics on the fertilization rates ofsea urchins Biol Bull 212195minus205

Lauzon-Guay JS Scheibling RE Barbeau MA (2009) Model-ling phase shifts in a rocky subtidal ecosystem Mar EcolProg Ser 37525minus39

Lawrence JM (1975) On the relationships between marineplants and sea urchins Oceanogr Mar Biol Annu Rev 13213minus286

Lee RKS (1973) General ecology of the arctic benthic marinealgae Arctic 2632minus43

Lees K Pitois S Scott C Frid C Mackinson S (2006) Charac-terizing regime shifts in the marine environment FishFish 7104minus127

Leinaas HP Christie H (1996) Effects of removing seaurchins (Strongylocentrotus droebachiensis) stability ofthe barren state and succession of kelp forest recovery inthe east Atlantic Oecologia 105524minus536

Leleu K Remy-Zephir B Grace R Costello MJ (2012) Map-ping habitats in a marine reserve showed how a 30-yeartrophic cascade altered ecosystem structure Biol Con-serv 155193minus201

Levin PS (1994) Small-scale recruitment variation in a tem-perate fish the roles of macrophytes and food supplyEnviron Biol Fishes 40271minus281

Lewontin RC (1969) Meaning of stability Brookhaven SympBiol 2213minus24

Ling SD (2008) Range expansion of a habitat-modifying spe-cies leads to loss of taxonomic diversity a new andimpoverished reef state Oecologia 156883minus894

Ling SD Johnson CR (2012) Marine reserves reduce risk ofclimate-driven phase shift by reinstating size-and habi-tat-specific trophic interactions Ecol Appl 221232minus1245

Ling SD Johnson CR Frusher SD Ridgeway KR (2009)Overfishing reduces resilience of kelp beds to climate-driven catastrophic phase shift Proc Natl Acad Sci USA10622341minus22345

Mann KH (1973) Seaweeds their productivity and strategyfor growth Science 182975minus981

Marzloff MP Johnson CR Little LR Soulie JC Ling SDFrusher SD (2013) Sensitivity analysis and pattern-oriented validation of TRITON a model with alternativecommunity states insights on temperate rocky reefsdynamics Ecol Model 25816ndash32

Mathieson AC Moore GE Short FT (2010) A floristic com-parison of seaweeds from James Bay and three contigu-

ous Northeastern Canadian Arctic sites Rhodora 112396minus434

Matsunaga K Kawaguchi T Suzuki Y Nigi G (1999) Therole of terrestrial humic substances on the shift of kelpcommunity to crustose coralline algae community of thesouthern Hokkaido Island in the Japan Sea J Exp MarBiol Ecol 241193minus205

May RM (1977) Thresholds and breakpoints in ecosystemswith a multiplicity of stable states Nature 269471minus476

McLean JH (1962) Sublittoral ecology of kelp beds of theopen coast area near Carmel California Biol Bull 12295minus114

McNaught DC (1999) The indirect effects of macroalgae andmicropredation on the post-settlement success of thegreen sea urchin in Maine PhD dissertation Universityof Maine Orono ME

Meidel SK Scheibling RE (1998) Annual reproductive cycleof the green sea urchin Strongylocentrotus droebachi -ensis in differing habitats in Nova Scotia Canada MarBiol 131461minus478

Meidel SK Scheibling RE (2001) Variation in egg spawningamong subpopulations of sea urchins Strongylocentrotusdroebachiensis a theoretical approach Mar Ecol ProgSer 21397minus110

Mikhaylova TA (1999) The initial stages of experimentalforming of Laminaria communities in the White Sea BotZh 8456ndash66

Miller RJ (1985) Succession in sea urchin and seaweedabundance in Nova Scotia Canada Mar Biol 84275minus286

Miller RJ Nolan SC (2000) Management of the Nova Scotiasea urchin fishery a nearly successful habitat basedmanagement regime Fisheries and Oceans ScienceCanadian Stock Assessment Secretariat Ottawa

Miner CM Altstatt JM Raimondi PT Minchinton TE (2006)Recruitment failure and shifts in community structurefollowing mass mortality limit recovery prospects ofblack abalone Mar Ecol Prog Ser 327107minus117

Mohr JL Wilimovsky NJ Dawson EY (1957) An arcticAlaskan kelp bed Arctic 1045minus52

Moore DS Miller RJ (1983) Recovery of macroalgae follow-ing widespread sea urchin mortality with a description ofthe nearshore hard-bottom habitat on the Atlantic coastof Nova Scotia Can Tech Rep Fish Aquat Sci 1230

Mumby PJ Hastings A Edwards HJ (2007) Thresholds andthe resilience of Caribbean coral reefs Nature 450 98minus101

Norderhaug KM Christie HC (2009) Sea urchin grazing andkelp re-vegetation in the NE Atlantic Mar Biol Res 5515minus528

North WJ (1971) The biology of giant kelp beds (Macrocys-tis) in California introduction and background NovaHedwigia 321minus68

Norton TA (1992) Dispersal by macroalgae Br Phycol J27293minus301

Oshurkov VV Bazhin AG Lukin VI Sevostyanov VF (1988)Sea otter predation and the benthic community structureof Commander Islands Biol Mora 650minus60

Paine RT (1969) A note on trophic complexity and commu-nity stability Am Nat 10391minus93

Pearce CM Scheibling RE (1990) Induction of metamorpho-sis of larvae of the green sea urchin Strongylocentrotusdroebachiensis by coralline red algae Biol Bull 179304minus331

Pearse JS Hines AH (1979) Expansion of a central Californiakelp forest following the mass mortality of sea urchinsMar Biol 5183minus91

Pearse JS Clark ME Leighton DL Mitchell CT North WJ(1970) Marine waste disposal and sea urchin ecology InNorth WJ (ed) Kelp Habitat Improvement Project Cali-fornia Annual Report 1969 to 1970 California Instituteof Technology Pasadena CA p 1minus93

Pecorino D Lamare MD Barker MF (2012) Growth mor-phometrics and size structure of the Diadematidae seaurchin Centrostephanus rodgersii in northern NewZealand Mar Freshw Res 63624minus634

Peterson CH (1984) Does a rigorous criterion for environ-mental identity preclude the existence of multiple stablepoints Am Nat 124127minus133

Petraitis PS Dudgeon SR (2004) Detection of alternative sta-ble states in marine communities J Exp Mar Biol Ecol300343minus371

Petraitis PS Latham RE (1999) The importance of scale intesting the origins of alternative community states Ecology 80429minus442

Pinnegar JK Polunin NVC Francour P Badalamenti F andothers (2000) Trophic cascades in benthic marine ecosys-tems lessons for fisheries and protected-area manage-ment Environ Conserv 27179minus200

Propp MV (1977) Ecology of the sea urchin Strongylocentro-tus droebachiensis of the Barents Sea metabolism andregulation of abundance Sov J Mar Biol 327minus37

Raffaelli D Hawkins S (1996) Intertidal ecology Chapman ampHall London

Round FE (1967) The ecology of benthic algae In JacksonDE (ed) Algae and man Plenum Press New York NYp 138minus184

Sala E Boudouresque CF Harmelin-Vivien M (1998) Fish-ing trophic cascades and the structure of algal assem-blages evaluation of an old but untested paradigmOikos 82425minus439

Scheffer M Carpenter S Foley JA Folke C Walker B (2001)Catastrophic shifts in ecosystems Nature 413591minus596

Scheibling RE (1984) Echinoids epizootics and ecologicalstability in the rocky subtidal off Nova Scotia CanadaHelgol Meeresunters 37233minus242

Scheibling RE (1986) Increased macroalgal abundance fol-lowing mass mortalities of sea urchins (Strongylocentro-tus droebachiensis) along the Atlantic coast of Nova Scotia Oecologia 68186minus198

Scheibling RE (1996) The role of predation in regulating seaurchin populations in eastern Canada Oceanol Acta19421minus430

Scheibling RE Hamm J (1991) Interactions between seaurchins (Strongylocentrotus droebachiensis) and theirpredators in field and laboratory experiments Mar Biol110105minus116

Scheibling RE Lauzon-Guay JS (2010) Killer storms NorthAtlantic hurricanes and disease outbreaks in sea urchinsLimnol Oceanogr 552331minus2338

Scheibling RE Hennigar AW Balch T (1999) Destructivegrazing epiphytism and disease the dynamics of seaurchinminuskelp interactions in Nova Scotia Can J FishAquat Sci 562300minus2314

Scheibling RE Feehan CJ Lauzon-Guay JS (in press) Cli-mate change disease and the dynamics of a kelp-bedecosystem in Nova Scotia In Fernaacutendez-Palocios JMNascimiento LD Hernaacutendez JC Clement S Gonzaacutelez ADiaz-Gonzaacutelez JP (eds) Climate change perspectivesfrom the Atlantic past present and future Servicio dePublicaciones de la Universidad de La Laguna Tenerifep 41minus81

Schiel DR (1990) Macroalgal assemblages in New Zealandstructure interactions and demography Hydrobiologia19259minus76

Schiel DR Andrew NL Foster MS (1995) The structure ofsubtidal algal and invertebrate assemblages at theChatham Islands New Zealand Mar Biol 123355minus367

Sharp G Allard M Lewis A Semple R Rochefort G (2008)The potential for seaweed resource development in sub-arctic Canada Nunavik Ungava Bay J Appl Phycol 20491ndash498

Shears NT Babcock RC (2003) Continuing trophic cascadeeffects after 25 years of no-take marine reserve protec-tion Mar Ecol Prog Ser 2461minus16

Shears NT Babcock RC (2007) Quantitative description ofmainland New Zealandrsquos shallow subtidal reef communi-ties Science for Conservation 280 Department of Con-servation Wellington

Shears NT Grace RV Usmar NR Kerr V Babcock RC (2006)Long-term trends in lobster populations in a partiallyprotected vs no-take marine park Biol Conserv 132222minus231

Simenstad CA Estes JA Kenyon KW (1978) Aleuts seaotters and alternate stable-state communities Science200403minus411

Sivertsen K (1996) Incidence occurrence and distribution ofthe nematode Echinomermella matsi in its echinoid hostStrongylocentrotus droebachiensis in northern NorwayMar Biol 126703minus714

Sivertsen K (1997) Geographic and environmental factorsaffecting the distribution of kelp beds and barrengrounds and changes in biota associated with kelpreduction at sites along the Norwegian coast Can J FishAquat Sci 542872minus2887

Sivertsen K (2006) Overgrazing of kelp beds along the coastof Norway J Appl Phycol 18599minus610

Sjoslashtun K Fredriksen S Rueness J (1998) Effect of canopybiomass and wave exposure on growth in Laminariahyperborea (Laminariaceae Phaeophyta) Eur J Phycol33337minus434

Skadsheim A Christie H Leinaas HP (1995) Populationreductions of Strongylocentrotus droebachiensis (Echin-odermata) in Norway and the distribution of its endo -parasite Echinomermella matsi (Nematoda) Mar EcolProg Ser 119199minus209

Steneck RS (1997) Fisheries-induced biological changesto the structure and function of the Gulf of Maine eco -system In Wallace GT Braasch EF (eds) Proceedingsof the Gulf of Maine Ecosystem Dynamics ScientificSymposium and Workshop Regional Association forResearch on the Gulf of Maine Hanover MEp 151ndash165

Steneck RS Dethier MN (1994) A functional group ap -proach to the structure of algal-dominated communitiesOikos 69476minus498

Steneck RS Graham MH Bourque BJ Corbett D ErlandsonJM Estes JA Tegner MJ (2002) Kelp forest ecosystemsbiodiversity stability resilience and future EnvironConserv 29436minus459

Steneck RS Vavrinec J Leland AV (2004) Acceleratingtrophic-level dysfunction in kelp forests ecosystems ofthe Western North Atlantic Ecosystems 7323minus332

Steneck RS Leland A McNaught DC Vavrinec J (2013)Ecosystem flips locks and feedbacks the lasting effectsof fisheries on Mainersquos kelp forest ecosystem Bull MarSci 8931minus55

Sutherland JP (1974) Multiple stable points in natural com-munities Am Nat 108859minus873

Tamaki H Takahashi H Fukaya A Fukuda M Arai SMuraoka D (2005) The distributional patterns of Undariapinnatifida (Harvey) Suringar and sea urchins related towater velocity condition in Ogatsu Bay Miyagi Prefec-ture Jpn J Phycol 53131minus135

Tamaki H Kusaka K Fukuda M Arai S Muraoka D (2009)Undaria pinnatifida habitat loss in relation to sea urchingrazing and water flow conditions and their restorationeffort in Ogatsu Bay Japan J Water Environ Technol7201minus213

Tegner MJ Dayton PK (1981) Population structure recruit-ment and mortality of two sea urchins (Strongylocentro-tus franciscanus and S purpuratus) in a kelp forest MarEcol Prog Ser 5255minus268

Tegner MJ Dayton PK (1987) El Nintildeo effects on southernCalifornia kelp forest communities Adv Ecol Res 17243minus279

Tegner MJ Dayton PK (1991) Sea urchins El Nintildeos andthe long term stability of Southern California kelp forestcommunities Mar Ecol Prog Ser 7749minus63

Tegner MJ Dayton PK (2000) Ecosystem effects of fishing inkelp forest communities ICES J Mar Sci 57579minus589

Tegner MJ Dayton PK Edwards PB Riser KL (1996) Is thereevidence for the long-term climatic change in southernCalifornia kelp forests Calif Coop Ocean Fish InvestigRep 37111minus126

Tegner MJ Dayton PK Edwards PB Riser KL (1997) Large-scale low-frequency oceanographic effects on kelp for-est succession a tale of two cohorts Mar Ecol Prog Ser146117minus134

Thibaut T Pinedo S Torras X Ballesteros E (2005) Long-term decline of the populations of Fucales (Cystoseiraspp and Sargassum spp) in the Albegraveres coast (FranceNorth-western Mediterranean) Mar Pollut Bull 501472minus1489

Trowbridge CD Little C Pilling GM Stirling P Miles A(2011) Decadal-scale changes in the shallow subtidalbenthos of an Irish marine reserve Bot Mar 54497minus506

Tuya F Sanchez-Jerez P Haroun RJ (2005) Influence of fish-ing and functional group of algae on sea urchin control ofalgal communities in the eastern Atlantic Mar Ecol ProgSer 287255minus260

Vadas RL Steneck RS (1988) Zonation of deep water benthicalgae in the Gulf of Maine J Phycol 24338minus346

Valentine JP Johnson CR (2005) Persistence of sea urchin(Heliocidaris erythrogramma) barrens on the east coastof Tasmania inhibition of macroalgal recovery in theabsence of high densities of sea urchins Bot Mar 48106minus115

Vanderklift MA Kendrick GA (2005) Contrasting influenceof sea urchins on attached and drift macroalgae MarEcol Prog Ser 299101minus110

Vaacutesquez JA (1992) Lessonia trabeculata a subtidal bottomkelp in northern Chile a case study for a structural andgeographical comparisons In Seeliger U (ed) Coastalplants of Latin America Academic Press San Diego CAp 77minus89

Vaacutesquez JA (2008) Production use and fate of Chileanbrown seaweeds resources for a sustainable fisheryJ Appl Phycol 20457minus467

Vaacutesquez JA Buschmann AH (1997) Herbivoreminuskelp interac-tions in Chilean subtidal communities a review Rev ChilHist Nat 7041minus52

Vaacutesquez JA Castilla JC Santelices B (1984) Distributionalpatterns and diets of four species of sea urchins in giantkelp forest (Macrocystis pyrifera) of Puerto ToroNavarino Island Chile Mar Ecol Prog Ser 1955minus63

Vaacutesquez JA Vega JA Buschmann AH (2006) Long termvariability in the structure of kelp communities in north-ern Chile and the 1997minus98 ENSO J Appl Phycol 18505minus519

Vega JMA Vasquez JA Buschmann AH (2005) Populationbiology of the subtidal kelps Macrocystis integrifolia andLessonia trabeculata (Laminariales Phaeophyta) in anupwelling ecosystem of northern Chile interannual vari-ability and El Nintildeo 1997minus98 Rev Chil Hist Nat 7833minus50

Watanuki A Aota T Otsuka E Kawai T Iwahashi Y Kuwa-hara H Fujita D (2010) Restoration of kelp beds on anurchin barren Removal of sea urchins by citizen diversin southwestern Hokkaido Bull Fish Res Agency 3283minus87

Watson J Estes JA (2011) Stability resilience and phaseshifts in rocky subtidal communities along the west coastof Vancouver Island Canada Ecol Monogr 81215minus239

Wernberg T Russell BD Moore PJ Ling SD and others(2011) Impacts of climate change in a global hotspot

for temperate marine biodiversity and ocean warmingJ Exp Mar Biol Ecol 4007ndash16

Wharton WG (1980) The distribution of sea urchin-dominated barren grounds along the south shore of NovaScotia Can Tech Rep Fish Aquat Sci 95433minus47

Wharton WG Mann KH (1981) Relationship betweendestructive grazing by the sea urchin Strongylocentro-tus droebachiensis and the abundance of American lob-ster Homarus americanus on the Atlantic coast of NovaScotia Can J Fish Aquat Sci 381339minus1349

Wiencke CM Clayton N Goacutemez I Iken K and others (2007)Life strategy ecophysiology and ecology of seaweedsin polar waters In Amils R Ellis-Evans C Hinghofer-Szalkay HG (eds) Life in extreme environments SpringerDordrecht p 213ndash244

Wilce RT Pedersen PM Sekida S (2009) Chukchia pedi -cellata gen et sp nov and C endophyticanov combArctic endemic brown algae (Phaeophyceae) J Phycol45 272ndash286

Wilmers CC Estes JA Edwards M Laidre KL Konar B(2012) Do trophic cascades affect the storage and flux ofatmospheric carbon An analysis of sea otters and kelpforests Front Ecol Environ 10409minus415

Region Kelp species Source

Alaska USAStefensson Sound Laminaria solidungula Saccharina latissima Dunton et al (1982)Demarcation Point L solidungula S latissima Wiencke et al (2007)Camdem Bay L solidungula S latissima Wiencke et al (2007)Chukchi Sea L solidungula S latissima Mohr et al (1957)Prince Patrick Island S latissima Wiencke et al (2007)

Canadian ArcticBylot Island S latissima Wilce et al (2009)Cape Hatt S latissima Alaria esculenta Cross et al (1987)Pangnirtung Fiord Laminaria sp Cross et al (1987)Brock Island L solidungula Lee (1973)Ungava Bay A esculenta L solidungula L digitata S latissima Sharp et al (2008)Lancaster South A esculenta L solidungula S groenlandica S latissima Cross et al (1987)Foxe Basin A esculenta L solidungula S groenlandica S latissima Chapman amp Lindley (1981)Hudson Bay A esculenta L solidungula L digitata S latissima Mathieson et al (2010)

GreenlandSiorapoluk to Nuuk S latissima Agarum clathratum L solidungula Krause-Jensen et al (2012)Disko Island S latissima Bischoff amp Wiencke (1993)Young Sound S latissima Glud et al (2009)

Northern EuropeSvalbard L digitata L solidungula S latissima A esculenta Saccorhiza Hop et al (2002)

dermatodeaKingsfjorden L digitata L solidungula S latissima A esculenta S dermatodea Wiencke et al (2007)White Sea L digitata L hyperborea S latissima A esculenta S dermatodea Mikhaylova (1999)

Appendix Documented kelp beds in the Beaufort Sea Canadian Arctic Greenland and northern Europe

Editorial responsibility Charles Peterson Morehead City North Carolina USA

Submitted June 27 2013 Accepted September 23 2013Proofs received from author(s) November 1 2013

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

tem from one stable (ie robust to relatively smallperturbations) community state to another (Lawrence1975 Steneck et al 2002) Sea urchin barrens havemuch lower primary productivity and habitat struc-tural complexity than kelp beds and consequentlycan be considered a collapse of the kelp state(Simenstad et al 1978 Chapman amp Johnson 1990Sivertsen 1996 Graham 2004 Christie et al 2009)Since kelp beds are key components of coastal eco-systems that provide important services to residentcommunities (Mann 1973 Levin 1994 Krumhansl

amp Scheibling 2012) understanding the factors thatcause phase shifts to urchin barrens and that enablekelp beds to recover is crucial for the proper man-agement of these ecosystems

Of particular concern to managers is the possibilitythat sea urchin barrens are a stable state of the sub-tidal ecosystem maintained by various feedbackmechanisms that prevent recovery of the kelp-dominated state after the initial driver of the phaseshift has been relaxed or reversed (Lauzon-Guay etal 2009 Ling et al 2009) This type of transition is

Fig 1 (A) Destructive grazing front of sea urchins Strongylocentrotus droebachiensis advancing into a kelp bed near HalifaxNova Scotia Canada Photo credit R E Scheibling (B) Extensive urchin (S polyacanthus) barrens in the Aleutian IslandsUSA Photo credit B Konar (C) Urchins S droebachiensis on scoured coralline algae in barrens in Norway Photo credit C WFagerli (D) Range-expanding urchin Centrostephanus rodgersii forming patchy barrens in a kelp bed in southeast Tasmania

Photo credit S D Ling (E) S nudus grazing a kelp bed in Japan Photo credit D Fujita

Term Definition

Examples

sl amp

Cminus

minus19

85minus

)Brarr

minus19

70minus

es amp

minus19

70minus

ABrarr

nminus

minus79

87minus

aBrarr

minus20

AKrarr

79minus

73minus

sBrarr

minus77

minus19

59minus

intildea

minus20

aacutesq

minus96

Vaacutes

ez amp

minus19

AKrarr

sminus20

minus98

Bminus

minus87

)Brarr

seminus

00minus

re amp

minus19

78minus

(Krarr

r Brarr

ndash n

Kminus

minusA

66minus

minus10

ayKrarr

minus93

seKrarr

70sminus

)Brarr

minus18

minus91

dBrarr

minus20

minus93

yminus

minus08

sKrarr

01minus

aacutend

gminus

02minus

aacutend

aBrarr

minus87

minus09

minusminus

78minus

seKrarr

5minus20

minus20

minus01

dKrarr

0)Krarr

78minus

minus79

minus19

30sminus

alKrarr

minus60

sminus90

Kminus

minus07

iaBrarr

minus05

iaBrarr

5minus15

minus86

Urchin fishery

Otter extirpation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Urchin disease

Storm eventsa

Present

Urchin fishery

Present

Higher temperaturea

High predation

(marine reserves)

Groundfishcollapse

Otter recovery

Urchin increase

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Term Definition

Examples

sl amp

Cminus

minus19

85minus

)Brarr

minus19

70minus

es amp

minus19

70minus

ABrarr

nminus

minus79

87minus

aBrarr

minus20

AKrarr

79minus

73minus

sBrarr

minus77

minus19

59minus

intildea

minus20

aacutesq

minus96

Vaacutes

ez amp

minus19

AKrarr

sminus20

minus98

Bminus

minus87

)Brarr

seminus

00minus

re amp

minus19

78minus

(Krarr

r Brarr

ndash n

Kminus

minusA

66minus

minus10

ayKrarr

minus93

seKrarr

70sminus

)Brarr

minus18

minus91

dBrarr

minus20

minus93

yminus

minus08

sKrarr

01minus

aacutend

gminus

02minus

aacutend

aBrarr

minus87

minus09

minusminus

78minus

seKrarr

5minus20

minus20

minus01

dKrarr

0)Krarr

78minus

minus79

minus19

30sminus

alKrarr

minus60

sminus90

Kminus

minus07

iaBrarr

minus05

iaBrarr

5minus15

minus86

Urchin fishery

Otter extirpation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Urchin disease

Storm eventsa

Present

Urchin fishery

Present

Higher temperaturea

High predation

(marine reserves)

Groundfishcollapse

Otter recovery

Urchin increase

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Term Definition

Examples

sl amp

Cminus

minus19

85minus

)Brarr

minus19

70minus

es amp

minus19

70minus

ABrarr

nminus

minus79

87minus

aBrarr

minus20

AKrarr

79minus

73minus

sBrarr

minus77

minus19

59minus

intildea

minus20

aacutesq

minus96

Vaacutes

ez amp

minus19

AKrarr

sminus20

minus98

Bminus

minus87

)Brarr

seminus

00minus

re amp

minus19

78minus

(Krarr

r Brarr

ndash n

Kminus

minusA

66minus

minus10

ayKrarr

minus93

seKrarr

70sminus

)Brarr

minus18

minus91

dBrarr

minus20

minus93

yminus

minus08

sKrarr

01minus

aacutend

gminus

02minus

aacutend

aBrarr

minus87

minus09

minusminus

78minus

seKrarr

5minus20

minus20

minus01

dKrarr

0)Krarr

78minus

minus79

minus19

30sminus

alKrarr

minus60

sminus90

Kminus

minus07

iaBrarr

minus05

iaBrarr

5minus15

minus86

Urchin fishery

Otter extirpation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Urchin disease

Storm eventsa

Present

Urchin fishery

Present

Higher temperaturea

High predation

(marine reserves)

Groundfishcollapse

Otter recovery

Urchin increase

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Term Definition

Examples

sl amp

Cminus

minus19

85minus

)Brarr

minus19

70minus

es amp

minus19

70minus

ABrarr

nminus

minus79

87minus

aBrarr

minus20

AKrarr

79minus

73minus

sBrarr

minus77

minus19

59minus

intildea

minus20

aacutesq

minus96

Vaacutes

ez amp

minus19

AKrarr

sminus20

minus98

Bminus

minus87

)Brarr

seminus

00minus

re amp

minus19

78minus

(Krarr

r Brarr

ndash n

Kminus

minusA

66minus

minus10

ayKrarr

minus93

seKrarr

70sminus

)Brarr

minus18

minus91

dBrarr

minus20

minus93

yminus

minus08

sKrarr

01minus

aacutend

gminus

02minus

aacutend

aBrarr

minus87

minus09

minusminus

78minus

seKrarr

5minus20

minus20

minus01

dKrarr

0)Krarr

78minus

minus79

minus19

30sminus

alKrarr

minus60

sminus90

Kminus

minus07

iaBrarr

minus05

iaBrarr

5minus15

minus86

Urchin fishery

Otter extirpation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Urchin disease

Storm eventsa

Present

Urchin fishery

Present

Higher temperaturea

High predation

(marine reserves)

Groundfishcollapse

Otter recovery

Urchin increase

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

sl amp

Cminus

minus19

85minus

)Brarr

minus19

70minus

es amp

minus19

70minus

ABrarr

nminus

minus79

87minus

aBrarr

minus20

AKrarr

79minus

73minus

sBrarr

minus77

minus19

59minus

intildea

minus20

aacutesq

minus96

Vaacutes

ez amp

minus19

AKrarr

sminus20

minus98

Bminus

minus87

)Brarr

seminus

00minus

re amp

minus19

78minus

(Krarr

r Brarr

ndash n

Kminus

minusA

66minus

minus10

ayKrarr

minus93

seKrarr

70sminus

)Brarr

minus18

minus91

dBrarr

minus20

minus93

yminus

minus08

sKrarr

01minus

aacutend

gminus

02minus

aacutend

aBrarr

minus87

minus09

minusminus

78minus

seKrarr

5minus20

minus20

minus01

dKrarr

0)Krarr

78minus

minus79

minus19

30sminus

alKrarr

minus60

sminus90

Kminus

minus07

iaBrarr

minus05

iaBrarr

5minus15

minus86

Urchin fishery

Otter extirpation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Urchin disease

Storm eventsa

Present

Urchin fishery

Present

Higher temperaturea

High predation

(marine reserves)

Groundfishcollapse

Otter recovery

Urchin increase

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

sl amp

Cminus

minus19

85minus

)Brarr

minus19

70minus

es amp

minus19

70minus

ABrarr

nminus

minus79

87minus

aBrarr

minus20

AKrarr

79minus

73minus

sBrarr

minus77

minus19

59minus

intildea

minus20

aacutesq

minus96

Vaacutes

ez amp

minus19

AKrarr

sminus20

minus98

Bminus

minus87

)Brarr

seminus

00minus

re amp

minus19

78minus

(Krarr

r Brarr

ndash n

Kminus

minusA

66minus

minus10

ayKrarr

minus93

seKrarr

70sminus

)Brarr

minus18

minus91

dBrarr

minus20

minus93

yminus

minus08

sKrarr

01minus

aacutend

gminus

02minus

aacutend

aBrarr

minus87

minus09

minusminus

78minus

seKrarr

5minus20

minus20

minus01

dKrarr

0)Krarr

78minus

minus79

minus19

30sminus

alKrarr

minus60

sminus90

Kminus

minus07

iaBrarr

minus05

iaBrarr

5minus15

minus86

Urchin fishery

Otter extirpation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Urchin disease

Storm eventsa

Present

Urchin fishery

Present

Higher temperaturea

High predation

(marine reserves)

Groundfishcollapse

Otter recovery

Urchin increase

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Cminus

minus19

85minus

)Brarr

minus19

70minus

es amp

minus19

70minus

ABrarr

nminus

minus79

87minus

aBrarr

minus20

AKrarr

79minus

73minus

sBrarr

minus77

minus19

59minus

intildea

minus20

aacutesq

minus96

Vaacutes

ez amp

minus19

AKrarr

sminus20

minus98

Bminus

minus87

)Brarr

seminus

00minus

re amp

minus19

78minus

(Krarr

r Brarr

ndash n

Kminus

minusA

66minus

minus10

ayKrarr

minus93

seKrarr

70sminus

)Brarr

minus18

minus91

dBrarr

minus20

minus93

yminus

minus08

sKrarr

01minus

aacutend

gminus

02minus

aacutend

aBrarr

minus87

minus09

minusminus

78minus

seKrarr

5minus20

minus20

minus01

dKrarr

0)Krarr

78minus

minus79

minus19

30sminus

alKrarr

minus60

sminus90

Kminus

minus07

iaBrarr

minus05

iaBrarr

5minus15

minus86

Urchin fishery

Otter extirpation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Urchin disease

Storm eventsa

Present

Urchin fishery

Present

Higher temperaturea

High predation

(marine reserves)

Groundfishcollapse

Otter recovery

Urchin increase

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Kminus

minusA

66minus

minus10

ayKrarr

minus93

seKrarr

70sminus

)Brarr

minus18

minus91

dBrarr

minus20

minus93

yminus

minus08

sKrarr

01minus

aacutend

gminus

02minus

aacutend

aBrarr

minus87

minus09

minusminus

78minus

seKrarr

5minus20

minus20

minus01

dKrarr

0)Krarr

78minus

minus79

minus19

30sminus

alKrarr

minus60

sminus90

Kminus

minus07

iaBrarr

minus05

iaBrarr

5minus15

minus86

Urchin fishery

Otter extirpation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Urchin disease

Storm eventsa

Present

Urchin fishery

Present

Higher temperaturea

High predation

(marine reserves)

Groundfishcollapse

Otter recovery

Urchin increase

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Urchin fishery

Otter extirpation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Urchin disease

Storm eventsa

Present

Urchin fishery

Present

Higher temperaturea

High predation

(marine reserves)

Groundfishcollapse

Otter recovery

Urchin increase

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Urchin fishery

Otter extirpation

of kelp

Low recruitment

Urchin mortality

Groundfishcollapse

Urchin disease

Storm eventsa

Present

Urchin fishery

Present

Higher temperaturea

High predation

(marine reserves)

Groundfishcollapse

Otter recovery

Urchin increase

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Kelp state

Barren state

Urchin recruitment

Detritus

Destructive grazing

Detritus

Urchin migration

Predator abundance

mortality

Kelp recruitment

Spore production

Urchin migration

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

LITERATURE CITED

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

cite21

cite27

cite52

cite122

cite58

cite150

cite125

cite89

cite128

cite40

cite156

cite15

cite159

cite184

cite71

cite77

cite34

cite111

cite90

cite65

cite117

cite96

cite170

cite145

cite22

cite148

cite173

cite28

cite176

cite179

cite84

cite10

cite41

cite100

cite103

cite47

cite72

cite131

cite134

cite78

cite109

cite162

cite137

cite165

cite35

cite60

cite168

cite91

cite97

cite29

cite54

cite123

cite151

cite11

cite126

cite154

cite129

cite182

cite157

cite42

cite48

cite79

cite30

cite36

cite61

cite112

cite140

cite67

cite115

cite92

cite143

cite98

cite146

cite24

cite174

cite149

cite177

cite55

cite80

cite86

cite12

cite43

cite18

cite104

cite74

cite107

cite132

cite135

cite31

cite138

cite163

cite166

cite37

cite169

cite62

cite68

cite99

cite25

cite50

cite121

cite56

cite81

cite124

cite87

cite152

cite127

cite13

cite180

cite155

cite158

cite44

cite19

cite183

cite75

cite32

cite110

cite113

cite94

cite69

cite144

cite20

cite119

cite172

cite147

cite51

cite26

cite175

cite178

cite82

cite57

cite88

cite14

cite45

cite102

cite70

cite105

cite130

cite76

cite133

cite108

cite136

cite161

cite164

cite33

cite139

cite167

cite39

cite64

cite95

Documents

Sea urchin barrens as alternative stable states of collapsed … · Filbee-Dexter & Scheibling: Urchin barrens as alternative stable state termed a discontinuous phase shift (Fig