Table Of ContentHarmful Algae xxx (xxxx) xxx–xxx
ContentslistsavailableatScienceDirect
Harmful Algae
journal homepage: www.elsevier.com/locate/hal
Pelagic harmful algal blooms and climate change: Lessons from nature’s
experiments with extremes
Vera L. Trainera,⁎, Stephanie K. Moorea, Gustaaf Hallegraeffb, Raphael M. Kudelac,
Alejandro Clementd, Jorge I. Mardonese, William P. Cochlanf
aEnvironmentalandFisheriesSciencesDivision,NorthwestFisheriesScienceCenter,NationalMarineFisheriesService,NationalOceanicandAtmosphericAdministration,
2725MontlakeBlvdE,Seattle,WA98112,USA
bInstituteforMarineandAntarcticStudies,UniversityofTasmania,PrivateBag129,Hobart,Tasmania7001,Australia
cOceanSciencesDepartment,UniversityofCaliforniaSantaCruz,SantaCruz,CA95064,USA
dPlanctonAndinospA,LaboratorioPuertoVaras,Terraplén869,PuertoVaras,Chile
eCentrodeEstudiosdeAlgasNocivas(CREAN),InstitutodeFomentoPesquero(IFOP),PadreHarter574,PuertoMontt,Chile
fEstuary&OceanScienceCenter,RombergTiburonCampus,SanFranciscoStateUniversity,3150ParadiseDr.,Tiburon,CA94920,USA
ARTICLE INFO ABSTRACT
WededicatethisworktoProfessorTheodore Timeseriesnowhavesufficientdurationtodetermineharmfulalgalbloom(HAB)responsestochangingclimate
Smayda,knownbygenerationsofHAB conditions, including warming, stratification intensity, freshwater inputs and natural patterns of climate
scientistsasTed,forhisinsightsandgenerosity variability,suchastheElNiñoSouthernOscillationandPacificDecadalOscillation.Againstthecontextoftime
toothersinunravellingthemysteriesofHAB series, such as those available from phytoplankton monitoring, dinoflagellate cyst records, the Continuous
dynamics.Heisdeeplymissedintheworld-
PlanktonRecordersurveys,andshellfishtoxinrecords,itispossibletoidentifyextremeeventsthataresig-
wideHABcommunity.
nificantdeparturesfromlong-termmeans.Extremeweathereventscanmimicfutureclimateconditionsand
providea“dressrehearsal”forunderstandingfuturefrequency,intensityandgeographicextentofHABs.Three
Keywords:
casestudiesofextremeHABeventsaredescribedindetailtoexplorethedriversandimpactsoftheseoceanic
Harmfulalgalblooms
Pelagic outliersthatmaybecomemorecommoninthefuture.Oneexampleisthechain-formingdiatomofthegenus
Climatechange Pseudo-nitzschia inthe U.S. PacificNorthwest andits responseto the 2014-16 northeastPacific marineheat
Timeseries wave. The other two case studies are pelagic flagellates. Highly potent Alexandrium catenella group 1 dino-
Extremeevents flagellateblooms(upto150mg/kgPSTinmussels;4humanpoisonings)during2012-17createdhavocforthe
seafoodindustryinTasmania,south-easternAustralia,inapoorlymonitoredareawheresuchproblemswere
previouslyunknown.Earlyevidencesuggeststhatchangesinwatercolumnstratificationduringthecoldwinter-
springseasonaredrivingnewbloomscausedbyapreviouslycrypticspecies.AnexpansionofPseudochattonella
cf.verruculosatothesouthandA.catenellatothenorthoverthepastseveralyearsresultedintheconvergenceof
bothspeciestocausethemostcatastrophiceventinthehistoryoftheChileanaquacultureintheaustralsummer
of2016.Together,thesetwomassivebloomswerecolloquiallyknownasthe“Godzilla-Redtideevent”,resulting
inthelargestfishfarmmortalityeverrecordedworldwide,equivalenttoanexportlossofUSD$800million
whichwhencombinedwithshellfishtoxicity,resultedinmajorsocialunrestandrioting.Bothbloomswere
linkedtothestrongElNiñoeventandthepositivephaseoftheSouthernAnnularMode,thelatteranindicatorof
anthropogenicclimatechangeinthesoutheasternPacificregion.Foreachofthesethreeexamples,representing
recentcatastrophiceventsingeographicallydistinctregions,additionaltargetedmonitoringwasemployedto
improvetheunderstandingoftheclimatedriversandmechanismsthatgaverisetotheeventandtodocument
thesocietalresponse.ScientistsmustbepoisedtostudyfutureextremeHABeventsasthesenaturalexperiments
provideuniqueopportunitiestodefineandtestmultifactorialdriversofblooms.
1. Introduction oceanareunderthemostpressurebecausetheyrespondmorerapidly
andstronglytovariationsinatmosphericconditions,suchaswarming,
The world’s oceans are experiencing severe environmental and relativetodeeperwaters.Forexample,theglobalaveragewarmingof
ecological stress from climate change. The near surface layers of the theupper75moftheoceanfrom 1971to2010hasbeen0.11°Cper
⁎Correspondingauthor.
E-mailaddress:vera.l.trainer@noaa.gov(V.L.Trainer).
https://doi.org/10.1016/j.hal.2019.03.009
Received17March2019;Accepted18March2019
1568-9883/ © 2019 Published by Elsevier B.V.
Please cite this article as: Vera L. Trainer, et al., Harmful Algae, https://doi.org/10.1016/j.hal.2019.03.009
V.L.Trainer,etal. Harmful Algae xxx (xxxx) xxx–xxx
decade,decreasingto0.015°Cperdecadeat700mdepth(Rheinetal., species determined in ungrazed, monospecific, nutritionally optimal
2013).Conditionsintheupperoceanarealsomoresensitivetochanges cultures are used in models to evaluate potential range shifts and
infreshwaterinput,whichinfluencessalinityandverticalstratification changes in bloom phenology (e.g., Moore et al., 2011; Gobler et al.,
(ortheabilityofsurfacewaterstomixvertically),oceancurrents,and 2017). Townhill et al. (2018) offer a different approach that borrows
ocean biogeochemistry. In the future, the upper ocean is expected to fromlandscapeecology,usingobservedoccurrencesofHABspeciesin
undergoevenmoresignificantchangeastheglobalclimatecontinuesto naturetoidentifyrelationshipswithenvironmentalparametersthatare
warm, to include alterations in underwaterlight fields, thefrequency climate sensitive. However, extrapolating these relationships beyond
andintensityofupwelling,thermohalineoverturning,nutrientcycling the range of parameters observed introduces uncertainty in the pro-
fromdepth,precipitationandstorms(BoydandDoney,2002).Ofthe jectionsoffutureHABrisks.
organismsthatinhabittheupperocean,thephytoplankton,including Conventionistoacquireatleast30consecutiveyearsofmonitoring
thosethatcauseharmfulalgalblooms(HABs),willbeamongthefirstto ata siteto identifya “climatenormal” andlink phytoplanktonabun-
respondtochangesinoceanconditions.Thisisbecausetheyhaverapid dancerecordstoenvironmentalvariables(Daleetal.,2006).Inmany
generation times and short lifetimes (Irwin et al., 2015) and because regionsoftheworld,sustainedHABspeciesandtoxinmonitoringnow
their distribution, survival and success are closely controlled by the allowforananalysisoflong-termtrendsthatprovidesomeevidencefor
physicalandchemicalcharacteristicsofthewatermassesinwhichthey strong climate drivers of HABs over 30+ years. For example, mon-
reside. itoringofHABtoxinsinshellfishtoprotectpublichealthdatebackto
Mosteffortstobetterunderstandhowclimatechangewillinfluence the1950sinthePacificNorthwestoftheU.S.InPugetSound,aninland
HABshavefocusedontheeffectsofincreasedtemperature,akeystone fjordinWashingtonState,increasedconcentrationsinshellfishofpa-
parameter of climate change that is known to directly and indirectly ralytic shellfish poisoning (PSP) toxins, produced by dinoflagellate
affectphytoplanktoninavarietyofways.Temperatureinfluencescri- speciesofAlexandrium,covarywithwarmphasesofthePDO;however,
ticalphysiologicalprocessesinphytoplanktonsuchaspotentialgrowth nosuchrelationshipexistsforElNiñoevents(Mooreetal.,2011).In
rates, photosynthesis, motility, and nutrient acquisition rates (cf. re- contrast, on the outer coasts of Washington and Oregon, increased
viewbyWellsetal.,2015).Increasedtemperaturemayalterthephe- concentrationsofdomoicacid(DA),producedbyspeciesofthediatom
nologyandgeographyofsomeHABs.Forexample,seasonalwindowsof Pseudo-nitzschia, coincide with or closely follow the warm phases of
growthforsomeharmfulalgaemayexpandorcontractinsomeregions bothPDOandENSO(McKibbenetal.,2017;McCabeetal.,2016)but
(e.g., Moore et al., 2009), while other harmful algae may respond by this pattern does not hold forsites further south (Sekula-Woodet al.,
expanding or contracting their geographic range (Gobler et al., 2017; 2011).
Townhill et al., 2018). Temperature may indirectly affect phyto- The longest time series, dating back to the 1940s in the North
plankton populations also via changes in zooplankton grazing Atlantic, has been collected by the Continuous Plankton Recorder
(Lewandowska et al., 2014) which could dramatically influence phy- (CPR),auniqueplatformthatisusedtomonitorlargephytoplankton
toplanktoncommunitydiversity. and zooplankton. The frequency and geographical range of the CPR
Warming of the upper ocean will increase stratification of coastal samplingenablessamplestobecollectedoveralargerspatialscaleto
waters (Hays et al., 2005; Hallegraeff, 2010; Paerl and Scott, 2010), determine whether changes are more localized or widespread. The
suppressingnutrientexchangeviamixingwithdeeperwatersanden- limitationsofCPRdataincludeitscollectiononlyofsubsurfacesamples
hancing light levels in the upper stratified surface layer (Behrenfeld at 4–7m with poor taxonomic resolution unless techniques such as
etal.,2006).Thesubsequentrapiddepletionofsurfacenutrientsmay scanning electron microscopy are employed to confirm species iden-
favor phytoplankton, including harmful algae, with unique nutrient tities.However,thehigh-throughputnatureofsamplecollectiondoes
acquisition strategies. The strategies used by harmful algae to suc- convey some benefits. Although the CPR was originally designed to
cessfully compete under nutrient limiting conditions include mixo- collectzooplanktonwithits∼270μmmesh,thelargevolumesofwater
trophy, unique trace metal uptake capabilities, elevated ‘surge’ mac- filtered can contain high densities of phytoplankton that either effec-
ronutrient uptake rates, and swimming to areas where nutrients are tively reduce the mesh size or stick to the microthreads of the silk
more available (Smayda, 2010). Changes in the intensity of ocean material (Batten et al., 2003; Stern et al., 2018). Although there are
mixing or stratification, and the associated changes in nutrient avail- veryfewHABgenerathatcanbeeffectivelysampledusingtheCPR,a
ability, will be a central link between climate change and phyto- recent study has demonstrated its value in characterizing changing
planktondistributionandcomposition(Haysetal.,2005). temporal and spatial patterns of Pseudo-nitzschia species in the North
Ultimately,ourabilitytopredicttheimpactsofclimatechangeon Pacific (Stern et al., 2018). Decades of data from the CPR have also
HABsis limitedby ourskillinreplicating futureconditionsinthe la- demonstrated the linkage between rising SST and milder winter tem-
boratory or extrapolating observed responses of HABs in natural sys- peratures with a shift from diatoms to dinoflagellates in the North
temstopastclimateconditions,aswellasourinabilitytopredicthow Atlantic, the North Sea (Edwards et al., 2001; Hickel, 1998) and the
strainandspeciesplasticityandevolutionwillaltercurrenttraits.The Baltic Sea (Wasmund et al., 1998). As many HAB species are dino-
majority of studies have been conducted using a single species to de- flagellates, these data may suggest that earlier and more frequent di-
termine the response to a single variable. Controlled culture experi- noflagellateHABsmayresultfromboththedirecteffectsofincreasing
mentsinthelaboratory,intheabsenceofcompetition,grazing,anda temperatureandindirecteffectssuchaswarming-inducedstratification
multitude of other factors that are present in nature, will inevitably (EdwardsandRichardson,2004)thatwillfavorswimmersthatareable
yield either different resultscompared within situpopulations, orsi- toaccesslightnearthesurfaceandmacronutrientsatdepth.
milarresultsbutatvaryingmagnitudes.Multi-factorialexperimentsare Dinoflagellate cyst records in sediment cores can also be used to
criticaltogainamorecompletepictureofthecomplexenvironmental reconstructtimeseriesandevaluaterelationshipswithclimatedrivers.
regulatorsofHABcellsuccessandtoxinproduction;however,theselab Manydinoflagellatesproduceanon-motile,restingcyststagethatcan
experimentsarechallengingduetothelargematricesthatareneeded be buried and preserved in bottom sediments and entrained into the
(Wellsetal.,2015; GriffithandGobler,thisissue,2019).Becauseof sedimentary record (Dale, 2001). Enumerating cysts and dating sedi-
thesechallenges,multifactorialexperimentsaretheexceptionandnot mentlayersenablesreconstructionofhistoricalpatternsinabundance
thenormandfewexisttoinformmodelsusedtoprojectHABresponses thatmaybeindicativeofbloomsofthemotileforms.Althoughthecyst
tofutureclimatechange.Asaconsequence,growthresponsesofHAB record is taxonomically very limited in its coverage of HAB species,
2
V.L.Trainer,etal. Harmful Algae xxx (xxxx) xxx–xxx
analysis of cysts has been used to recreate bloom patterns of dino- including both floods and drought resulting from anthropogenic
flagellatesinthegenusAlexandriuminPugetSound,WashingtonState warming(Swainetal.,2018).Theseinclude“whiplash”eventswhere
(Coxetal.,2008;Feifeletal.,2012).Warmerairtemperaturesandsea the transitions from wet to dry and dry to wet periods become more
surfacetemperatures(SST)weresignificantlyandpositivelycorrelated rapid.Suchrapidswingsfromwettodryyearswillfavororganismsthat
withcystabundancesfromthelate1800sto2005(Feifeletal.,2012). demonstrate the ability to adapt to environmental temperature ex-
AncientDNAisolationfromsedimentcores,albeitaresearchfieldstill tremesasdemonstratedbythesuccessofPseudo-nitzschiaalongmuchof
initsinfancy,offersconsiderablepromiseformorecompleteanalysisof theU.S.westcoastduringtheanomalouslywarmconditionsof2015.
historicalcommunitystructure(e.g.,Klouchetal.,2016). Precipitationalsowillvarymarkedlybyyearandseason,especiallyin
EvenforlongHABtimeseriesthatspanmultiplecyclesofnatural southernCalifornia.Thecontrastbetweenlong,drysummersandbrief,
patternsofclimatevariability,theHABresponsetothewarmperiodsof wetwinters isexpectedtobecomeeven moreexacerbated inthe21st
thepastmaynotberepresentativeoffutureconditions.Thisisbecause century(Swainetal.,2018).
thewarmperiodsofthepastareprojectedtobecoolerthaneventhe Studying HABresponsestoextreme eventsprovidesessentialdata
coolperiodsofthefuture(OverlandandWang,2007).Thatis,climate onthesuccessofHABspeciesrelativetonon-HABspeciesundereco-
change is expected to result in future ocean conditions that will be logicallystressfulconditions.Aswithalllivingorganisms,theresponses
unlike anything experienced in the documented past, limiting our ofphytoplanktontoenvironmentalstressorsarenotnecessarilygradual
abilitytolooktothepasttopredictthefuture.Thus,eventsthattoday or linear. Many organisms have threshold responses and can be very
are considered “extreme” are likely at the lower threshold of what sensitivetoshortperiodsofenvironmentalextremesthatdeviatefrom
might be expected in the future, but nevertheless serve as important “normal”conditions(Glynn,1984; Stocket al.,2011). Separatingthe
naturalexperimentstoforecastwhatistocome. driversofHABeventsduringextremeeventsischallengingbutcanbe
furtherresolvedusingculturestudieswithnaturalisolatesfromthese
1.1. Extremeevents events.Usingisolatesfromthesenaturalexperimentsallowsustostudy
organisms’ responses to shifting and often interactive environmental
In many cases, extreme weather events mimic future climate con- pressures under controlled conditions. Important questions to ask of
ditions and provide a “dress rehearsal” for understanding future fre- thesenaturalexperimentsare:Whatarethedriverswithineachofthese
quency, intensity and geographic extent of HABs. Extreme weather examplesthatallowedtheeventtooccur?Whatevidencedowehave
eventsarenaturallyoccurring,short-term,intenseperturbationstothe thateachoftheseeventsareduetoclimatechange?Whatisthesocio-
ocean-climate system. The short-term weather changes often may be economicimpactofeachorhowdowebetterquantifytheseimpacts?
similartochangesexpectedunderclimatechange,providinganatural Andfinally,howdotheseeventsinformourunderstandingofHABsin
laboratorytostudytheeffectsofmultiplestressorsonHABs.Extreme thefutureocean?
weather will become more common with climate change. These ex- HerewediscussthreecasestudiesofextremeHABeventsthatmay
tremes willhavedramatic effectsneartheair-seainterfacewherepe- beindicativeoffutureclimatechangeconditions.Thefirstisastudyof
lagicphytoplanktonreside.Heavyandextremeraineventshavebeen a diatom species of the genus Pseudo-nitzschia while the second and
associatedwithhighSST(EASAC,2013).Thisresultsfromtheelevated thirdexaminepelagicflagellateblooms.TheseHABeventscanbeca-
capacityoftheatmospheretoholdwatervaporasairtemperaturein- tegorizedas“extreme”becauseextensivetimeseriesallowsthemtobe
creases.Thiselevatedatmosphericmoisturewillfuelstormsonseveral distinguishedfrom“normal”blooms.Inallthreeeventsdescribedhere,
spatial scales and will result in a greater likelihood of precipitation. once the extreme HAB was differentiated from normal blooms, addi-
Extra precipitation will cause a positive feedback loop, resulting in tionaltargetedmonitoringwasemployedtoimprovetheunderstanding
moreintenseprecipitationevents.Ontheotherextreme,warmingwill oftheclimatedrivers,themechanismsthatgaverisetotheeventandto
enhance evaporation resulting inearlier snow melt and causing great documentthesocietalresponse.Thesefocused,opportunisticstudiesof
summer drying. More frequent and longer heat waves with corre- nature’s experiments allow for the identification of complex climate
spondinglyfewercolddaysandnightswillbeobserved.Theresultwill driversofHABs,andtoquantifythecapacityofcoastalcommunitiesto
begreaterextremesinprecipitationanddroughtthatwillcontinueto withstandthem.Thethreecasestudiesdescribedherearenotmeantto
increasewithglobalwarming(Trenberth,2011). beanexhaustivedescriptionofallthatisknownaboutclimatechange
The opportunity to study the response of HABs under extreme andpelagic HABs,butrather theyhighlightthe valueofstudyingex-
conditions provides a greater understanding of the multifactorial en- treme events in clarifying the multitude of factors that must be con-
vironmentaleffectsthatwillcombinetoproducethemostdevastating sideredwhenanalyzingtheimpactsofclimaticchangeonHABs.
events. According to the Intergovernmental Panel on Climate Change
(IPCC,2013),bothgradualandmoreerraticchangesinweatherwillbe 2. Casestudies
observed which, in turn, willbe a major driver ofchanging HABs. In
recentyears,Europehassufferedarisingnumberofextremeweather 2.1. Pseudo-nitzschiaanddomoicacidmonitoringintheU.S.Pacific
events - from unprecedented heat waves and droughts to record- Northwest
breaking floods,wind storms and freezes (Fig. 1). The number of hot
days has increased 3-fold and the length of heat waves has doubled Long term trends are emerging that link toxic Pseudo-nitzschia
since 1880. Insurance industry data clearly show that the number of blooms and corresponding increases in DA concentrations in shellfish
loss-related weather extremes has significantly increased on a global withwarmeroceantemperatures.Acorrelationwasobservedbetween
scale.Thereisincreasingevidencethatglobalwarmingdrivessomeof anincreasedgrowthratepotentialofP.australisandthePacificDecadal
these trends. These extreme events include damage due to heavy Oscillation(PDO)witha1-monthlaganda3-monthlagwiththemean
rainfall, drought, heat waves and fewer extremely cold days (EASAC, OceanNiñoIndex(ONI;McCabeetal.,2016).Similarly,a25-yearre-
2013). cordofrazorclamDAshowedastrongrelationshipwiththePDOwith
Modelling studies are showing that extreme weather will also be- zerolaganda3-monthlagwiththeONI(McCabeetal.,2016).Thelag
comemoreprevalentinotherpartsoftheworld.Forexample,thestate periodmaybearequisite“pre-conditioning”ofseawaternecessaryfor
of California in the U.S. will show increased hydroclimatic extremes theobservedbiologicalresponse.ThepositiverelationshipbetweenP.
3
V.L.Trainer,etal. Harmful Algae xxx (xxxx) xxx–xxx
Fig.1.Trendsinnumbersofextremeeventsworldwide,definedasthosenon-linearclimateanomaliescausingaboveaveragesocio-economiclosses.(A)Trendsin
typesofdifferentcatastrophesworldwide,1980–2002.Dottedlinesconnectannualaveragedata;solidlinesshowlineartrends;(B)Relativetrendsofloss-relevant
naturalextremeeventsindifferentpartsoftheworld.Forbothpanels,1980levelsaresetat100%;datafromMunichReNatCatSERVICE);reprintedwithpermission
(EASAC,2013).
australisgrowthratepotentialandwarmanomalies,includingthePDO, anomaly contributing to cascading effects throughout the ecosystem.
El Niño as well as the Pacific warm anomaly in 2015, illustrate the This warm water did not support living marine animals due to the
likelihood of more frequent DA contamination of shellfish during depleted nutrients and collapse of the food web offshore, similar to
warmer years (McCabe et al., 2016; McKibben et al., 2017). This in- nutrient depletion experienced during other warm (El Niño) events
formation linking damaging DAclosures towarmer years currentlyis (Chavez et al., 2002). Specifically, a record number ofGuadelupe fur
beingusedbycoastalmanagers,therebyenablinggreaterpreparedness seals,Northernfurseals,andCaliforniasealionpupswererescuedby
incoastalcommunities. the Marine Mammal Center in Sausalito, California in 2015. The sea
In late Spring 2015, a massive, nearly simultaneous bloom of lion pups were younger and thinner than typically seen, further evi-
Pseudo-nitzschia occurred along the west coast of North America, denceofthelackoffoodinthecoastalseas(TMMC,2015).TheMarine
spanningtheCaliforniaCurrentSystem(CCS)fromcentralCaliforniato Mammal Center stated “we cared for more than 200 patients at our
at least as far north as British Columbia. Harmful consequences re- hospital every single day for 3 solid months”; animals suffered from
sultingfromthefoodwebtransferofDAcausedclosuresofDungeness starvation,DApoisoningorboth.Thescarcepreythatwasavailablefor
crab,rockcrab,razorclamandotherfisheries,aswellasthemortality mammals, such as sardines and anchovies, were highly contaminated
of numerous marine mammals and seabirds (McCabe et al., 2016; withDA.Allplanktivorousfish,includingNorthernanchovy(Engraulis
McKibben et al., 2017). Coastal communities suffered severe losses mordax),Pacificjackmackerel(Trachurussymmetricus),Pacificsardine
fromfisheriesclosures-thecommercialDungenesscrabfisheryalone (Sardinopssagax),Pacificherring(Clupeapallasii),andeulachon(Tha-
experiencedaUSD$97.5milliondecreaseinrevenuein2015relativeto leichthyspacificus)collectedprimarilyinCaliforniacoastalwatersand
2014dueinparttotheDAclosures(NMFS,2016). analyzed for DA (n=45) contained measurable DA with the highest
A sequence of environmental changes occurred in response to a concentrationmeasuredinanchovyvisceraatanastounding3239ppm
warmwateranomaly˜3°CaboveaverageSSTtocausea“perfectstorm” (Fig. 2). There was a general trend of decreasing toxicity with in-
of shellfish closures and animal mortalities. This northeast Pacific creasing latitude, consistent with the survey sampling northern lati-
marineheatwave,commonlyknownasthe“Blob”,begantodevelopin tudeslateinthebloomwhenDAwasdeclining.
the Gulf of Alaska in late 2013 and continued to spread in 2014 and Byspring2015,thewarmanomalymovedtowardthecoastofNorth
2015(Bondetal.,2015).TheSSTanomalieswerecausedbyreduced America.AresearchcruiseoffNewport,Oregon,inApril2015showed
heatexchangebetweentheoceanandatmosphere,relativelypoorad- thattoxicP.australiscellswerepresentwithinthewarmwatermass(W.
vection of cool water in the upper ocean, and reduced storm-driven Peterson, pers. comm.). A mooring off the Oregon coast showed that
mixing (Bond et al., 2015). Nutrient renewal from vertical transport peakSSTsreachedthecoastinApril2015,followedimmediatelybythe
wasrestrictedinwinterof2014-15duetoincreasedstratification(Peña spring transition to upwelling (Du et al., 2016). A series of spring
et al., 2018). Climate model simulations combined with observations stormsinMay2015broughtthetoxiccellstothecoastwheretheywere
suggestthatmarineheatwavesintheNorthPacific,liketheBlobandits fueledbymacronutrientsfromupwelling,resultinginacoastalbloom
associated drought that was particularly notable in California, may thatcontaminatedshellfishcoastwidewithDAatconcentrationsabove
intensify with climate change (Wang et al., 2014, 2015; Yoon et al., the regulatory threshold beginning on 8 May 2015 (Fig. 3, vertical
2015;DiLorenzoandMantua,2016). dottedline).
During the 2014–16 northeast Pacific marine heatwave, low pri- How did P. australis survive in these nutrient-deprived, warm wa-
mary productivity was noted (Whitney, 2015) and anomalously low ters,andthenthrivetodevelopintoamonospecific,highlytoxicand
nutrient concentrations were observed off central Oregon (Du et al., devastatingHABunderthissequenceofchangingconditions?ClearlyP.
2016)andstation P(Peñaetal., 2018)inthenortheast subarcticPa- australis can sustain itself under extreme environmental conditions,
cific.Recordnumbersofmarinemammalswereobservedclosetoshore positioningitselftooutcompeteotherphytoplanktonandbloomtohigh
inCalifornia,withdecreasednutrientconcentrationsinthewarmwater densities when provided with adequate nutrients for exponential
4
V.L.Trainer,etal. Harmful Algae xxx (xxxx) xxx–xxx
Fig. 2.Temperature anomaly and DA in fish
viscera during a west coast cruise. (A)
AnomalouslywarmsurfacewaterintheNorth
Pacific Ocean as shown by mean sea surface
temperature anomalies (°C) from the NCEP/
NCARReanalysisforJune-August2015;(B)All
fish viscera (northern anchovy, herring, Jack
mackerel, sardine) collected from the 2015
westcoastsurvey(transitfromsouthtonorth;
June - Sept 2015) were positive for DA with
valuesrangingfrom0.02to3238.6ppm.The
regulatoryclosurelevelof20ppmisshownin
red on the right y-axis. (For interpretationof
thereferencestocolourinthisfigurelegend,
thereaderisreferredtothewebversionofthis
article).
growth.Asuiteofphysiologicalcharacteristicsmayhavepermittedthis anomalouslylowSiO4:NO3ratios,ledtosilicateexhaustioninsurface
particularspeciestorespondinsuchadramaticfashionin2015.These waters, whereas nitrate remained available throughout the bloomde-
include its ability to acquire strongly complexed iron, even when velopment(Ryanetal.,2017).GiventhatcellularDAquota(particulate
available at very low concentrations, using a high-affinity iron acqui- DApercell)dependsonthebalancebetweenratesofDAsynthesisand
sition system that requires copper and the production of DA (Wells ratesofcelldivision,silicatelimitationmayhavebeentheproximate
et al., 2015); in fact, another massive bloom of Pseudo-nitzschia was causeofelevatedDAtoxicityinMontereyBaybydecreasingcelldivi-
reportedadecadeearlierinthenorthernCCSunderlowFeconditions, sion rates without any increase in per-cell DA production rates. Here
but when macronutrients were plentiful (Trainer et al., 2009; Trick theplentifulnitratesupportedexceedinglyhighabundancesofPseudo-
etal.,2018).Inaddition,thehighaffinityofP.australisfornitrateand nitzschiawhichwereoverwhelminglyP.australis(Bowersetal.,2018).
ammonium (Cochlan et al., 2008) would provide this species with a Domoicacidquotasreachedhighlevelswithinthesedensepopulations
competitive advantage for acquiring nitrogen (N) under N-depleted (Ryan et al., 2017) due to the eventual exhaustion of silicate, while
conditions,butespeciallyfollowingupwellingduringN-repletecondi- nitrate remained available for continued DA biosynthesis rates as re-
tionswhereitsmaximalrateofnitrateuptakeexceedsthoseofvirtually portedinculturestudiesfollowingsilicate(orphosphate)depletionfor
all the other phytoplankton species commonly found in upwelling otherPseudo-nitzschiaspecies(cf.reviewsbyBates,1998;Lelongetal.,
systems (Kudela et al., 2010). It is also possible that P. australis may 2012; Trainer et al., 2012). In contrast, further north in the CCS off
share the short-term ‘surge uptake’ capability seen for smaller-celled Oregon, the cool nutrient-rich upwelled waters did not initially elim-
species of Pseudo-nitzschia (Auro, 2007; Bill, 2011), where N-starved inatetheanomalouslywarmwatersatthecoastalmargin.Theupwelled
cells can rapidly respond to elevated N concentrations by acquiring watermixedwiththeoverlyingandadjacentwarmwatersoftheBlob
nitrogen at uptake velocities that greatly exceed cellular growth re- and P. australis bloomed not only because of the supply of upwelled
quirements, but this has not yet been examined for P. australis. Al- nutrients, but also due to the direct effect of temperature on cellular
though the details are not entirely clear, the ability of P. australis to growth,seeninthethermalgrowthresponsecurvesforthreestrainsof
survive marine heat waves, and the resultant conditions of warmer, P.australisisolatedfromthisbloom(McCabeetal.,2016).Here,unlike
nutrientdepletedwaterssuggestthatthistoxicspecieswillbeamore MontereyBay,ambientsilicateconcentrationsrarelywerelowerthan
frequentoccupantofthefutureocean. nitrate (Du et al., 2016), and warmer temperature was likely the pri-
While it is clear that the spring upwelling provided the needed maryphysiologicalfactorinthedevelopmentofthetoxicdiatombloom
macronutrients to support the growth P. australis to high densities inthenorthernCCS.Evenfurthersouth(e.g.,theSouthernCalifornia
throughouttheCCS,theseupwelledwatershadsubtle,butpotentially Bight), there were no toxic blooms in 2015, presumably due to the
verydifferentimpactsonthetoxicityofthe2015bloom,inparticular supraoptimal temperatures (Smith et al., 2018). Here, substantial DA
during its initial development. In Monterey Bay (southern CCS), the concentrations have never been seen at water temperatures>19°C
upwelledwatersinitiallyeliminatedanomalouslywarmsurfacewaters (Smithetal.,2018).
associatedwiththeBlob.Thus,althoughwerenodirectimpactsonP. ExaminingtheresponseofPseudo-nitzschiaacrossthefulldomainof
australis growth rate from increased water temperature, the unusual the CCS also reinforces the difficulty of broadly generalizing from
chemical composition of these upwelled waters, specifically the limited studies. P. australis generally does well inwarm (butnot hot)
5
V.L.Trainer,etal. Harmful Algae xxx (xxxx) xxx–xxx
Fig.3.ShellfishharvestingbeachesandDAconcentrationsinrazorclams.(A)RazorclamDAatbeachsitesinOregon(OR),Washington(WA)andBritishColumbia
(VancouverIsland,VI)withnorth-southwindstress.DottedlineshowsthatonMay8,shellfishbeachesshowedasimultaneousincreaseinDAconcentrations.Winds
abovethezerolineindicatestormsornorthwardwindstress;the3stormsinlateAprilandearlyMaytransportedPseudo-nitzschiacellstowardthecoast;(B)Mapof
beacheswhererazorclamsormussels(SpringCoveonly)weremonitoredforDA(pinkcircles);datashowninpanelA.WinddatawereaccessedfromtheNational
DataBuoyCentermooringatCapeElizabeth,WA(46041;bluecircle).(Forinterpretationofthereferencestocolourinthisfigurelegend,thereaderisreferredtothe
webversionofthisarticle).
water with weak to moderate upwelling (Kudela et al., 2004). Re- evidence for a thermal limit of>19°C for production and accumula-
structuringoftheEastPacificprovideda“windowofopportunity”for tion of DA from extant natural populations (Smith et al., 2018 and
Pseudo-nitzschia dominance in place of the usual springtime centric Fig.5),optimalgrowthrateswerereportedat23°CforonestrainofP.
diatomsthatincludeChaetocerosspp.,Skeletonemacostatumandothers australis isolated from southern California (Zhu et al., 2017). Here
(e.g.,Venrick,1998).Conditionshavebeenshiftingforthelastseveral cellularDAconcentrationswerenotevendetecteduntilthistempera-
yearstowardswarmer,nutrient-depletedwater,amplifyingtheunusual ture was reached, and then DA increased exponentially with rising
conditions in 2015 (Fig. 4). However, even though there is strong temperature above 23–30°C, even as growth rates were crashing,
Fig. 4.Data collected from the Santa Cruz Municipal Wharf (Monterey Bay, CA) weekly from 2012 to 2016 show significant trends of increasing sea surface
temperatureanddecreasingnitrateovermultipleyears.
6
V.L.Trainer,etal. Harmful Algae xxx (xxxx) xxx–xxx
Fig.5.InsouthernCalifornia,itwastoohottosupportPseudo-nitzschiagrowthandDAproduction.(A)Thetemperature-salinityplotshowsparticulateDA(pDA)
concentrations from a 2015 west-coast survey (McCabe et al., 2016). The highest DA (and presumably P. australis) was associated with cool(er), salty water,
indicativeofBlobwatermixedwithlocalupwelling.ThebluedashedpolygonshowstheT-Sspace(0–10m)forCalCOFILine80(offSantaBarbara,inset,blue
transect)immediatelyaftertheSpringTransition,whiletheredpolygonshowstheT-SspaceforLine90(offsouthernCalifornia,inset,redtransect).Astheseason
progressedthoseregionswarmed,reducingtherealizedhabitatfortoxicP.australisinthesouth,whileexpandingitnorthwardintothePacificNorthwest;(B)The
surveydatafortotalDAversustemperature.ConsistentwithSmithetal.(2018),nosubstantialDAwasdetectedabove18°C.(Forinterpretationofthereferencesto
colourinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle).
resulting in cellular DA quotas increasing by 70-fold in this upper suchasnutrientavailability(Ryanetal.,2017),andthereappearstobe
temperaturerange. considerablespeciesandstrainvariabilitythatcouldfavorbothmoreor
Pseudo-nitzschia species matter as do their specific responses to lesstoxicorganismsunderfutureclimateconditions.
changingenvironmentalconditions.Duringthe2015event,P.australis Thesocioeconomicandculturaldamagetocommunitiesasaresult
emerged from a background of multiple Pseudo-nitzschia species, and ofthe2015HABeventweredescribedinarecentstudyoftwoU.S.west
dominated along almost the entire U.S. west coast, whereas during coast fishing communities that are both highly dependent on the
other years, less toxic species such as P. delicatissima can dominate, commercialDungenesscrabfisheryforincomeandhaveasubstantial
resulting in large blooms but with lower toxicity (Smith et al., 2018; recreationalrazorclamfishery(Ritzmanetal.,2018).Inperson,semi-
Bowersetal.,2018).Underfutureconditions,wemayseeselectionfor structuredinterviewsassessedthesensitivityandadaptivecapacityof
speciesorstrainsthathavenotemergedinthehistoricalrecord,andare thesecommunitiestotheHABevent.Documentedeconomichardship
therefore poorly characterized. This could be exacerbated when con- spanned the communities, and included processors and fish markets,
sidering multiple stressors such as warming, increased hypoxia, and restaurants,hotels,retailandtourismsectors.Someintervieweesfeared
ocean acidification - stressors that are all expected to intensify in the thatarepeateventwouldresultinaculturalshiftawayfrombothre-
CCS as a result of anthropogenic change (cf. reviews by Capone and creational and commercial fishing. Understanding the social, cultural
Hutchins,2013;Ravenetal.,2019thisissue).Recently,theinteractive and economic hardships faced by coastal communities is an essential
effectsofincreasedtemperatureandpCO2wereexaminedwithN-lim- partofthescientificstorytellingofHABevents.Communitymembers
ited continuous cultures of natural assemblages from southern Cali- emphasizedtheimportanceofpreparingforthefuturewhichincludes
forniacontainingP. multiseriesandP.hasleana.Althoughtemperature fully describing and communicating impacts to legislative authorities
individually appeared to have the strongest influence on cellular DA whocan sponsormitigative strategiesto reducethe shockof extreme
quotas,therewereconflictingpDAresponsestopCO2dependingonthe HABs.
Nsubstrateusedforgrowthandthetemperaturetested,andtheauthors
concluded that there was noconsistent toxicresponse topCO2 nor to 2.2. AlexandriumdinoflagellateandtoxinmonitoringinTasmania,
theinteractionoftemperatureandpCO2(Tattersetal.,2018) southeastAustralia
Itisclearfromthiscasestudythatwarmingwasasignificantfactor
in the unprecedented west coast-wide Pseudo-nitzschia bloom, but the
Startingin1985,theTasmanianshellfishindustryhadbecomeused
bloomwasnotnecessarilytheresultofalinearphysiologicalresponse
to closures and public warnings of paralytic shellfish poisoning (PSP)
to increased temperature. The northeast Pacific marine heatwave al-
risksinflicted byGymnodiniumcatenatumdinoflagellatebloomsinthe
lowed P. australis to expand or shift northward but exceeded the ap-
Huon River and d’Entrecasteaux Channel near the capital city of
parenttemperaturethresholdforDAproductioninthesouthernregions
Hobart,Tasmania,Australia(Hallegraeffetal.,1995).Afterexperien-
of the CCS (Fig. 5) where toxic blooms did not develop in 2015, and
cing recurrent shellfish farm closures in this region for a decade, this
where substantial DA concentrations have never been observed at
problemwaslargelysolvedbydeclaringtheaffectedestuariesunfitfor
temperatures>19°C (Smith et al., 2018). The temperature signal is
shellfishaquaculture.EarlyHABsurveysofotherTasmanianlocations,
almost certainly a proxy for other co-varying environmental factors
including the east coast, had revealed low concentrations of
7
V.L.Trainer,etal. Harmful Algae xxx (xxxx) xxx–xxx
Fig.6.SignageandSSTduringtheanomalousPSTeventineasternTasmania.(A)Thefirsttoxicalgalbloomwarningsignspostedalongtheentireeastcoastof
Tasmaniatopreventfurtherhumanpoisoningsandprotecttourism.Courtesy:https://www.dhhs.tas.gov.au/__data/assets/pdf_file/0005/264227/Marine_Biotoxin_
Warning_6pp_DL.PDF;(B)MapofTasmania,southofthemainlandofAustralia,showingseasurfacetemperatureson27September2015duringpeakPST,withthe
EastAustralianCurrent(inred)interactingwiththecontinentalshelf.ThelocationsofthemainaffectedshellfishfarmareasMoultingBay,GreatOysterBay,Little
SwanportandSpringBayareindicated.Source:oceancurrent.imos.org.au.(Forinterpretationofthereferencestocolourinthisfigurelegend,thereaderisreferredto
thewebversionofthisarticle).
Alexandrium tamarense group 5 (non-toxic; now designated A. aus- by analytical laboratories. Starting in 2018, Tasmania invested in a
traliense)andonlyoncegroup4(weaklytoxic;nowdesignatedA.pa- dedicatedbiotoxinlaboratory.
cificum) and the area was therefore classified as a low biotoxin risk, Preliminary investigations of meteorological and oceanographic
subject to limitedplankton and biotoxin monitoring. Unexpectedly in conditions from 2012-17 indicate that the annually recurrent winter-
October 2012, a shipment of cultured blue mussels was tested by Ja- spring Alexandrium catenella blooms (June-October) occur within a
paneseimportauthoritiesandshowntobecontaminatedwith10mg/ narrow temperature window (10–15°C) under two distinct sets of
kg paralytic shellfish toxin (PST; 0.8mg/kg is the regulatory closure conditions: (1) following high rainfall and land run-off; and (2) fol-
level).Thisledtoaglobalproductrecallandlosstothelocaleconomy lowing periods of anomalously low air temperatures and associated
ofAUD$23M.Thecausativeorganismprovedtobehighlytoxicblooms cooling of shallow coastal waters. The common driver of blooms ap-
ofA. tamarensegroup1(nowdesignatedasA. catenella)notdetected pearstobestratificationincoastalwatersviasalinityandtemperature
previouslyinanyAustralianorNewZealandwaters(Hallegraeffetal., gradients. The ocean environment east of Tasmania is a well-docu-
2017). Following low toxicity during 2013 and 2014, more severe mented climate change hotspot (Fig. 6), characterized by a stronger
bloom events occurred during 2015, 2016 and 2017 (up to 300,000 East Australian Current (EAC) and rapidly increasing ocean tempera-
AlexandriumcellsL−1;up150mgkg−1PSTinmussels,22mgkg−1PST tures (2.3°C warmer since the 1940s) and associated changes in fish,
inoysters;14mgkg−1PSTinrocklobsterviscera;1.3mgkg−1PSTin plankton and kelp communities (Johnson et al., 2011). Thermal stra-
abalone viscera), also causing 4 people to be hospitalized after con- tificationusuallyoccursonthecontinentalshelfoverlatespring(No-
sumption of wild shellfish (human poisoning syndromes reported by vember)butsince2009thestrengtheningEAChasbeenaccompanied
Edwardsetal.,2018).Bothcultureexperimentsbutnotablyfieldesti- by earlier developmentofstratified conditions.While there isno evi-
matessuggestedahighcellulartoxincontentupto100–500pgSTXeq dence that blooms originate offshore, the offshore environment may
cell-1, explaining why even very low Alexandrium concentrations play a role in moderating coastal water conditions. At times, Alexan-
(50–100cellsL-1)occasionallycancauseshellfishtoxicity. driumbloomsoffeasternTasmaniamayalsobeinfluencedbyclimate
To protect tourism and human health, the east coast of Tasmania drivenchangesassociatedwithseasonalrainfall.Bloomeventsin2012,
haspostedpermanentpublicHABwarningssince2017(Fig.6),which 2013 and 2016 followed high rainfall, resulting in enhanced coastal
is a first for Australia. Until these bloom events began in 2012, the stratification.Coastalstratificationalsointensifiedduringthe2015and
island state of Tasmania had very limited biotoxin analytical cap- 2017bloomperiods,butthroughadifferentmechanisminvolvingthe
abilities,withsamplesroutinelyshippedandtestedinSydney,thereby subsurfaceoutflowofanomalouscoldwaterfromGreatOysterBayand
leaving theTasmanianseafoodindustry inlimbofor7–10 dawaiting SpringBay(Condieetal.,2019).Decreasedsilicateconcentrationsas-
testresults.Toassisttheindustry,amajoreffortwasmadetointroduce sociatedwithastrengtheningEACwouldfurtherfavordinoflagellates
and validate rapid (20min) immunological test kits, allowing for andselectagainstcompetingdiatomblooms(Thompsonetal.,2009).
shellfish farmers to perform PST tests at their farms before making ClimateforecastsfortheTasmanianregionpredictgraduallydecreasing
decisions on harvesting (Dorantes-Aranda et al., 2017). Extensive total rainfall, but an increase in the frequency of heavy downpours
training courses were organised to ultimately provide individual (Grose et al., 2010). With ever increasing water temperatures also in
farmerswithcertificatesofcompetence.Negativetestresultswereac- wintermonthsprojectedforthedecadestocome,thiscouldreducethe
ceptedbyregulators,whilepositivetestresultsrequiredconfirmation seasonalbloomwindowforthecolder-waterA.catenella.
8
V.L.Trainer,etal. Harmful Algae xxx (xxxx) xxx–xxx
Fig.7.Shellfishtoxicity(mgSTXeq.kg−1)from2012to2017
in Moulting Bay and Little Swanport oysters and Spring Bay
mussels. Orange arrows indicate the seasonal 10–15°C tem-
peraturewindow.The2016bloomwasprecededbyamajor
rainfall event while anomalously cold water in Great Oyster
Bay (Little Swanport) may explain the 2015 bloom. (For in-
terpretation ofthereferencestocolourinthisfigurelegend,
thereaderisreferredtothewebversionofthisarticle).
Our current understanding of these novel Alexandrium blooms is ofalgalbloomsonthesalmonindustrydatebackto1983(Lembeyeand
thattheyarenotasimpleresponsetoincreasingwatertemperatures, Campodónico, 1984). In 1988, a bloom (100,000 cells ml−1) of the
sincetheyoccurduringthecoldwinter-springmonths(Fig.7).Genetic raphidophyteHeterosigmaakashiwokilled5000tonsofsalmonvalued
evidence(Johnetal.,2018)indicatesthattheTasmanianA.catenella at USD$11M (Clement and Lembeye, 1993). This extreme biological
group 1 populations exhibit a unique microsatellite signature and eventin1988triggeredthestartofanextensiveplanktonmonitoring
therefore do not represent a ballast water introduction nor range ex- program by the Chilean Salmon Association which was followed by
pansion (Sydney coastal waters exclusively host A. pacificum). Pre- other private and governmental organizations (i.e., Plancton Andino
liminarysedimentDNAresults(Shawetal.,2019)suggestthatA.ca- SpAandtheInstituteforFisheriesDevelopment–IFOP).Insummerof
tenelladidexistinTasmanianwatersasearlyas1987butsimplyhad 2000,thesamespeciesbloomedagainandcausedthemortalityof40%
been overlooked. Alexandrium cyst surveys during August2016 along ofthefarmedsalmonintheReñihuefjord.Inthatevent,asfewas40
theentireeastcoastofTasmaniafoundconsistentlylowabundancesof cells ml-1 of H. akashiwo coupled with a high concentration of Lepto-
cysts(0.1–3cystspergramofsedimentwetweight;Hallegraeffetal., cylindrusdanicusproducedaverylargesalmonkillatonlyonesalmon
2017).Thisalsosuggeststhatthesebloomsarerecentevents.Palaeo- farm,pointingtothehighpotentialichthyotoxicityofthisflagellateina
genomic researchisinprogressusing ancientDNA methodsondated synergistic action with the diatom. However, the most damaging and
sediment depth cores (11,000 years old) from the area to document extremeeventinthehistoryofthesalmonindustryinvolvedtwoother
historic shifts in A. tamarense ribotypes 1, 4 and 5, in support of this ichthyotoxic microalgae; the dictyochophyte, Pseudochattonella, (re-
recentbloomingphenomenon.Nohistoricrecordsofshellfishpoisoning ferred to until 2011 as the raphidophyte Chattonella) and the dino-
areknown fromtheTasmanianregionbyAboriginaltribesforwhom flagellate,Alexandriumcatenella.
shellfish comprised a significant part of their diet (Jones, 1978). Our DuringthewarmJanuaryof2004,Pseudochattonellacf.verruculosa
workinghypothesisisthatthesedinoflagellatesrepresentapreviously (8cells/ml)wasrecordedforthefirsttimetogetherwithanoutbreakof
cryptic genotype newly stimulated by climate-driven increased water Vibrio parahaemolyticus in Los Lagos region, causing the mortality of
columnstratificationduringwinterandearlyspring.Themajorlessons 2–5% of the total farmed salmon in the Reloncavi Sound (Mardones
learnedfromthisextremeTasmanianHABeventarethatpredictionof etal.,2012).Thisspecieswasregularlydetectedin2004,2009,2011
the impact of global climate change on marine HABs is fraught with and 2016 in the same area. During the first recorded bloom, P. cf.
difficulties, and that the greatest problems for human society will be verruculosacelldensityreachedamaximumof22cellsml−1atawater
causedbyrangeexpansionsortheincreaseofalgalbiotoxinproblems temperature of 15°C, whereas in later events peak abundances
inpoorlymonitoredareas. reached>400cellsml−1atwatertemperaturesof18.5°C(Mardones
et al., 2012). Even higher concentrations recently were detected in
2016inthenorthernPatagonianfjords,consideredthemostsignificant
2.3. RecordHABeventsinChileanfjordsin2016byPseudo-chattonellacf. event to date in that area. Altogether, these spatial-temporal records
verruculosaandAlexandriumcatenella show an increase in bloom intensity and a clear expansion of Pseu-
dochattonellacf.verruculosatowardsthesouthernareas(Fig.8B).
TheChileaninlandmarinewaters(Fig.8A)areauniqueSouthern Incontrast,Alexandriumcatenellabloomshaveshownanexpansion
Hemisphere ecosystem that supports a variety of economic activities, towardsthenorthofChileinrecentyears(Fig.8C).Thisdinoflagellate
especially fishing and aquaculture, providing employment and eco- wasfirstreportedinChilein1972intheBellBay-MagallanesRegion
nomic servicesforthousandsofpeople.To date,mostoftheshellfish (∼54°S)(GuzmanandCampódonico,1975).Subsequently,A.catenella
andsalmonfarmingactivityiswithintheLosLagosregion(41–44°S) wasdetectedoutsidethemouthoftheAysénfjord,nearChurrecueIs-
with an increasing development further south. Salmon farming com- landinMay1992(Muñozetal.,1992).Itisbelievedthatthiswasthe
menced in the Chilean fjords in 1982 and currently is valued at USD first sign of a gradual northward dispersion of the species. Later in
$4.3billionperyear(León-Muñozetal.,2018).Overthepastdecades 1998,cellsofthistoxicdinoflagellatewereobservedinnettowsamples
thesalmonindustryhasfacedanumberofsetbacksfromdiseaseout- collected near the Quellón area, Chiloé Archipelago (Lembeye et al.,
breaksandfish-killingalgalblooms(Littleetal.,2015).Localimpacts
9
V.L.Trainer,etal. Harmful Algae xxx (xxxx) xxx–xxx
Fig.8.LocationofHABspeciesandtheirrelationshiptoclimateindicesduringthe2016GodzillaHABeventincomprisontopreviousyears.(A)Patagonianfjordsin
theSouthofChile;(B)SouthernexpansionofPseudochattonellasp.(Mardonesetal.,2012;Clementetal.,2016);and(C)northernexpansionofAlexandriumcatenella
(MardonesandClement,2016)bloomsinChileanfjords,and(D)bloomsthathaveexpandedthespatialdistributionofPseudochattonellasp.andA.catenellainthe
ChileanfjordsassociatedwiththeElNiño3.4index(blueline)andtheMarshallSouthernAnnularMode(SAM)index(orangeline).TheintensityoftheA.catenella
andP.verruculosabloomsisindicatedbytheintensityoftheredandgreenverticalbars,respectively.(Forinterpretationofthereferencestocolourinthisfigure
legend,thereaderisreferredtothewebversionofthisarticle).
1998)atleast350kmnorthfromAysénfjord.However,themostim- 1184 cells ml−1, forming patches of reddish water discoloration.
portantoutbreakwithamaximumof778cellsml−1wasinsummerof Duringthesummerof2005and2006,bloomsofAlexandriumcatenella
2002inthesouthernChiloéArchipelago(43°S)(Clementetal.,2002). (by then a well-known PST producer) were detected in the northern
ThefirstfarmedtroutkillduetoA.catenellaoccurredinChurrecue fjords of the Aysén region. The bloom caused 7 human intoxications
Island,nearAysenfjord,inMarch1996withamaximumabundanceof andonefatality.Unusualbehavioroffarmedsalmon,coincidentwith
10
