ebook img

The novel interaction between Phytophthora ramorum and wildfire elicits elevated ambrosia beetle PDF

13 Pages·2014·1.35 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview The novel interaction between Phytophthora ramorum and wildfire elicits elevated ambrosia beetle

ForestEcologyandManagement318(2014)21–33 ContentslistsavailableatScienceDirect Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco The novel interaction between Phytophthora ramorum and wildfire elicits elevated ambrosia beetle landing rates on tanoak, Notholithocarpus densiflorus Maia M. Beha,⇑, Margaret R. Metza, Steven J. Seyboldb, David M. Rizzoa aDepartmentofPlantPathology,UniversityofCalifornia,Davis,OneShieldsAve,Davis,CA95616,UnitedStates bUSDAForestService,PacificSouthwestResearchStation,Davis,CA95616,UnitedStates a r t i c l e i n f o a b s t r a c t Articlehistory: The2008wildfiresintheBigSurregionofCalifornia’scentralcoast—thefirsttooccurinforestsimpacted Received25August2013 byPhytophthoraramorum,thenon-native,invasivepathogenthatcausessuddenoakdeath—providedthe Receivedinrevisedform6January2014 rareopportunitytostudytheresponseofscolytidandothersubcorticalbeetlestothisnoveldisturbance Accepted7January2014 interaction.Weusedstickycardtrapsattachedtothemainstemoftanoak,Notholithocarpusdensiflorus, thetreespeciesmostsusceptibletoP.ramorum,todeterminewhichsubcorticalbeetlespeciesmaybe usingtanoakasahostandtocompareinsectlandingratesonthesetreesinforestplotsimpactedbynei- Keywords: therdisturbance,eitherwildfireorP.ramorumdisturbancealone,orbothdisturbancescombined.Xyle- Anthaxia borinus saxesenii and Gnathotrichus pilosus, two species of ambrosia beetles (Coleoptera: Scolytidae), Flatheadedborers composedthemajority(48%and40%,respectively)ofsubcorticalbeetleslandingontanoaksduringboth Gnathotrichuspilosus Metallicwoodboringbeetles yearsofthestudy.Adultsoftwospeciesofasmall,branch-feedingflatheadedborer(Anthaxiasp.;Cole- Suddenoakdeath optera:Buprestidae)werealsocapturedinrelativeabundancelandingontanoaksinthecombineddis- Xyleborinussaxesenii turbanceplotsduringthesecondyearofthestudy.Allbuttwoofthe2779scolytidbeetlescollectedin thisstudyweretrappedontanoaksinforestplotsdisturbedbyP.ramorumand/orfire,and75%ofthese scolytidsweretrappedduringthefall2009season.Themajorityofscolytidsweretrappedontanoaksin plotscontainingbothdisturbances(81%in2009and79%in2010),and,ofthetwodisturbances,more scolytids were trapped on tanoaks in burned plots than in P. ramorum-infested plots (92% more in 2009and476%morein2010).Semiochemicalsemanatingfromthetanoaksuponwhichthestickycards wereattached—eitherintheformofhostvolatilecompoundsorscolytidaggregationpheromones—pre- sumablyaffectedambrosiabeetlelandingrates,andgreaterquantitiesofmoribundandrecently-killed treesintheplotsdisturbedbothbyP.ramorumandfiremayhaveledtogreaterpopulationdensities ofambrosiabeetlesintheseareas.OurfindingsofelevatedambrosiabeetlelandingratesinBigSurfor- ests with mixed disturbances suggest a heightened threat to tanoak in these areas, but additional researchisneededtodeterminetheactualfrequencyofambrosiabeetlegalleryinitiationinlivingtano- aksandwhethercolonizationhastensorleadstotreemortality. (cid:2)2014ElsevierB.V.Allrightsreserved. 1.Introduction ofaparticularecosystem(Paineetal.,1998),butinteractionsthat involveanoveldisturbanceandapre-existingdisturbanceregime Bothbioticandabioticdisturbancesshapeforestecosystemsby areofparticularsignificanceduetotheirincreasingfrequencyand influencingtheircomposition,structure,andfunctionalprocesses potential for dramatic and sustained changes to landscape struc- (Pickett and White, 1985). Yet even as disturbances proliferate tureandfunction(BumaandWessman,2011). duetoclimateandhabitatchange,biologicalinvasions,andother The feedbacks between invasive forest pests and wildfires are processes, little is known about the interactions among multiple examples of interacting disturbances that are of great relevance disturbances(Turner,2010). Disturbanceinteractions ofanytype intheUnitedStates(USA).From1990to2006,therewasanearly are important because they may exceed the ecological resilience threefoldincreaseintherateofdetectionofestablished‘highim- pact’forestpathogensandinsects(thosespeciesofregulatorysig- ⇑ nificance or that have caused notable damage to forest trees) Correspondingauthor.Tel.:+1(530)7549894. E-mailaddresses:[email protected](M.M.Beh),[email protected](M.R. compared to the previous 130yr (Aukema et al., 2010). Metz),[email protected](S.J.Seybold),[email protected](D.M.Rizzo). http://dx.doi.org/10.1016/j.foreco.2014.01.007 0378-1127/(cid:2)2014ElsevierB.V.Allrightsreserved. 22 M.M.Behetal./ForestEcologyandManagement318(2014)21–33 Additionally, the frequency of large wildfires in the western USA the tree bole caused by the beetle galleries (McPherson et al., hasincreasedsignificantlysincethemid-1980sinassociationwith 2008,2010,2013). warming temperatures and lengthened fire seasons (Westerling DespitethesusceptibilityoftanoaktoP.ramorum,theimpacts etal.,2006). of scolytid colonization of P. ramorum-infected tanoak have re- Asaspecificexampleoftwointeractingdisturbancesinwestern ceivedlittleattentioncomparedtothoseoncoastliveoak.Tanoak NorthAmerica,nativebark beetles[Coleoptera: Scolytidae(sensu has been recorded as a host for western oak bark beetles, Pseud- Wood,2007;Bright,2014)]frequentlyattackfire-damagedconif- opityophthorus pubipennis (LeConte) (Bright and Stark, 1973; Fur- erous trees. They are considered the most significant insects fol- nissandCarolin,1977;SwieckiandBernhardt,2006)andseveral lowing fires in western North America because they can kill ambrosia beetles, Monarthrum scutellare (LeConte) (Wood and weakenedtreesthatmayhaveotherwiserecoveredfromtheburn Bright,1992;SwieckiandBernhardt,2006)andXyleborinussaxese- injury (reviewed in McCullough et al., 1998; Parker et al., 2006; nii(Ratzeburg)(BrightandStark,1973).Noadditionalhostrecords Jenkins et al., 2008). Ambrosia beetles (Coleoptera: Scolytidae) forscolytidsontanoakwerereportedinsupplementstotheworld are also associated with burned, coniferous forests; they tend to catalog for this family (Bright and Skidmore, 1997, 2002; Bright, utilize fire-killed trees, and, in the process, enhance the rate of 2014). deterioration in this wood by introducing stain fungi (Lowell In2008,twomajorwildfiresintheBig Surregioncreated the etal.,1992;KelseyandJoseph,2003).Incontrasttoconiferousfor- uniquecircumstancesinwhichtoinvestigatetheresponseofsco- ests,thereislittleinformationontheresponseofscolytidbeetles lytidstothenovelinteractionbetweenwildfireandanon-native, toburnedhardwoodtreesinwesternNorthAmerica,eveninfor- destructive pathogen in the coastal forests of California. Specifi- estswherewildfireispartofanestablisheddisturbanceregime. cally,wewereinterestedinthescolytidsandotheropportunistic Interactionsamonginsects,wildfire,andforestpathogensalso subcorticalbeetlesassociatedwithtanoakinthisregion.Ifoppor- occur but have been rarely studied (Parker et al., 2006), and re- tunisticsubcorticalbeetlesareattractedtotanoaksinBigSurfor- searchonthree-waydisturbanceinteractionsinvolvinganon-na- estsdisturbedbywildfireorP.ramorum,thencompromisedtrees tive forest pest has been even scarcer. An opportunity to study willhavetocontendwiththeadditionalstressofattemptedcolo- such a disturbance interaction occurred in the Big Sur region of nizationbythesebeetles.Opportunisticbeetlesmayreactunpre- California’s central coast duringthe summer of 2008 when wild- dictablytotreesinforeststhatarebothburnedandP.ramorum- firesoccurredforthefirsttimeinforestsimpactedbyPhytophthora infested. For example, beetles may demonstrate a compounded ramorum S. Werres & A.W.A.M. de Cock, a non-native, invasive or multiplicative attractionto tanoakin forestswithbothdistur- pathogen(Metzetal.,2011).Thisdestructiveoomycetepathogen bances,potentiallycausingdamaginglevelsofbeetlecolonization causessuddenoakdeath,anemergingforestdisease,which,since andsecondarydisturbancetothisvulnerabletreespecies. the mid-1990s, has killed millions of tanoaks, Notholithocarpus Inthisstudy,weassessedwhetherP.ramoruminfestation,wild- densiflorus (Hook. & Arn.) Manos, Cannon, & Oh [syn. Lithocarpus fire,orthecombinationofbothdisturbancesaffectedthelanding densiflorus(Hook.&Arn.)Rehd.],andoaks,Quercusspp.,inCalifor- ratesofscolytidsontanoaktreesintheBigSurregion.Thespecific nia’scoastalforests(Rizzoetal.,2002,2005;Meentemeyeretal., objectives of our research were to determine:(1) Which scolytid 2011).P.ramorumcausesmortalityinthesesusceptibletreespe- species and/or other subcortical beetles landed on sticky cards ciesthroughtheformationofbolecankers,whichultimatelylead positioned on the main stems of tanoaks in the Big Sur region; tocambiumdeathandxylemplugging,butthepathogencanalso (2) Whether scolytid abundance was influenced by seasonality causenon-lethalfoliarandtwiginfectionsonawidevarietyofhost and/or forest disturbance; and (3) Whether two forest distur- plants (Rizzo et al., 2005; Parke et al., 2007). Of the tree species bances,P.ramorumandfire,interactedtoelicitanelevatedscolytid killedbyP.ramorum,tanoakisthemostsusceptibletothepatho- landingresponseontanoaks. genandat-riskofextensiveareawidemortalityfromsuddenoak death(Haydenetal.,2011;Cobbetal.,2012).IntheBigSurregion, oneofthemostecologicallydiverseareasinCaliforniabutalsoone oftheearliestinfestedandmostimpactedbyP.ramorum(Masch- 2.Methods erettietal.,2008;Meentemeyeretal.,2008),tanoakmortalitylev- els are especially high (Maloney et al., 2005; Davis et al., 2010; 2.1.Studyregion Metzetal.,2012). Severalspeciesofscolytids,mostofwhichareambrosiabeetles, TheBigSurecoregion(MontereyCo.,CA),situatedonthewes- are associated with P. ramorum-infected oaks and tanoaks (Rizzo tern flank of the Santa Lucia Mountains, is a rugged landscape et al., 2002; McPherson et al., 2005) and have been shown to be crisscrossedbynumeroussteepslopesanddrainages(Fig.1).The preferentially attracted to P. ramorum-infected coast live oak, Q. elevation in the region ranges from sea level to 1571m within agrifolia Née (McPherson et al., 2008). Scolytid entrance holes on 5km of the Pacific Coast, facilitating the wide variety of climatic P. ramorum-infected trees are aggregated in the cankered region zonesandplantcommunitiesthatcontributetoBigSur’secological of the bole and frequently occur before there are any symptoms richness(DavisandBorchert,2006;Meentemeyeretal.,2008;Da- ofdisease in thecrownof thetree (McPhersonetal., 2000). This vis et al., 2010). Redwood-tanoak forests, in which P. ramorum is targeted colonization of live trees is an unusual behavior for mostdestructive(Maloneyetal.,2005),aregenerallylocatedinra- ambrosia beetles of temperate forests (McPherson et al., 2008), vinesandrivervalleysatlowelevations,whereasmixedevergreen asthesescolytidsareknownprimarilyasopportunisticpeststhat forests,dominatedbybroadleafhardwoods,arelocatedpredomi- bore intothesapwood and heartwood of nearlyor recentlydead nantlyonmoistslopesinlowermontanezonesandhighereleva- trees in order to inoculate the symbiotic ambrosia fungi upon tions (Meentemeyer et al., 2008; Davis et al., 2010). The fire whichtheyfeed(Furnissand Carolin,1977). P.ramorum-infected regimehistoryfortheregionisnotwellcharacterized,butthere- coastliveoaksandtanoaksthatarecolonizedbyambrosiabeetles turnintervalbetweenfiresisthoughttorangefrom5to75yrwith displaya65–80%reductioninsurvivaltimecomparedtoP.ramo- anaveragereturnof24yr(Davisetal.,2010).Inthesummerand rum-infectedoaksnotcolonizedbythesebeetles(McPhersonetal., fallof2008,wildfiresburnedlargesectionsoftheBigSurregion: 2010). This hastened mortality may be due to wood decay fungi The Basin Complex-Indians Fire was ignited by lightning strikes thatgainentryintothetreesonthebeetlesastheyborethrough inJuneandburnedmorethan95,000hauntilitwascontainedin the bark to the xylem, or to the weakened structural integrity of late July; the Chalk Fire, located to the south of the first wildfire M.M.Behetal./ForestEcologyandManagement318(2014)21–33 23 Fig.1. Locationofthestudyarea;beetletrappingplotswithassociateddisturbancetreatments;andtheBasinComplex-IndiansFireandChalkFireperimetersintheBigSur region,MontereyCo.,CA. (Fig.1),beganinSeptemberandburnedanadditional6400habe- 2.2.Experimentaldesign foreitwascontainedlaterthatmonth(USDAForestService,2008). We conducted our research in the Big Sur Forest Monitoring Scolytids and other subcortical beetles were sampled from 12 Network, a collection of 280 plots established in 2006 and 2007 forest plots in August–November of 2009 (=‘‘fall’’ 2009 trapping to examine the feedbacks among P. ramorum, its various hosts, season)andinMarch–Juneof2010(=spring2010trappingseason), and thephysicalenvironmentofBig Sur (Haas etal.,2011;Metz approximately 1 and 1.5yr, respectively, following the fires. We et al., 2011, 2012). The plots were 500m2 and distributed in a used a balanced, fully crossed two-factor design based on three stratified-randomdesignacrossredwood-tanoakandmixed-ever- replicate plots for each of four disturbance combinations: (1) P. green forest types in areas with and without P. ramorum (Haas ramorumandfiredisturbanceabsent=nodisturbance;(2)P.ramo- etal.,2011;Metzetal.,2012).Withineachplot,weoriginallymea- rumdisturbancepresentandfiredisturbanceabsent=P.ramorum sured all stemsP1cm diameter at breast height (dbh) (1.37m), disturbance; (3) P. ramorum disturbance absent and fire distur- confirmedthepresenceofthepathogenorotherpests,estimated bancepresent=firedisturbance;and(4)P.ramorumandfiredistur- cover class by species, and measured the cylindrical volume of bancepresent=mixeddisturbance. logsP20cmindiameter(Metzetal.,2011).Immediatelyfollow- The12plotsusedinthisstudywerechosenfortheirrelatively ingcontainmentoftheBasinComplex-IndiansFire,burnseverity highdensityofpre-firetanoak[eachplothadatleast15–25%tan- was quantified in 61 of the 121 plots burned in the wildfires by oakcover(Mueller-DomboisandEllenberg,1974;KentandCoker, using the Composite Burn Index (Key and Benson, 2006; Metz 1992);seeAppendixTableA1formoredetails],andallbuttwoof et al., 2011). In the summer of 2009, all stemsP1cm dbh were the research plots were located in redwood-tanoak type forests. surveyedagainforsurvivalandhealth,andtheliveanddeadbasal Nevertheless, the species composition in our plots was generally areaofeachhostwascalculatedbyusingsizemeasurementsfrom quitediverseandnoneoftheplotsapproachedatanoakmonocul- plotestablishment(Metzetal.,2012). ture. Of the six plots with P. ramorum, the percentage of P. 24 M.M.Behetal./ForestEcologyandManagement318(2014)21–33 ramorum-symptomatictanoakpriortothefiresaveraged70%,and season,rectangular‘‘cages’’madeofvinyl-coatedhexagonalchick- of those burned plots located within the Basin Complex-Indians en wire (dimensions of 17(cid:2)23(cid:2)3cm with 2.2(cid:2)2.2cm open- Fireperimeter,themeanCompositeBurnIndexrangedfrom1.85 ings) were used to surround the sticky cards and prevent the to2.18onascaleof0–3(AppendixTableA1).Thetotalbasalarea unintended bycatch of birds, mammals, and reptiles. The cages (m2/plot) and the proportion of P. ramorum host basal area that were attached to the trap trees with horticultural wire and the diedbetweenthetimeofplotestablishmentin2006/2007andfall sticky cards were positioned inside the cages so that they were of2009wasgreatestinthemixeddisturbanceplotsandthefiredis- flush against the tree bole while the cage extended about 3cm turbance plots (Appendix Table A1). The elevations of the plots fromeachsideofthestickycardsurface. range from 119 to 670m, and the minimum distance between Trapcatchesonthestickycardsweretransportedtothelabora- plotsis400m,althoughthedistancesbetweenmostplotsareon tory (Davis, CA) for beetle removal and identification. To extract thescaleofkilometers(Fig.1,AppendixTableB1).Thespatialdis- the beetles from the sticky card surface, a droplet of xylenes tancebetweenplotswaschosentoensureindependenceforpath- (EMDChemicals,Gibbstown,NJ,USA)wasplacedatopeachspec- ogen effects and beetle activity, but because neither sudden oak imentodissolvetheadhesive,andthepointofascalpelwasused deathnorwildfirewasrandomlydistributedamongtheplotnet- toliftthebeetlesfromthecard.Allbeetlesfromthesameplotand work,thereissomeclusteringofdisturbanceconditionsoverthe deploymentperiodwerestoredinavialofxylenesforseveraldays largespatialscaleofthenetwork(Fig.1).Burnedplotsareneces- todissolveanyresidualstickytrapadhesive.Priortospeciesiden- sarilywithinoneoftwodifferentfireperimeters,whereasP.ramo- tification by microscopy, vacuum filtration through filter paper ruminfestationismorecommoninthenorthofBigSurthaninthe wasusedtoseparatethespecimensfromthesolvent.Allscolytids south(Fig.1).However,wecheckedmodelresidualsintheanaly- wereidentifiedbyMMBandSJSbyusingtaxonomickeysandspe- sesdescribedbelowforspatialautocorrelationbyusingMoran’sI ciesdescriptions(BrightandStark,1973;Wood,1982;Hobsonand andthespdeppackageintheprogramminglanguageR(RDevelop- Bright,1994),exceptforonespecimenofCyclorhipidionbodoanum ment Core Team, 2009; Bivand, 2013) and found no evidence of (Reitter) (=Xyleborus californicus Wood), which was identified by spatialautocorrelation. D.E.Bright(ColoradoStateUniversity,FortCollins,CO).Buprestid beetleswereidentifiedbyR.Westcott(OregonDepartmentofAgri- 2.3.Beetlesampling culture, Salem, OR) and S. Bíly´ (Department of Entomology, Na- tional Museum, Prague, Czech Republic). Other subcortical Beetlesweresampledbyusing14(cid:2)20cmyellowstickycards beetles were identified to the species or family level by MMB (AlphaScentsInc.,Portland,OR,USA)attachedtothreelivetanoaks andSJS.VoucherspecimenshavebeendepositedwiththeBohart perplot(=traptrees).Twostickycardswereusedpertraptree,and Museum of Entomology, University of California, Davis, CA; the each one was placed approximately at breast height (1.37m) on CaliforniaAcademyofSciences,SanFrancisco,CA;andtheNational theeast-andwest-facingsideofthetreebole,foratotalof6traps Museum,Prague,CzechRepublic(Buprestidaeonly). perplot.Althoughweendeavoredtouseonlytraptreeslargerthan In this paper, we have elected to use the original family-level 20cmdbhinthisstudy,thedbh’softraptreesrangedfrom8to nomenclatureforbarkandambrosiabeetles(Coleoptera:Scolyti- 83cmwithameanof42.5[±3.19(S.E.)]cm.Thethreetraptrees dae)basedontheargumentpresentedinWood(2007)andamore per plot provided a consistent number of tanoak sample units extensivetreatmentoftheissuedevelopedbyD.E.Bright(personal acrossthefullrangeoftreespeciesdiversityavailableonthe12re- communication),whichistobepublishedinhisthirdsupplement search plots. In some instances, plots may only have had four or to the world catalog of the Scolytidae and Platypodidae (Bright, fivelivetanoaktreesavailableintheappropriatesizeclass. 2014).Inessence,morphologicalandfossilevidenceofadultsco- Thethreetanoaksselectedfortraptreeswithineachplotwere lytids support the family-level treatment, whereas similarity in representativeoftheoveralltanoakinthatplot.Forinstance,with- scolytid and curculionid larval morphology supports a subfamily infiredisturbanceplots,thebolesoftraptreesweretotallyorpar- placement. Because this issue is not entirely resolved, we prefer tiallycharred,andwithinP.ramorumdisturbanceplots,P.ramorum to take the more conservative and economical approach of using cankerswerepresentonsome,butnotall,traptrees.Withinmixed theoriginalnomenclature. disturbance plots, most trap trees were charred and some were knowntohaveP.ramorumcankerspriortothefires(thepost-fire 2.4.Dataanalyses charringmadeitdifficulttoseetheircurrentcankerstatus),andin controlplots,traptreeswereneitherburnednordiseased.Asthe Withineachfieldseason,weanalyzedthelandingratesof(1)all damagefromthedisturbancesdifferedamongplots(asshownin scolytidspeciescombined,and(2)eachofthefourmostabundant AppendixTableA1),thedamagelevelandhealthofthetraptrees scolytidspeciesinthetrapcatches,byusingpooledcountsoftrap alsodifferedamongplots. catchesfromeachstickycarddeploymentperiodfromeachofthe Samplingin2009commencedinmid-Augustinmostplotsand four disturbance treatments. The seasonal activity of the beetles concludedinearlyNovember,whereassamplingin2010ranfrom waspresentedbyplottingthesemeasuresversusthemodecollec- mid-MarchtothesecondhalfofJune(AppendixTableB1).During tiondatesofthestickycardsineachtrappingseason.Foranalysis thetwosamplingseasons,thestickycardswerecollectedperiod- oftherelationshipbetweenscolytidcatchesanddisturbancetype, icallytoassessbeetlelandingratesandreplacedwithcleansticky wesummedtrapcatchesfromallstickycarddeploymentperiods cards. The deployment lengths of sticky cards ranged from 7 to withineachfieldseasonperplotforeachbeetlespeciesorthefam- 31days, with a mean deployment period of 18days in fall 2009 ilyandthendividedbythetotalnumberofweeksoftrappingin and24daysinspring2010(AppendixTableB1).Insomeinstances, thatplot.Wecomparedthesestandardizedlandingrates(i.e.,the different tanoaks were used as trap trees during the spring 2010 number of beetles trapped/plot/week) with a two-factor analysis season compared to the fall 2009 season; this was due primarily ofvariance(ANOVA)andthecrossedfactorsofP.ramorumandfire. to structural failure or death of trap trees during the winter of Tukey’s ‘‘honestly significant difference’’ (HSD) test was used for 2009–2010. posthocpairwisecomparisonsamongtreatmentmeans.Alltests During the fall 2009 sampling season, sticky cards were at- and analyses were performed with the statistical programming tachedtothetraptreeswithgreenhorticulturalwire(Woodstream language R (R Development Core Team, 2009) and significance Co.,Lititz,PA,USA)wrappedtautlyaroundthetreeboleatboththe wasdeterminedforcriticalvaluesofa=0.05,althoughresultswith top and bottom of the cards. During the spring 2010 sampling P<0.1werealsonotedtoindicatesuggestivetrendsinthedata. M.M.Behetal./ForestEcologyandManagement318(2014)21–33 25 Prior to performing the ANOVAs, each standardized data set abundant of these beetles were several species of Anthaxia wastestedforhomoscedasticitybyusingatestforoutlyingvari- (Buprestidae), includingA. cupriola Barr, A. retifer LeConte,and A. ance(a‘Cochrantest’)containedintheRpackage‘outliers’(Kom- strigata LeConte (Table 1). These buprestids were trapped exclu- sta,2011),andthosedatasetsthatdidnotmeettheassumptionsof sively in the spring 2010 trapping season and primarily in the homoscedasticityweretransformed.Datawerealsoexaminedfor mixeddisturbanceplots(Table1).Anthaxiaretifer(108of148iden- normalitybyevaluatingthemodelresiduals.Inmostcases,alog- tifiedspecimens)wasthedominantspecies(Table1). Otherbee- arithmic transformation [x0=log(x+1)] was used, but in certain tles trapped in small numbers included: Scobicia declivis instances, other transformations were required in order to make (LeConte) (Bostrichidae); Colydima lineola Say (Zopheridae); Ptili- the data homoscedastic and meet assumptions of normality in nus basalis LeConte (Anobiidae); and members of the following the model residuals, including a square root transformation families:Anobiidae,Ciidae,Elateridae,Monotomidae,andTenebri- p (x0= x) and a rank transformation (Conover, 1980). There was onidae.Largerwoodborerstypicallyassociatedwithhardwoodsin oneinstanceofspeciesdatathatremainedheteroscedasticdespite California[e.g.,theflatheadedappletreeborer,Chrysobothrisfemo- transformationandonwhichanANOVAwasnotperformed[Mon- rata (Olivier) (Buprestidae), the oak cordwood borer, Xylotrechus arthrumscutellare(LeConte)landingratesfromspring2010]. nauticus (Mannerheim), and Neoclytus conjunctus (LeConte) (both Cerambycidae)]werenottrappedinourstudy(SwieckiandBern- hardt,2006). 3.Results 3.1.Speciescollected 3.2.Seasonalflightactivity Overthecourseofthe2009and2010trappingseasons,atotal Within each field season, peak landing rates on tanoak trap of 2779 scolytids were collected and identified from the sticky trees were variable for different beetle species as well as within trapsattachedtotanoak;75%ofthesescolytidsweretrappeddur- individualspeciesindifferentdisturbancetreatments.Forallscol- ingthefall2009season(Table1).Thiscollectioncomprisedseven ytidspeciescombinedinfall2009,thepeaklandingrateontanoak species, fiveof whichare ambrosiabeetlesand two of whichare in the mixed disturbance plots occurred between August 24 and barkbeetles.Twooftheambrosiabeetlespecies,X.saxeseniiand September 3, but there was no definitive peak catch for any of Gnathotrichuspilosus(LeConte),madeup48%and40%ofthetotal the other treatments (Fig. 2A). The peak in the mixed disturbance scolytidcatch,respectively.M.scutellareandM.dentiger(LeConte), plots was largely driven by the extreme spike in the number of alsoambrosiabeetles,werethenextmostfrequentlytrappedsco- X.saxeseniicaughtduringthistimeperiod(Fig.3A),thoughM.scu- lytids,makingup6%and5%ofthetotalscolytids,respectively.The tellarealsohadaprominentpeakcatchatthistime(Fig.3C).Dur- remaining1%ofthetrappedscolytidswassplitamongthreespe- ing the 2010 field season, the highest catches for all scolytid cies:P.pubipennisandHylocurushirtellus(LeConte),bothbarkbee- species combined and across all disturbancetreatments were re- tles,andC.bodoanum,anambrosiabeetle. cordedbetweenMay29andJune22(Fig.2B).Thistrendwasinflu- Inadditiontothesescolytidspecies,avarietyofothersubcorti- encedbycatchesofX.saxesenii(Fig.4A),G.pilosus(Fig.4B),andM. cal beetles were collected from the sticky cards. The most dentiger(Fig.4D). Table1 TrapcatchesofScolytidaeandBuprestidaecollectedfromstickytrapsinfourforestdisturbancetreatmentsduringthefall2009andspring2010trappingseasonsintheBigSur region,MontereyCo.,CA. Beetlescollected Fieldseason Disturbancetreatments Totalsfromalltreatments None P.ramorum Fire Mixed Scolytidae Xyleborinussaxesenii(Ratzeburg) 2009 0 62 217 757 1036 2010 1 1 51 247 300 Gnathotrichuspilosus(LeConte) 2009 1 47 16 810 874 2010 0 1 19 217 237 Monarthrumscutellare(LeConte) 2009 0 18 3 117 138 2010 0 19 0 21 40 Monarthrumdentiger(LeConte) 2009 0 4 16 5 25 2010 0 4 46 68 118 Pseudopityophthoruspubipennis(LeConte) 2009 0 0 0 1 1 2010 0 0 3 5 8 Hylocurushirtellus(LeConte) 2009 0 0 0 0 0 2010 0 0 0 1 1 Cyclorhipidionbodoanum(Reitter)a 2009 0 0 0 1 1 2010 0 0 0 0 0 Totalscolytids 2009 1 131 252 1691 2075 2010 1 25 119 559 704 Buprestidae Anthaxiaspp.b 2009 0 0 0 0 0 2010 8 0 13 141 162 Boldvaluesareusedforthenumbersoftotalscolytids. a Cyclorhipidionbodoanum(Reitter)(=XyleboruscalifornicusWood). b Ofthe162Anthaxiaspp.collected,148specimenswerecuratedandidentifiedas:A.retiferLeConte(108specimens;99inmixed,8inFire,and1inNodisturbance);A. strigataLeConte(39specimens,allinMixed);andA.cupriolaBarr(1specimeninMixed). 26 M.M.Behetal./ForestEcologyandManagement318(2014)21–33 Fig.2. Mean(±SE)numberoftotalscolytidstrappedperplot(n=3)onstickycardsinfourforestdisturbancetreatmentsatdifferentcollectiondatesinfall2009(A)and spring2010(B)intheBigSurregion,MontereyCo.,CA.Thecollectiondateslistedforeachtrappingseasonarethemodecollectiondatesforall12plots. Fig.3. Mean(±SE)numberofXyleborinussaxesenii(A),Gnathotrichuspilosus(B),Monarthrumscutellare(C),andMonarthrumdentiger(D)trappedperplot(n=3)onsticky cardsinfourforestdisturbancetreatmentsatdifferentcollectiondatesinfall2009intheBigSurregion,MontereyCo.,CA.Thecollectiondateslistedarethemodecollection datesforall12plots. 3.3.Effectsofdisturbancesandtheirinteraction 92% more scolytids were collected in fire disturbance plots com- paredtoP.ramorumdisturbanceplots,andonlyonescolytidbeetle Allscolytids.Infall2009,81%ofallscolytidscollectedduring wastrappedintheplotswithnodisturbances(Table1).Log-trans- thatseasonweretrappedontanoakinthemixeddisturbanceplots, formed landing rates of all scolytid species combined differed M.M.Behetal./ForestEcologyandManagement318(2014)21–33 27 Fig.4. Mean(±SE)numberXyleborinussaxesenii(A),Gnathotrichuspilosus(B),Monarthrumscutellare(C),andMonarthrumdentiger(D)trappedperplot(n=3)onstickycards infourforestdisturbancetreatmentsatdifferentcollectiondatesinspring2010intheBigSurregion,MontereyCo.,CA.Thecollectiondateslistedarethemodecollection datesforall12plots. significantlywithP.ramorumdisturbance(P=0.01)andfiredistur- significantdifferencesinmeancatchesperweekamongthetreat- bance(P=0.006)(Table2),andthemeanlandingrateperweekper mentgroups(Table3). plotwassignificantlyhigherinthemixeddisturbanceplotsthanin Individual species. The effects of the disturbances on landing the no disturbance and P. ramorum disturbance plots (Table 3). rates of the four most abundant ambrosia beetle species were, However,eventhoughlandingratesontraptreesinthemixeddis- forthemostpart,similartotheresultsforallscolytidspeciescom- turbance plots were more than an order of magnitude larger bined(Tables2and3).Therewerethreeresultsthatwerenotably than those in the otherplots, the interaction term in the ANOVA different for individual species than for all scolytids. First, in the modelwasnotsignificant(Table2).Thiswaslikelyduetothefact fall2009trappingseason,G.pilosuswastheonlyspeciescollected that the logarithmic transformation that we employed most fre- inthisstudyinwhichbothdisturbancefactorsweresignificant(P. quently(andisappropriatefordatainwhichthemeanandvari- ramorum: P=0.006; fire: P=0.014) and the interaction between ance are positively correlated) often eliminates significant thesedisturbanceswasclosetosignificant(P=0.058)intheANO- interactiontermsbecauserelationshipsthataremultiplicativeon VA model of landing rates (Table 2). The mean G. pilosus landing alinearscaleareadditiveonalogarithmicscale(GotelliandElli- rateperweekperplotwassignificantlygreaterinthemixeddistur- son,2004). bance plots than in all other treatments, including fire (Table 3). Inspring2010,therewerefewerscolytidstrappedthaninfall Second, in contrast to the trends noted for the other species 2009, but the majority of scolytids (79%) were still collected in trapped in this study, more M. scutellare were trapped overall in themixeddisturbanceplots(Table1).Inplotswithonlyonedistur- theP.ramorumdisturbanceplotsthaninthefiredisturbanceplots bance,476%morescolytidsweretrappedontanoakinthefire-dis- in both trapping seasons (Table 1). P. ramorum disturbance was turbed plots than in the P. ramorum-disturbed plots, and, as in the only significant factor (P=0.017) in the 2009 ANOVA model 2009,onlyonescolytidbeetlewastrappedintheplotswithnodis- of M. scutellare catches (Table 2). Third, M. dentiger was the only turbances(Table1).IntheANOVAmodelforallscolytidstrapped speciescollectedinthisstudyinwhichthegreatesttotalcatchin inspring2010,neitherofthedisturbancefactorsnortheirinterac- atrappingseasondidnotoccurinthemixeddisturbanceplots.In tion were significant, although fire disturbance fell just short of the case of M. dentiger in 2009, the majority of individuals were being considered significant at P=0.05 (Table 2). There were no trappedinthefiredisturbanceplots(Table1). 28 M.M.Behetal./ForestEcologyandManagement318(2014)21–33 MontereyCo.,CA,withPhytophthoraramorum P.ramorumfire(cid:2) NS1.03;0.341,8NS0.12;0.741,8 NS2.82;0.131,8NS0.57;0.471,8 (cid:2)4.84;0.0581,8NS0.10;0.771,8 NS0.12;0.741,8– NS0.86;0.381,8NS0.003;0.961,8 ndfiretreatmentj(2levels,present/absent); (cid:2)***ficancesymbol:P<0.1;P<0.05;P<0.01; geneityofvariancewasconfirmedfollowing bSftwbpdab4ohumeyo.rielseoertDltrAboshmtPueirelrwm.sereodpccbisgirsbonitunaasiierclsoagtnsmldooeuenscbscloicrtehlioetbtarehasieeunncage,(dtmtnbdhaeleecewelixdlnsteercahibstwelfntbeldruphetdoeefithtsig/ermsroatretecwensfrlsoetotsaro.urtfimitalniedarIcoplpseMynklfuptou.ydntew.sschOiddectnaeserfdaitgocrascrneuntttdhrtuhitetahnseetraertalsegbaralenalepttsoyovrtwtdpcehaaao3ec)oceskpn,0udhtli0rofetoanirom0antdpnersnfdpdeaaoftooosmjooraoitrrernskerbttsipiduastgrtttltnioynnhpohssopiiettlasifioousalantotcfkfriwastcabbsPslsncaeil.uitndonetnrhoblanitacysnlcttetmoetuoheasusnmrroie,rnyltetrbyifihoBeiucneraroamiaPdggeertfl. Table2aANOVAmodelforlandingratesperplotperweekofallscolytidspeciespooledandforthefourmostabundantscolytidspeciestrappedinfall2009andspring2010intheBigSurregion,disturbance,firedisturbance,andtheinteractionofP.ramorumandfiredisturbancesasfactors. cbSourceFieldseasonTransformationDisturbancefactors P.ramorumFire ***0Allscolytids2009x=log(x+1)11.07;0.0114.07;0.0061,81,8(cid:2)NS02010x=log(x+1)0.98;0.355.31;0.051,81,8 (cid:2)*Xyleborinussaxesenii2009None4.45;0.06810.53;0.0121,81,8(cid:2)NS02010x=log(x+1)0.57;0.473.71;0.091,81,8 ***0Gnathotrichuspilosus2009x=log(x+1)13.72;0.0069.84;0.0141,81,8NS***2010Rank0.60;0.4629.17;<0.0011,81,8 (cid:2)*Monarthrumscutellare2009Rank8.95;0.0174.78;0.061,81,8d2010NA––pNSNS0=x0.036;0.860.54;0.48Monarthrumdentiger2009x1,81,8NSNS02010x=log(x+1)0.08;0.791.78;0.221,81,8 aleANOVAmodel:Y=+(P.ramorum)+(fire)+(P.ramorumfire)+whereYisthelandingrateinreplicateplotk(#1–3)withdiseasetreatmenti(2levels,present/absent)a(cid:2)ijkijijijkijkleisthetruegrandmeanandistheerrorterm.TherewasnospatialautocorrelationdetectedintheresidualsofanyofthenineANOVAmodels(allP>0.81,Moran’sITest).ijkbForeachtableentryinthe‘disturbancefactors’columns,thefirstnumberistheF-valuewithdegreesoffreedomassubscripts,andthesecondnumberistheP-valuewithasigni***P<0.001.cDataforthestandardizedcatchesweretransformedtomeettheassumptionsofhomogeneityofvariancespriortoANOVA,exceptintheinstanceofX.saxeseniiinfall2009.HomotransformationwithaCochrantest;normalitywasconfirmedbyanexaminationofmodelresiduals.dLandingratedataforM.scutellarein2010didnotmeettheassumptionsofhomogeneityofvarianceswithanytransformation,andthusanANOVAwasnotperformed. 1t((b–1atttaactooihfiMtWwisaCw(sbt1aateMoriorhc4nnnsrWh2rrhhh1ioaetcoobnmnsnll.fnf9e9u9ooanoae1arshhagovmuru))aeeeroleudddsNl89su9tohteetpam.ylMsoOMibdteieltsssiothneahntH222cnBwnyontaooabiSp,lSeota.eedCtretfhdcnnmehieod.r;;;tnhtuthigstnohet.leluudmdr.etGhtguityoudseeant2altobeheeessDaddeHRachmmmeh,tCtsy.em.siwdrrcnit0bbrpcnanSdaeesannaf(iieAaatkroa1aeoipuOtornrto1koBusisnonnGdsneobnltloh.l,bront9triliotssruNldsbfdn3ntisrrroltnyeoeah1e.eHoaftr(oefafsooo8imtBhti.rogei)lsofiaoBlnoeypgaich9rgoertevtNluasrsfcl2o.nNrienatgtesutketaoahkrlirOrsurmhn7tdaieilrrt‘ilrr;iCpneaetntsCigQgsau‘trrofeataahslntt3eBupglthehceaot;sebuAed(aaiihtoabtehsSrstuphswbu)haootoortrs,uriaorcemueebanstetmnutae.teaAsepaeeb(yanrtsitluisp,rttaenhdPsaeOonedrnonaSBmhretadeesmu,hysa,i1Bdadlneacn.sdaatexnnawtttn(ildtfet.d2elshfau9lt,cnibBalaiyBevykseKclaefrpodleHegtefipse0tH.raiausot29ullSd,ndeorerrrhnr,(ueetstiraeyeo1nBssiiehteisnfti.3rS02nsulmoaiscnres2sbgsgmgchtaonSpsp3tipgapduano)1nn)oaabgrtIpCstih0ahhitoiorttire.or)rnaoipobsteicoUkrrnl2odyoesaMhpksol0.tMeetbfcvfChAyrrtaecrtraynsite,a),trrrX,Si.lravr6daeuahsynstttoeaAniuln]renyl-koe(2n.e1fAeeitht.cipam.lwe)r1io,(naakudSs(rnr.dd,,gdocnhyeen0.9ssaddoBnBce2boeld9danrdinice21,ialsuititoabl0gIe9tlscrbecahdomtra0dsnu7truysoetytx0iM9,nssuni8rhacii4Shcopacntubnoy—1apegt3rcgwettelio17acttcdtog;euaatpa;nygarecnsiheps,3erhush;iseesaaos3(od3MnethntegnpiNstpyenesdWer)NitrsebothpBrnnbtni)vs;sWetnadoe.eeGnrieskewtwa]loeCleapiaefiserisaeedaAbvWaisfchyouoo.l,ounb.isensprrteraloelnotanscXoee,eki,naopns1formtpAupbteoonioxntento[iuadotodrrAldh.nsuodtiidnte9hTdeaeensalhrelceaGlenndlndos.eanoicoscosirlattrrctc7hkaiiomlSxenotztyS.,adthaoctrdeerlrteawgsnciiesith3lefteytaehieadefteoetnusxys,ltmyprpoy1ppanthda;neeSlmttMnnrssaCowdefi1sduielo[rN9leistograaetrsdelolSsvooaafmttapCauk9coohtkrcieo8ienunloeefcnrhatatBealtssgntinlto,.htocfris8,chraam.ont2s,litolstootoneeeerk,armocude,iadf1iyautepn.2utn2eiiio1;dzridosodv2aotlemdsnsitovs9rgtunr;ueelrs0ame9ashienrk,bht0dnCtyeWe—reMah7cootvhatnmso0Wnr7l(tnau0orpafitBasiy–lioin3tseanSwAarie8il4inshyn3steoono8nsl,e,ranainceirpa;vbHota.il;z.tadoebn;vcmownoo)yfid1orfi3sibtcdpesneossbceS]oac[Safbiuittfidtm9apvr)tWpeh,psvaiaeneindshoakelerasso,aratftiNtnie9atlbreeetlninoednnniefapeyirnihneybrmaoftdeaootglaotnes2nuccaecsd,nkinahotdeletonilrfnucanatunroioMit)tytaAotnyaeecryttsofihmceadruhtaMdboarnhddr’eahetc.rinatiAa,’aeuontrcieBBoselrneaeeuel(ls,odocCestenr(td.s.ctaruleBBagp2tWnstaakveinucdBaodlallarnoe1ocdTrolsvlhseodo0rrorrnelttmwplct,rniryuisutr9huchiiebbnaehyrcoasgnHtrri1lrwieuggupdopoCnoaonahnu8gttettrdeeaaisaooorif3hhidnnlreeeeeraepAznvpghoaeo2rerterrmiootessdst)iinttdddddykyygaeaeeaaeesssssr---r---tttt);:;ff,,, M.M.Behetal./ForestEcologyandManagement318(2014)21–33 29 d these species, but the larvae of Anthaxia spp. generally feed on n CA. Mixed 46.42±17.53b13.73±12.26a 20.78±5.17b6.07±5.44a 22.24±15.44b5.33±5.18b 3.21±3.01b0.52±0.17 0.14±0.14a1.67±1.63a meachsourcea hbbaorloroieatsnnhtc,sc1oAhm9f.oe7csmru7pAo)o.rf.nSiorlcpyeloeatnficfoaiiefenuresdnr(osdBA,fa.ohCrrnuare,rptydi1rfeewe9lsrl7osohu1owas;dvasNoen,redbalwnsePoehdinnnitsuecehsotflrhluolaaebwlvc.s,ete,e2rbwd0se0h(feoF8enurr)aelriagansnistinsedtsgdhaaoaednnsuadlflldatCorsuvawlotars--fl Co., afro ers from a wide array of plant species (Barr, 1971; Nelson et al., nterey ondat 2el0e0v8a)t.edThiunsc,earlttahionupglhotthtyepirespionpuoluartisotnuddye,ntshietieressmpoanysehsavoeftbheeesne Mo ed beetlesmayhavebeendirectedtowardthecolorofthestickycard m on, or traps (similar in color to flowers) and not to volatiles from the egi erf tanoaksinthevariousdisturbanceclasses. r p BigSur 2ab1a 2ab4a 9a0b 5ab 4a3a mation 4.2.Seasonalflightactivity asonsinthe Fire 7.09±6.12.92±2.1 6.07±5.71.25±1.1 0.49±0.00.47±0.2 0.08±0.00 0.44±0.41.13±1.1 eoftransfor pcareuWvseioepuceshalyoksflreeigtchootrtdpreaedpriosdduusbrcfioonrrgtMict.ahslecsbueteeelstlaelearsesoainnnsdthXien.sfaacxloleasasentnadilishCpaarvliinefgobrebneeina- e p gs ty (Bright and Stark, 1973; S.J. Seybold, unpublished data). Of these pin for twoseasons,fargreatertotalcatchesofscolytidsoccurredinthe p 2 2010tra toTable ftsahulemls(emesbepere.ect[ilMaels.l—ydiednnutgriigenengrerlwaatla—esdAtihusegpueorsnsteeanmedxocSreeepprtteieoamndibalyemri)on,nitnghdetihcfeaatllisncogorltylhattaidet spring Refer s2p0e1c0iecso,maspiatrewdatsotrtahpepfeadllionfg2r0e0a9te,rasnwumerbeetrhseintwthoeplraitmeasrpyriAnngtho-f thefall2009and P.ramorum 3.51±2.44a0.61±0.54a 1.66±0.92a0.02±0.02a 1.26±1.06a0.03±0.03a 0.48±0.40ab0.47±0.39 0.11±0.07a0.10±0.10a wisecomparisons. awaHsilxaisaisoastbotsberhpeeyepte,ln.2ec,a0wdms1ieh3saip)dic]neeh.rFosiwanaullerltrahmsenteuadoysdngpylbayr,eilpnltetrrghraeypev(SpiisoaneeunidatstisaadoctunoBiorlalminrenbcogiatnsirttoahcncoeMosansosupdtfsarueAilcnu.CigmvsatemlriiotfgooofanrNtantamhaiathsbu.abrrAvoeaes--l during orpair ccaounsdeitiotnrseeosfthmeatyypibcaellyshtroetssaendddfroyllsouwminmger.thFuerthderromugohret-,ltihkee eatments aments HSDtestf ltaahntedesepfarbolelveirtdaleeinsf,yowosdheoafsoseorngthamlelealrayirevsbaeme(ucEosbntedblueincmgiv,oe1is9tt7of8ot)rh.tLehaendfdeuvinnegglourpsamtteoesgnmrtoaowyf 3)onstickytrapsinfourforestdisturbancetr nDisturbancetreat None 0.03±0.03a0.03±0.03a 0a0.03±0.03a 0.03±0.03a0a 0a0 0a0a ficantlydifferent(P<0.05)basedonTukey’s etoalackofhomogeneityofvariances. aaiht(ohfisrspnrseppltvaeieeestdfgprervdeeoreihuir.eancos)teccdthgdtrrdtrdithaieai,ois2uaopevipdvnct0mrnseeuouis1innspscdsat0itbigseegnuipacdiesdseloktotseacduthncystrontsasoheailiicopifnypebongasnttptlrrpruriteheSiddahrdrbneaetiteeecnuhgettctranehgartpraotienrstieoclpgoffiotehntfiisshfhtsinsetcesece2ia4sisnaoref0tb.psaw3lunr1oyytela)scah0lfeta.tyifir,eekids2Arenrssce0oddesoiohsop0nftudluiihfryti9teingtlethieatthnirieinobttdgectsoohnerssesefeancp2atptvpitalor0hatsnelhiuity1tuennotes,0trmtlgnehielhmtt/ka.htiemeseermroHousegalefted,ymoepirnrunoiw(arpfidtm1srgpahiicee)nnno(eaavot2gdsforttterfeih)eftflrotscfPhee,hpaeinwt.gestleao.(hlprhirsttAoarBeoeotoreenmraetfnnpmtwepccoothe2naapmitoarra0auhtgiuisdxl0oinaseemlaiid8yygaaesr-l plotperweek(n= Fieldseaso 20092010 20092010 20092010 20092010 20092010 entlettersaresigni nthe2010datadu sfsautpcurtetudAchayileetllhrsayorocoeuoracgntchhcduleirousruispenirdeoflrrnuasesefestnuaeicbrnlteoswsgusreotuefctaghtotgheneeescsltltifufonetdrhueehamsdittsbttmdroeiarrospsitseputsiirnnsbocgaffo.anltlychl,eteiswdsosepniecncsaictnfiohncleyostBtcidiomglsyaSetkuaider- er er o sonalitybecause:(1)wedidnotrepeattrappingduringthespring ytidstrappedp ollowedbydiff werenotmade admtinuvadretienftagdollathtstaeeemaamstpoieondru-airsntuurmmeresmue,laetmrircpoholiersstyiwuteeriaesnr.,tseA;armn(;2dba)nrtoodrstahi(pa3ep)rbinwecegletimlwdesaiadstaincrneoofttqaucccootitolnelredsscu,etcnactnseliidd-- Table3Mean(±SE)numberofscol Source Allscolytids Xyleborinussaxesenii Gnathotrichuspilosus bMonarthrumscutellare Monarthrumdentiger aDatainthesamerowffieldseason.bPairwisecomparisons mattflotshaicrnuoneoieccnalcttrmyshseuoteipcai(fxdltreeediio.mrdogumsn.yrdap,saetieltsetaochtiirnuowie,eonresabftorhnaocr(ednfeeeWscalspeteenehohvpavoeyairlnadstoctiri,htiociPmsaan.1bvlsi9rnileaadte8rgnemes2sdsvuo)oce.sfrcalluolohndFomcpioisagshmretphdaereeeivedsrrnpextstueraai,oalnrmlebtxvhfltaipdrmaeinlegitecfmiihnf,etoetyunersvpmneatlafnoodrobnittoreadsturth;dafiseuoiosspenriftonateuhgcrflriteeFenubinairceagneudnn.trlihcc1aaisneee-r-) 30 M.M.Behetal./ForestEcologyandManagement318(2014)21–33 may have contributed to differences in scolytid abundance and aggregation pheromones in suitable situations, and apart from seasonalityamongdisturbancetreatments.Coastalfogandarapid host-producedvolatiles,thesepheromonesarethedominantsemio- increaseintemperaturewithhigherelevationsarecharacteristicof chemicalsthatstimulatemassattackuponatree[seeSeyboldand themarineinversionlayeralongtheBigSurcoastline(Davisetal., Vanderwel(2003)andreferencestherein].Thelargenumbersofsco- 2010),makingweatherconditionsquitevariableevenamongsites lytidstrappedonspecifictanoaksatcertaintimesinthisstudy(often locatedrelativelyclosetooneanother. whenveryfewbeetlesweretrappedonothertraptreesinthesame plot)suggeststhataggregationpheromonesmayhavebeenreleased 4.3.Effectsofdisturbancesandtheirinteraction fromtreescolonizedbyambrosiabeetlesinthedisturbedplots. To explain why greater numbers of beetles were trapped on Althoughambrosiabeetlelandingratesontanoakweresignifi- tanoakinplotswithbothdisturbancescomparedtoindividualdis- cantlyhigherinforestsdisturbedbothbyfireandP.ramorum,itis turbances,itmaybesimplertoexaminetheoveralllevelsoftree important that the landing rates presented in this study are not mortalityinthedifferentdisturbancetreatmentsratherthanspec- interpretedasratesofhostcolonizationbecausewedidnotquan- ulate on the synergistic effects between host volatiles or phero- tifythenumberofentranceholesonthetraptreesorverifythat mones produced in P. ramorum-infected and burned trees or the theambrosiabeetlespeciestrappedonthestickycardswereactu- volume of scorched wood onthelandscape. Althoughthe data on allyconstructinggalleriesandsuccessfullyrearingoffspringinthe plot-level mortality from the time of plot establishment in 2006/ tanoaks. One hypothesis of how scolytids find appropriate host 2007tothefallof2009areincomplete,thetotalbasalareaaswell treesisthattheylandrandomlyontreesandtesttheirsuitability astheproportionoftreeswithintheplotthatdiedduringthosetwo by short-range olfaction or gustation (Moeck et al., 1981; Byers, orthreeyearsisquitelargeinthemixeddisturbanceplots(aswellas 1995;Gravesetal.,2008),andthusthelandingratesofambrosia in one of the fire disturbance plots) (Appendix Table A1). Greater beetlesarenotnecessarilyequivalenttotheiractualratesofcolo- quantitiesofmoribundandrecently-killedtreesinaplot,whether nization. Nevertheless, signs of ambrosia beetles colonization on fromP.ramorum,fire-injury,orboth,mayhaveledtogreaterpopu- tanoaks in the disturbed plots were frequently noted over the lationdensitiesofambrosiabeetlesinthoseareas. courseofthisstudy(includingonsometanoaksusedastraptrees), In addition to killing burned trees that would have ordinarily and,becausetanoakisthepredominanthardwoodcomponentof recoveredfromtheirinjuries,largepopulationsofbarkbeetlesthat theforestsin ourresearchsites,these observations stronglysug- emergefromfire-injuredtreesmayalsodispersetocolonizenear- gestthatambrosiabeetlelandingratesontanoakswerelikelycor- bygreentrees(FurnissandCarolin,1977;Wood,1982).Thissce- relatedwithratesofcolonization. nariohasnotbeenobservedwithambrosiabeetlesintheBigSur Wepositionedthestickycardtrapsonthelowerstemsoftano- region,norisitexpectedtooccurduetotheopportunisticnature aksinthisstudybecausewewantedtolearnwhichscolytidswere ofambrosiabeetlesandtheamplesupplyofrecentlydeadanddy- potentially using tanoak as a host, but this methodology also al- ingtreesinthearea.However,giventheimperiledstateoftanoaks lowedthe characteristics ofthe individual traptrees toinfluence in coastal California from P. ramorum-caused mortality, even the scolytid landing rates. Semiochemicals emanating from the trap lossoffire-injuredtanoakstoambrosiabeetlecolonizationwould treesverylikelyaffectedambrosiabeetlelandingrates,asthepro- haveamajorimpactontheforestsoftheBigSurregion.Further ductionofvolatilecompoundsthatareattractivetoscolytidshas researchisneededtodeterminetheactualfrequencyofambrosia beenproposedand,insomecases,demonstratedtooccurintrees beetlegalleryinitiationinlivingtanoaksandwhethercolonization infected with P. ramorum and injured by fire. McPherson et al. hastensorleadstotreemortality. (2005, 2008) suggested that ambrosia beetle entrance holes are concentratedinthecankerregionofP.ramorum-infectedtreesbe- Acknowledgements causeofolfactoryresponsestovolatilecompoundsproducedinthe cankerregion.Ockelsetal.(2007)usedgaschromatography–mass TheauthorswouldliketothankKerriFrangioso,MarkReaddie spectrometry to identify several different volatile phenolic com- andFeynnerAriasattheLandels-HillBigCreekReserve,theMon- poundsintheexudatesofP.ramorum-infectedcoastliveoaksthat terey Peninsula Regional Park District, California State Parks, Los maybeinvolvedinhostdefenseresponsesbutalsoactasattractants Padres National Forest, Don McQueen, and assorted members of toscolytids.ResearchbyKelseyetal.(2013)demonstratedthateth- theRizzolabfortheirassistanceonthisproject.WealsothankYi- anol, a compound known to elicit aggregation behavior in a wide genChen,UC-DavisDepartmentofEntomology,forhisadviceon rangeofambrosiabeetles(KelseyandJoseph,1999;Seyboldetal., thestatisticaltreatmentofthe data, and DonaldBright,Colorado 2000; Coyle et al., 2005; Miller and Rabaglia, 2009; Gandhi et al., StateUniversity,RichardWestcott,OregonDepartmentofAgricul- 2010), was present in higher concentrations in the sapwood of ture,andSvatoplukBíly´,NationalMuseum,Prague,CzechRepub- P.ramorum-infectedcoastliveoaksneartothecankerregioncom- lic, for insect identifications. We are also grateful to Andrew pared to sapwood located further from the canker region, as well Cline,CaliforniaDepartmentofFoodandAgriculture,Sacramento, ascomparedtosapwoodfromtreesnotinfectedwithP.ramorum. forassistancewithretrievalofbuprestidliteratureandproviding Ethanolconcentrationsinthesapwoodoffire-injuredponder- access to the California State Collection of Arthropods. Funding osa pines,Pinus ponderosa Douglas ex. C. Lawson,have also been was provided by USDA ForestService Pacific Southwest Research found to be muchgreater than in similar sapwoodsamples from Station, USDA Forest Service State and Private Forestry, the Na- unburned ponderosa pines, suggesting that scolytids may be at- tionalScienceFoundation(EF-0622770),andtheGordonandBetty tractedtoburnedconifersasaresultofthedetectionofthisetha- Moore Foundation. Partial funding for this project came from a nol (Kelsey and Joseph, 2003). As trees recover from fire injury, grant from the Jastro Shields Graduate Student Research Fellow- ethanolconcentrationshavebeenshowntodecline(KelseyandJo- ship,UniversityofCalifornia,Davis. seph, 2003), and this may be another reason that landing rates AppendixA werelowerinthespring2010trappingseasonthanthefall2009 trapping season. However, ethanol concentration are greatest in SeeTableA1. severely scorched and fire-killed trees (Kelsey and Joseph, 2003), andscolytidshavebeenknowntocolonizetreesforaslongasfour AppendixB years post-burn (Jenkins et al., 2008). Additionally, scolytids— including Gnathotrichus spp. and other ambrosia beetles—produce SeeTableB1.

Description:
this study were trapped on tanoaks in forest plots disturbed by P. ramorum and/or ramorum S. Werres & A.W.A.M. de Cock, a non-native, invasive.
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.