Stream temperature data collection standards for Alaska: Minimum standards to generate data useful for regional-scale analyses

(1)

ContentslistsavailableatScienceDirect

Journal

of

Hydrology:

Regional

Studies

jou rn a l h om ep a ge :w w w . e l s e v i e r . c o m / l o c a t e / e j r h

Stream

temperature

data

collection

standards

for

Alaska:

Minimum

standards

to

generate

data

useful

for

regional-scale

analyses

Sue

Mauger

a

_,

_Rebecca

_Shaftel

b,∗

_,

_E.

_Jamie

_Trammell

b,c

_,

_Marcus

_Geist

b

_,

_Dan

Bogan

b

a_Cook_Inletkeeper,₃₇₃₄_Ben_Walters_Lane,_Homer,_AK_99603,_United_States

b_Alaska_Natural_Heritage_Program,_University_of_Alaska_Anchorage,₃₂₁₁_Providence_Dr.,_Anchorage,_AK_99508,_United_States c_Geography_and_{Environmental}_Studies_Department,_University_of_Alaska_Anchorage,₃₂₁₁_Providence_Dr.,_Anchorage,_AK_99508,

UnitedStates

a

r

t

i

c

l

e

i

n

f

o

Articlehistory: Received10April2015

Receivedinrevisedform18July2015 Accepted23July2015

Availableonline24August2015 Keywords:

Streamtemperature Datacollection Minimumstandards

a

b

s

t

r

a

c

t

Studyfocus:Statewideinterestinthermalpatternsandincreasingdatacollectionefforts providesAlaska’sscientificandresourcemanagementcommunitiesanopportunitytomeet broaderregional-scaledataneeds.Abasicsetofstreamtemperaturemonitoringstandards isneededforAlaskanstobeginbuildingrobustdatasetssuitableforregionalanalyses. Thegoalofthisprojectistodefineminimum(base)standardsforcollectingfreshwater temperaturedatainAlaskathatmustbemetsothatobservationscansupportregional assessmentofstatusandrecenttrendsinfreshwatertemperaturesandpredictionsoffuture patternsofchangeintheseaquaticthermalregimesusingdownscaledclimateprojections. Newhydrologicalinsightsfortheregion:Wedefined10minimumdatacollectionstandards forcontinuousstreamtemperaturedatainAlaska.Thestandardscoverdataloggeraccuracy andrange,datacollectionsamplingfrequencyandduration,siteselection,loggeraccuracy checks,dataevaluation,fileformats,metadata,anddatasharing.Wehopethattheadoption ofminimumstandardswillencouragerapid,butstructured,growthincomparablestream temperaturemonitoringeffortsinAlaskathatwillbeusedtounderstandcurrentandfuture trendsinthermalregimes.

1. Introduction

TheavailabilityofstreamtemperaturedatainthecontiguousU.S.hasenabledrapidadvancesinourunderstandingof streamtemperaturedrivers,trends,andfutureprojections.Analysisofhistoricstreamtemperaturetrendsinthe West-ernU.S.indicatethatsomeaspectsofthethermalregimearecoherentacrossregionalscales,suchasincreasingsummer temperatures(Isaaketal.,2011),whileotheraspectsofthethermalregimearerespondingincomplexways,suchasdaily minimumsadvancingmorerapidlythanmaximums,butnotforallstreams,andnoconsistentchangestostream temper-aturevariability(Arismendietal.,2012,2013).Projectedincreasesintheannualmaximumweeklywatertemperaturesby

∗ Correspondingauthor.

E-mail addresses: [email protected] (S. Mauger), [email protected] (R. Shaftel), [email protected] (E.J. Trammell),

[email protected](M.Geist),[email protected](D.Bogan).

http://dx.doi.org/10.1016/j.ejrh.2015.07.008

(2)

2080areontheorderof2–5◦CforWashingtonState(Mantuaetal.,2010).Futureprojectionsofincreasingstream tem-peraturesacrossregionalrivernetworksindicatedecreasesinsuitablehabitatandfragmentationofexistinghabitatfor salmonidsintheWesternU.S.(Riemanetal.,2007;Isaaketal.,2010;Wengeretal.,2011;Rueschetal.,2012;Jonesetal., 2013).Theidentificationofimportantdriversofstreamtemperaturesallowsfortargetedmanagementstrategiesthatcan increaseresiliencyinaquaticecosystems,suchasimprovingriparianvegetationtoshadestreams,restoringstreamflows insummertimetodecreasestreamsensitivity,restoringfishpassagetoprovideaccesstothermalrefugia,andidentifying sensitiveareasforconservation(RiemanandIsaak,2010;Isaaketal.,2010).

InAlaska,climateischangingmorerapidlythaninthecontiguousU.S.;annualairtemperatureshaveincreasedinAlaska by1.7◦C(3◦F)overthelast60yearswhilewintertemperatureshaveincreasedby3.3◦C(6◦F,Chapinetal.,2014).Inaddition, datesofsnowmeltandfreeze-uphaveshiftedsothatthegrowingseasonisnow45%longerinInteriorAlaskathanitwas atthebeginningofthe20thcentury(Chapinetal.,2014).AsAlaskanscontinuetofeeltheimpactsofachangingclimate, theneedforresourcemanagerstounderstandhowthesechangeswillalteraquaticsystemsandfisheriesresourcesgrows. Streamtemperaturedatacollectioneffortshaveincreasedinrecentyearstobegintofillourgapsinknowledgeaboutcurrent thermalprofiles.Severalregionalanalyseshavebeenconductedinanefforttodifferentiatethewatershedcharacteristics drivingdifferencesinsummertimestreamtemperaturesacrossstreamandriversystems;importantfactorshaveincluded glaciercover(KyleandBrabets,2001;Fellmanetal.,2014),elevation(Mauger,2013;Lisietal.,2013),wetlands(Mauger, 2013),andlakes(Lisietal.,2013).DuetothelimitedspatialandtemporalcoverageofstreamtemperaturedatainAlaska, thereisalackofinformation describinghistorictrendsorgenerationoffutureprojections,especiallyastheyrelateto salmonids.

ArecentefforttocataloghistoricandexistingstreamtemperaturedataacrossAlaskafoundmorethan150continuous streamtemperaturesensorsdeployedacrossthestatemaintainedbyover15agencies.Theseagenciesarelikelyusingone ofthemanyexistingstreamtemperaturedataprotocolsspeciﬁctoAlaska,suchastheNationalParkService(Shearerand Moore2011Sergeantetal.,2013),CookInletkeeper(Mauger,2008),andtheUSGSincooperationwithU.S.FishandWildlife Service(Tooheyetal.,2014).Whiletheseprotocolsprovideexcellentguidanceregardingtemperaturemonitoring,they areoftenfocusedonspeciﬁcagencyproceduresandgoalsthatarenotapplicablebeyondtheirsourceentity.Noneofthe aforementionedprotocolsdirectthereadertowardclear,minimumstandardsregardingsamplefrequency,sampleduration, ordatamanagement.AbasicsetofstreamtemperaturemonitoringstandardsisstillneededforAlaskanstobeginbuilding robustdatasetssuitableforregionalanalyses.

Thegoalofthisprojectistodeﬁneminimum(base)standardsforcollectingfreshwatertemperaturedatainAlaskathat mustbemetsothatobservationscansupportregionalassessmentofstatusandrecenttrendsinfreshwatertemperatures andpredictionoffuturepatternsofchangeintheseaquaticthermalregimesusingdownscaledclimateprojections.By identifyingminimumdatastandards,ourobjectiveistoencouragerapid,butstructured,growthincomparablestream temperaturemonitoringeffortsinAlaskathatwillbeusedtounderstandcurrentandfuturetrendsinthermalregimes. Thesetrendscantheninformstrategiesformaintainingecosystemresilience.

2. Methods

Weidentifiedasequenceofstepsessentialtoanystreamtemperaturedatacollectionprojectandwithinthesesteps, identifiedcomponentswhereminimumstandardsshouldbeestablishedtoensurethatdatacouldbeusedinregional-scale analyses.Thestepsincludeselectionofadatalogger,datacollection,dataqualityassuranceandqualitycontrol,anddata storage.Weusedacombinationofempiricalevidence,publishedresearch,andexpertopinioninordertodefineeachof theminimumstandards.Foreachminimumstandard,wehavedescribedthemethodologyalongwithajustificationforthe finalstandard.

3. Results

Wedeﬁnedtenminimumdatacollectionstandardstogeneratedatausefulforregional-scaleanalysesofstreamthermal regimes.Thestandardscoverdataloggeraccuracyandrange;samplingfrequencyandduration;dataqualityassurancesteps includingaccuracychecks,siteselectionanddataevaluation;andﬁnally,metadata,datastorageandsharing(Table1).In somecaseswehaveincludedrecommendationsbeyondtheminimumstandardsforthereadertoconsider.Guidanceon howtoimplementthesestandardsisprovidedinaseparatereport:StreamTemperatureDataCollectionStandardsand ProtocolforAlaska(Maugeretal.,2014).

3.1. Datalogger

Therearetwominimumstandardsfordataloggers:accuracyof±0.25◦_C_and_range_from₋₄◦_to₃₇◦_C._The_accuracy andrangeminimumstandardsarebasedonthebestavailabletechnologyforwatertemperaturedataloggerscurrentlyon themarket.Wesettheminimumaccuracystandardat±0.25◦_C_as_opposed_to_0.2◦_C_to_be_clear_that_commonly_used_data loggerswithaccuracyspeciﬁcationsof0.21◦Careappropriate.Examplesofdataloggerscurrentlyavailablethatmeetthese speciﬁcationsincludeTidbiTv2,HOBOWaterTempProv2(OnsetComputerCorporation),LeveloggerEdge(SolinstCanada Ltd.)andYSI6920V2sonde(YSIIncorporated).Thereareadditionalbrandswithlessaccuracythatshouldnotbeused(e.g.,

(3)

Table1

Minimumdatacollectionstandardsforregionalanalysisofstreamthermalregimes.

Minimumstandards

Datalogger Accuracy ±0.25◦_C

Measurementrange −4◦_to₃₇◦_C₍₂₄◦_–99◦_F)

Datacollection Samplingfrequency 1hinterval Samplingperiod/duration 1calendarmonth

Qualityassuranceandqualitycontrol Accuracychecks Waterbathattwotemperatures:0◦Cand20◦Cbeforeandafterﬁeld deploymenttoverifyloggeraccuracy(varies≤0.25◦_C_compared_with_a

NIST-certiﬁedthermometer)

Siteselection Fivemeasurementsacrossthestreamwidthtoverifythatthesiteis well-mixed(i.e.,varies≤0.25◦_C)

Dataevaluation Removeerroneousdatafromthedataset Datastorage Fileformats CSVformatin2locations

Metadata Uniquesiteidentiﬁer,agency/organizationnameandcontact,datum, latitudeandlongitude,andsamplefrequency;storedwithtemperature data

Sharing Quality-controlledhourlydata

Daily Range

Dif

ference from Maxi

m um Daily T em p. 0.0 0.1 0.2 0.3 0.4 4 4.2 5.7 7.1 8.3 9 10.3 30 minutes 1 hour 2 hours 4 hours

Fig.1.Differenceindailymaximumstreamtemperaturesfor30min,1h,2h,and4hsamplingintervalsbasedonsevenstreammonitoringsitesinCook Inlet.Thehorizontallineindicatestheminimumstandardforloggeraccuracy,0.25◦_C.

iButtons).Introductionofadditionalmeasurementerrorintostreamtemperaturedatasetscanreduceourabilitytodetect trends.TherangeissetwellbeyondtheexpectedvaluesforstreamtemperatureinAlaska.

3.2. Datacollection 3.2.1. Samplingfrequency

Theminimumstandardforsamplingfrequencyisa1hinterval becauseitisthemaximumintervalthateffectively capturesthedailymaximumandminimumtemperatures.Theprobabilityofcapturingthedailymaximumorminimum givenaspeciﬁedsamplingintervalisaffectedbythedailyrangeinstreamtemperature.Dunhametal.(2005)compared severalsamplingintervalstotheirbaselineof30mintoestimatetheprobabilitiesofmissingthemaximumdailytemperature bymorethan1◦C.Givenadailyrangeof12◦C,thereislessthana2%probabilityofmissingthetruedailymaximumby morethan1◦Cusingatwo-hoursamplinginterval(Fig.5,Dunhametal.,2005).Theseresultsarerelevantforadatasetof48 non-glacialsalmonstreamsinCookInletwherethedailyrangeamongsitesvariedfrom3.9◦Cto11.6◦C(Table6,Mauger, 2013).However,a1◦Caccuracygoalmaynotbesensitiveenoughfortrackingmaximumandminimumtemperaturetrends duringspeciﬁcseasonsimportantforaquaticorganisms.

Weresampledstreamtemperaturedatacollectedat15minintervalsforsevenCookInletstreamswhosedailyranges variedfrom4.0◦to10.3◦C.Foreachofthesamplingintervalsstudied–30min,1h,2h,and4h–wecalculatedthedifference indailymaximumfromthe15minintervaldatasettodeterminethelossinaccuracyfromrecordingtemperaturesatlonger timeintervals(Fig.1).Errorbarsreﬂectthestandarderrorofthemeandifferencesbasedon76–149daysofdatawithinone yearforeachsite.A4hsamplingintervalresultsinareductiontothemaximumdailytemperatureof0.3◦Catthesitewith thelargestdailyrange.Thisintroducedbiasisgreaterthantheaccuracyofthedataloggersused.The2h,1h,and30min

(4)

Table2

FrequenciesofMWATandMWMTbymonthforstreamsinCookInlet,2008–2012.

June July August September

Maximumweeklyaveragetemperature(MWAT)

2008 1 17 22 0

2009 0 42 0 0

2010 1 25 11 1

2011 2 38 2 0

2012 17 1 10 0

Maximumweeklymaximumtemperature(MWMT)

2008 3 25 12 0

2009 0 42 0 0

2010 2 30 4 2

2011 3 38 1 0

2012 17 3 8 0

samplingintervalsresultinaminorlossofaccuracyinmeasuringthedailymaximum.Wechosea1hminimumstandard forthesamplingintervaltoreducethepossibilityofintroducingbiasintothedailymaximumandminimumvalues,which getcompoundedwhencalculatingmaximumweeklyvalues.A1hintervalalsominimizeswaitingtimewhenperforming synchronizeddataqualityassurancechecksinthefield.Whenconsideringinclusionofhistoricaldataforaregionalanalysis, a2hsamplingintervalmaybesufficientforcalculatingdailymeananddailymaximumvalues.But,historicaldatashould alsobecarefullyreviewedtoensurethatitmeetsalloftheminimumstandards(e.g.,thedatalogger’saccuracycanbe confirmedandallqualityassuranceandqualitycontrolstepsweretaken).

3.2.2. Samplingduration

Theminimumstandardforsamplingdurationisonecalendarmonth.However,werecommendyearrounddata collec-tionorasmuchoftheopenwaterseasonaspossible,andatleastthreeyearsofdatacollection.Wedecidedonaonemonth minimumstandardafterreviewingexistingregionalanalysesfortheshortestdurationofdatacollectionusefulfor under-standingstatusandtrendsinstreamthermalregimes.Severalregionalanalysesdevelopedstatisticalmodelsforpredicting monthlyaveragetemperatures(e.g.,Wehrlyetal.,2009;Hrachowitzetal.,2010;Mayer2012;Fellmanetal.,2014).The monthsmostcommonlymodeledwereJulyandAugust,butsomestudiesalsomodeledothermonthsoftheyear.

WerecommendyearrounddatacollectionbecausewinterairtemperaturesareincreasingfasterinAlaskathansummer temperatures(Chapinetal.,2014),andthesemayhaveimportantimplicationsonthedevelopmentandtimingoflifehistory eventsforsalmon(Bryant,2009).Adultsalmonmigrationtofreshwaterandsmoltmigrationtotheoceanarebothcloselytied tostreamtemperature.Adultsreturnearliertospawnincolderstreamsbecauseoftherequiredaccumulatedtemperature neededforeggincubation(Kovachetal.,2012;Lisietal.,2013).Smoltsleavefortheoceanearlierinwarmerstreamsbecause growthismorerapidandsizeisanimportantcueformigration(Richetal.,2009).Formultiplespeciesandlifehistoriesof salmoninawarmingstreaminSoutheastAlaska,bothadultandsmoltmigrationshaveadvancedoverthelast30–50years (Kovachetal.,2013).FuturepredictionsofcohoproductioninresponsetostreamtemperaturesanddischargeinWestern CookInletshowthattheresponsesvarydependingupontheinteractionofchangingtemperatureanddischarge(Leppietal., 2014).Thecomplexresponseofsalmontoclimatechangehighlightstheneedformonitoringstreamtemperaturesduring allseasonsoftheyear.

Themajorityofregionalanalysesevaluatingclimatechangeeffectsonﬁshdistributionshavemodeledoneormore measuresofthethermalmaxima(e.g.,MWAT,Eatonetal.,1995;andMWMT,Isaaketal.,2010).Inordertoprovidea recommendationforthedeploymentperiodneededtocapturethethermalmaximainSouthcentralAlaska,wereviewed ﬁvesummersofstreamtemperaturedatacollectedintheCookInletbasin.Weusedthedatesofmaximumweeklyaverage temperature(MWAT)andmaximumweeklymaximumtemperature(MWMT)toevaluatethetimingofthethermalmaxima (Table2).TheMWMToccursmorefrequentlyinJulythanMWAT,butwarmeventsinJuneandAugustindicatetheimportance ofmeasuringstreamtemperaturesforallthreesummermonths.FortwositesinsouthernCookInlet,theMWMTwas observedaslateasSeptemberin2010.WerecommendaminimumdeploymentperiodofJune1–August31tocapture thethermalmaximaforstreamsinSouthcentralAlaska.Thetimingofthermalmaximamaybedifferentinotherregions ofAlaska.Ifthetimingisnotknown,multipleyearsofdatashouldbecollectedovertheentireopenwaterperiodbefore narrowingyoursamplingperiodtotargetthethermalmaxima.Inaddition,climatechangemaybeshiftingthethermal maximaearlierinthesummerduetodecreasingsnowpackandincreasingtemperatures.

Itisalsoimportanttoconsiderinter-annualvariabilityinstreamtemperatureregimeswhenplanningstreamtemperature datacollectionefforts.ValuesforMWATandMWMTwerehighlyvariableoveraﬁveyearmonitoringperiodinCookInlet salmonstreams.Forstreamswithatleastthreesummersofdata(n=44),thedifferencebetweenthelowestandhighest MWATrangedfrom0.8◦ to6.4◦CandforMWMTrangedfrom1.4◦to7.3◦C.Werecommendatleastthreeyearsofdata collectioninordertoaccuratelycapturetheeffectofinter-annualvariabilityonastream’sthermalregime.Ifyouareunableto collectdatayearround,itisimportanttoconsidersamplingthesamemonth(orsetofmonths)yearafteryearforconsistency. Adataloggercanbeusedtorecordmeasurementsforseveralyearsasthebatterylifeforatypicalloggeris5yearsata1min

(5)

orgreaterinterval.But,duetolimitationsinstoragecapacityandrecommendedstepsforqualityassurance,loggersshould beretrievedannuallysothataccuracycheckscanbeperformedanddatacanbedownloadedbeforeredeploying.

3.3. Qualityassuranceandqualitycontrol 3.3.1. Accuracychecks

Theminimumstandardtoensureloggeraccuracyincludeswaterbathaccuracychecksattwotemperatures:0◦Cand 20◦C,beforeandafterfielddeploymentusingaNIST(NationalInstituteofStandardsandTechnology)traceable(calibrated andmaintained)thermometeraccurateto±0.25◦_C._{NIST-certified}_thermometers_can_be_{liquid-in-glass}_thermometers_or theycanbeadataloggerwhichhasbeenNIST-certified.Werecommenda4-point(0,10,20,30◦C)calibrationbythe manufac-turer.ItisgoodpracticetosendtheNIST-certifiedthermometerbacktothemanufacturerforre-calibrationeverytwoyears. Loggersmustmeasurewatertemperatureswithin0.25◦CoftheNIST-certifiedthermometerforatleast3measurementsin eachbath.Recordthesevaluesinalogbookorspreadsheet.

Theseaccuracychecksareneededtoverifythateachloggermeetsitstechnicalspeciﬁcationsthroughoutthedeployment periodandthatmeasurementdrifthasnotoccurredovertime.Loggersthatfailpre-deploymentaccuracychecksarenot used,whileloggersthatfailpost-deploymentaccuracychecksmayresultinafailuretomeettheminimumstandardsfor regionalanalysis.Werecommendthatloggersgothroughanaccuracycheckatleastonceayear.Whencollectingdatafor multipleyearsatasite,loggersshouldbeswappedoutonceayearifpossible.Accuracychecksensurethatdatacanbe conﬁdentlysharedwithotherusers.Wedonotrecommendusingthemtocalibrateormodifyvaluesrecordedbythedata logger.

3.3.2. Siteselection

Siteselectionatthereachscaleincludestwocomponents:identifyingastablelocationwithinthereachanddeploying theloggerinawell-mixedsectionofthestreamchannel.DuetothediversityinstreamandriverecosystemswithinAlaska, wecanonlyprovidegeneralguidanceforsiteselectionwithinthereach.Highvelocityhabitats,suchasthosefoundalong theoutsidebankofabend,shouldbeavoidedtoreducethelikelihoodoflosingaloggerduringhighﬂowevents.Lowvelocity habitats,likethosealongtheinsideofabendorineddiesorpools,shouldbeavoidedbecausesedimentdepositionmaybury alogger.Loggersshouldbedeployedwithintheactivechanneltopreventexposuretoairtemperaturesduringlowﬂows.

Iftheloggerdoesnotcomewithaprotectivecase(e.g.,TidbiTv2TempLogger),itshouldbeplacedinaﬂow-through housingtoprotecttheequipmentfromnatural,wildlife,orhumandisturbance.Housingsalsoprovideshadeforthelogger, protecttheloggerfrommovingdebris,andallowforsecureattachmentwithacable.Housingsshouldallowforgoodwater circulationpasttheloggerbutnotbeindirectcontactwiththetemperaturesensorbecausethehousingmayabsorbheat.If possible,loggerswithaprotectivecaseorwithinahousingshouldbeplacedinawellshadedlocationtoreducetheinﬂuence ofdirectsolarradiation.

Thechoicesmadetosecurealoggeratthesitewillhavethegreatestinfluenceonthesuccessfulcollectionofstream temperaturedata.Whenselectingadeploymentmethod,considerationshouldbegiventohowitwillworkathighand lowflows,howmuchstreambedmovementthereisatthesite,andhowtopreventpeoplefromaccidentallydislodging equipmentathightrafficsites(e.g.,trippingoverrebar,sandbagsorcables).Generally,attachingaloggertoabank-secured cableispreferredforstreamswithsoftsubstratesorsignificantmovementofthestreambedduringhighflows.Rebarcan beinsertedintothestreambankintreelessareas.Rebarorduckbillearthanchorssunkintothestreambedarepreferred ifthereisonlymoderatestreambedmovementandthestreamisshallow.Sandbagscanbeusedtoweighdownalogger, butthismethodisonlyrecommendedinstreamswithminimalstreambedmovement.Ifasitehaslargerocksorabridge support,underwaterepoxyisanotheralternative(Isaaketal.,2013).

Icemovementmightdestabilizeananchoringmethodifaloggerisleftinstreamoverthewinter.Anchoringthelogger tothestreambedasopposedtothestreambankshouldpreventlossduetoicemovementinthespring.Werecommend deployingtwologgersatasitetoprovidebackupintheeventthatoneislost.Whenasiteisﬁrstestablished,andespecially whendeployingloggersforwinterdatacollection,multiplebackuploggersusingdifferentanchoringmethodsisthebest waytoguaranteeloggersarerecoveredthefollowingyear.

Theminimumstandardforsiteselectionincludesﬁvemeasurementsofstreamtemperatureacrossthewidthofthe streamtoensurethattheloggerisdeployedinalocationthatvaries≤0.25◦_C._Temperature_loggers_should_be_placed_in_a well-mixedsectionofthemainstreamchannelifthedataaretobeusefulforregional-scaleanalysisofstreamtemperatures. Streamthermalregimescanbehighlyvariableatthereachscaledependinguponthediversityofhabitattypespresent. ThermalimagingoftheAnchorRiverinSouthcentralAlaskaindicatesthatsloughsandsidechannelsmaybewarmeror colderthanthemainchannelbyasmuchas4◦C(Table5,WatershedSciences,2010).Streamreachfeatureswithunique temperaturecharacteristics,suchasoff-channelhabitat,groundwaterupwellingareas,oranthropogenicfeatures(e.g.,dam orpointdischarge),shouldbeavoidediftheyarenotpartofthestudyobjectives.

Siteselectionalsoincludesthelocationofamonitoringsitewithinthestreamnetwork,whichistypicallyrelatedto projectobjectivesandmaynotbebasedonregionalanalysisofstreamtemperatures.Probabilisticdesigns,suchasthose usedforEPA’sNationalAquaticResourceSurveys(StevensandOlsen,2004),canbeusedtolocaterandomsamplingsites thatarespatiallybalancedacrossastreamnetwork,buttheyarelogisticallychallengingtoapplyinremotelocations.In ordertocapturetherangeofthermalregimesandinformpredictivemodelsofstreamtemperaturesacrossanetwork,sites

(6)

maybestrategicallyplacedtocapturethefullrangeofthedominantgeomorphicconditionsdrivingstreamtemperatures (Isaaketal.,2010;Lisietal.,2015).Dependingupontheregion,importantgeomorphicfactorstoconsiderincludeelevation, slope,streamsize,andwetlandandlakecoverage.SpatialdatafortheregioncanbeassembledinaGISandusedtoattribute thestreamnetworkwiththenecessarystreamorwatershedinformationrequiredforsiteselection.

Othersuggestionsforsiteselectionincludeutilizingconfluencesandtargetinguniquefeaturesinastreamnetwork. Confluencesprovideanopportunitytogatherinformationaboutthreedistinctstreamreachesbydeployingloggersinthe twoincomingtributariesandalsowithinthedownstreamreachbelowwherethetwosourcewatershavebecome well-mixed.Discretefeaturesinaregionthatmayaffectstreamtemperatures,suchasalargelakeorwildfire,canbebracketedto bettercapturetheireffect.Recentguidanceonsamplingdesignsforstreamnetworksrecommendsplacingmultiplesamples inclustersatconfluencesandalsosinglesamplesatoutletandheadwaterreaches(Sometal.,2014).

3.3.3. Dataevaluation

Theminimumstandardfordataevaluationisthatallerroneousdataareremovedfromthedataset.Dataevaluationsteps canonlybeperformedwithconfidencebythefieldstafffamiliarwiththesamplingeventsandsiteconditionsandshould occurimmediatelyafterreturningfromthefieldtopreventanylossofinformationsharingneededtodiagnoseerroneous data.Dataevaluationstepsincluderemovingairtemperaturemeasurementsbeforedeploymentandafterretrievaland screeningforanomalousreadingscausedbydewateringorburialofthelogger.Thereareseveralpublicationsthatprovide examplesofvisualchecksforanomalousdata:Maugeretal.(2014),SowderandSteel(2012),andTooheyetal.(2014). 3.4. Datastorage

Datamanagementandsharingstandardsarealsoincludedtoensuredatacollectedforregionalanalysisaremadeavailable inaneasilyexchangeableformat.

3.4.1. Fileformat

Theminimumstandardforfileformatisacomma-separatedvalue(csv)filestoredintwolocations.Wespecifieda softwareneutralfileformatsothatitiseasilyimportedintoavarietyofdatabaseandanalysisprograms,suchasExcel, Access,andR.Additionally,dataandassociatedmetadataneedtobestoredinatleasttwolocations,withoneofthose locationsbeingpubliclyaccessible.

3.4.2. Metadata

Regionalscaleassessmentsofstreamtemperatureswillrequirescientiststousedatafromnumeroussensorssourced frommanyagencies.Theminimumstandardrequiresthatmetadatainformationbestoredwiththetemperaturedatafiles sothatfutureuserscaneasilyusethedata.Thecreation,maintenance,anddistributionofmetadataarecritical.Asthe numberoftemperaturemonitoringdatasetsincreasesrapidly,ourabilitytodiscernwhichdatasetsareusefultoagiven researchinterestwillberelatedtoourcapacitytosortthroughmetadatawhichhavecommonfields.Usingconsistentfields andformatswillimprovecomparisonsbetweendatasetscollectedbydifferentgroupsandatdifferenttimes.Ataminimum, metadatashallincludethefollowingattributes:uniquesiteidentifier,datasourceagencyororganizationnameandcontact information,datum,latitude,longitude,andsamplefrequency(1h,30min,15min).Westronglyencourageinvestigators workinginAlaskatosubmitprojectmetadatatoAK-OATS(http://aknhp.uaa.alaska.edu/aquatic-ecology/akoats/).

3.4.3. Datasharing

Theminimumstandardforsharingdataisquality-controlledhourlydata,whichprovidestheinformationneededto characterizekeyaspectsofastream’sthermalregime(Dunhametal.,2005;Nelitzetal.,2007;Arismendietal.,2013). Althoughmanyregionalanalyseshavefocusedonstreamtemperatureresponsesassociatedwiththesummertimethermal maxima(e.g.,meanJulytemperatureorMWMT),therearemanyothercomponentstothestreamthermalregime:magnitude (e.g.,minimums),variability(e.g.,dailyrange),frequency(e.g.,numberofdaysthatexceedathreshold),duration(e.g., numberofcontiguousdaysaboveathreshold),andtiming(e.g.,dayofyear,Pooleetal.,2001).Wealsorecommendproviding dailysummariesofminimum,maximum,andmeanstreamtemperatures.Calculatingthesedailysummarystatisticsserves asanimportantqualityassurancestepbyforcingthedatacollectortoreviewthedatasoonafterdataretrievalsothat erroneousmeasurementscanbeidentiﬁedanddeleted.Dailysummary statisticsshouldonlybecalculated forquality controlleddatawithatleast90%ofdailymeasurements(e.g.,22hourlymeasurements).

4. Conclusions

ManyentitiesarecollectingstreamtemperaturedatainAlaskaforavarietyofpurposestomeetprojectoragencyspecific goals.StatewideinterestinthermalpatternsandincreasingdatacollectioneffortsprovidesAlaska’sscientificandresource managementcommunitiesanopportunitytoaddressbroaderregional-scaledataneeds.Wehaveendeavoredtoidentify minimumstandardsforstreamtemperaturedatacollectionthatwillresultindatasetsusefulforansweringmostresearch andmonitoringquestionsaskedattheregionalscale.Wehopethatinvestigatorswillconsidertheseminimumstandards whendevelopingafieldplan,astheywillreducethevariabilityofdataqualityduetodisparatesamplingmethods.Thiswill

(7)

enableresearcherstoeasilyevaluateaproject’smetadataanddeterminetheutilityofthedataforassessingpatternsand trendsinAlaska’sfreshwatersystems.Thesestandardsmayalsoprovideausefulstartingpointforcollaborativeeffortsto combinestreamtemperaturedatacollectedbymultipleentitiesacrossregionalscalesintheLower48.

Wehaveprovidedadditionalrecommendationsbeyondtheminimumstandardsasguidanceforentitieswhoseprimary objectiveistounderstandstreamthermalpatternsandalsoforthosewhohaveaninterestinmakingtheirdataasbroadly usefulaspossible.Mostnotablywerecommendatleastthreeyearsofyear-rounddatacollectionanddeployingtwodata loggersatasiteintheeventoneloggerfailsorislost.TipsareofferedtoaddressAlaska’suniquelychallengingconditions includingicemovement,highflowevents,treelessareas,andremoteaccess,whichallneedtobeconsideredwhen estab-lishingasamplingsite.Wehopeestablishingasetofstandardsandprovidingrecommendationswillencourageadditional groupstodeploytemperaturesensors,andparticularlybenefitfieldstaffwhoseprimarytasksmaynotbehydrologyor monitoringaswellaspersonnelatsmallerorganizations.

Insomecases,investigatorsmaychoosemorerigorousqualityassurancemethodsorshortersamplingintervals. For-tunately,thesedecisionswillnotprecludetheusefulnessofthesedataforregionalanalysisastheyareaboveandbeyond theminimumstandards.Werealizethatsomeproject-speciﬁcneeds,particularlyrelatedtosamplinglocation,maynotbe compatiblewiththesestandardsandwillnotresultinusefuldataataregionalscale.Nevertheless,inAlaska,wheretravel costscaneatupﬁeldbudgetsquickly,voluntaryadoptionofminimumstandardswillgoalongwaytohelpstretchlimited researchdollarsand,mostimportantly,togeneratevaluabledatasetsforunderstandingthermalpatternsacrossAlaska’s vastfreshwaterecosystems.

Acknowledgements

WewouldliketothankGretaBurkart,LauraEldred,JeffFalke,SteveFrenzel,AlanPeck,ChrisSergeant,BrockTabor,Ryan Toohey,andJohnTrawickifortheirguidanceandserviceontheTechnicalAdvisoryGroup;andJoeKlein,MegPerdue,and twoanonymousreviewersforprovidinghelpfulcommentsonanearlierdraft.WeareespeciallygratefultoJoelReynolds andKarenMurphyforassistanceandcoordinationthroughouttheprojectandfortheirguidanceinthepreparationof thisdocument.SupportforthiseffortwasprovidedbytheU.S.FishandWildlifeServiceonbehalfoftheWesternAlaska LandscapeConservationCooperative(F14AC00103).

AppendixA. Supplementarydata

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ejrh.2015.07.008.

References

Arismendi,I.,Johnson,S.,Dunham,J.,2012.Theparadoxofcoolingstreamsinawarmingworld:regionalclimatetrendsdonotparallelvariablelocal trendsinstreamtemperatureinthePaciﬁccontinentalUnitedStates.Geophys.Res.Lett.39,L10401.

Arismendi,I.,Johnson,S.L.,Dunham,J.B.,Haggerty,R.,2013.Descriptorsofnaturalthermalregimesinstreamsandtheirresponsivenesstochangeinthe PaciﬁcNorthwestofNorthAmerica.Freshw.Biol.58,880–894.

Bryant,M.D.,2009.GlobalclimatechangeandpotentialeffectsonPaciﬁcsalmonidsinfreshwaterecosystemsofsoutheastAlaska.Clim.Change95, 169–193.

Chapin,F.S.I.,Trainor,S.F.,Cochran,P.,Huntington,H.,Markon,C.,McCammon,M.,McGuire,A.D.,Serreze,M.,2014.ClimateChangeImpactsintheUnited States:TheThirdNationalClimateAssessment.In:Melillo,J.M.,Richmond,T.C.,Yohe,G.W.(Eds.).U.S.GlobalChangeResearchProgram,Alaska,pp. 514–536(Chapter22).

Dunham,J.,Chandler,G.,Rieman,B.,Martin,D.,2005.MeasuringStreamTemperaturewithDigitalDataLoggers:AUser’sGuide.U.S.Departmentof Agriculture,ForestService,RockyMountainResearchStation,FortCollins,CO.

Eaton,J.,McCormick,J.,Goodno,B.,O’Brien,D.,Stefan,H.,Hondzo,M.,Scheller,R.,1995.Aﬁeldinformation-basedsystemforestimatingﬁshtemperature tolerances.Fisheries20,10–18.

Fellman,J.B.,Nagorski,S.,Pyare,S.,Vermilyea,A.W.,Scott,D.,Hood,E.,2014.Streamtemperatureresponsetovariableglaciercoverageincoastal watershedsofSoutheastAlaska.Hydrol.Process.28,2062–2073.

Hrachowitz,M.,Soulsby,C.,Imholt,C.,Malcolm,I.a.,Tetzlaff,D.,2010.Thermalregimesinalargeuplandsalmonriver:asimplemodeltoidentifythe inﬂuenceoflandscapecontrolsandclimatechangeonmaximumtemperatures.Hydrol.Process.24,3374–3391.

Isaak,D.J.,Luce,C.H.,Rieman,B.E.,Nagel,D.E.,Peterson,E.E.,Horan,D.L.,Parkes,S.,Chandler,G.L.,2010.Effectsofclimatechangeandwildﬁreonstream temperaturesandsalmonidthermalhabitatinamountainrivernetwork.Ecol.Appl.20,1350–1371.

Isaak,D.J.,Wollrab,S.,Horan,D.,Chandler,G.L.,2009.ClimatechangeeffectsonstreamandrivertemperaturesacrossthenorthwestU.S.from1980to 2009andimplicationsforsalmonidﬁshes.Clim.Change113,499–524.

Isaak,D.J.,Horan,D.L.,Wollrab,S.P.,2013.Asimpleprotocolusingunderwaterepoxytoinstallannualtemperaturemonitoringsitesinriversandstreams. In:Gen.Tech.Rep.RMRS-GTR-314.U.S.DepartmentofAgriculture,ForestService,RockyMountainResearchStation,FortCollins,CO,21p.

Jones,L.,Muhlfeld,C.,Marshall,L.,Mcglynn,B.,Kershner,J.,2013.Estimatingthermalregimesofbulltroutandassessingthepotentialeffectsofclimate warmingoncriticalhabitats.RiverRes.Appl.30,204–216.

Kovach,R.P.,Gharrett,A.J.,Tallmon,D.A.,2012.Geneticchangeforearliermigrationtiminginapinksalmonpopulation.Proc.R.Soc.B:Biol.Sci.279, 3870–3878.

Kovach,R.P.,Joyce,J.E.,Echave,J.D.,Lindberg,M.S.,Tallmon,D.A.,2013.Earliermigrationtiming,decreasingphenotypicvariation,andbiocomplexityin multiplesalmonidspecies.PLoSONE8,e53807.

Kyle,R.E.,Brabets,T.P.,2001.WaterTemperatureofStreamsintheCookInletBasin,Alaska,andImplicationsofClimateChange.WRI01-4109,Anchorage, AK.

Leppi,J.C.,Rinella,D.J.,Wilson,R.R.,Loya,W.M.,2014.LinkingclimatechangeprojectionsforanAlaskanwatershedtofuturecohosalmonproduction. Glob.ChangeBiol.20,1808–1820.

(8)

Lisi,P.J.,Schindler,D.E.,Bentley,K.T.,Pess,G.R.,2013.Associationbetweengeomorphicattributesofwatersheds,watertemperature,andsalmonspawn timinginAlaskanstreams.Geomorphology185,78–86.

Lisi,P.J.,Schindler,D.E.,Cline,T.J.,Scheuerell,M.D.,Walsh,P.B.,2015.Watershedgeomorphologyandsnowmeltcontrolstreamthermalsensitivitytoair temperature.Geophys.Res.Lett.42,3380–3388.

Mantua,N.,Tohver,I.,Hamlet,A.,2010.Climatechangeimpactsonstreamﬂowextremesandsummertimestreamtemperatureandtheirpossible consequencesforfreshwatersalmonhabitatinWashingtonState.Clim.Change102,187–223.

Mauger,S.,2008.WaterTemperatureDataLoggerProtocolforCookInletSalmonStreams.CookInletkeeper,Homer,AK.

Mauger,S.,2013.StreamTemperatureMonitoringNetworkforCookInletSalmonStreams2008–2012.CookInletkeeper,Homer,AK.

Mauger,S.,Shaftel,R.,Trammell,E.J.,Geist,M.,Bogan,D.,2014.StreamTemperatureDataCollectionStandardsandProtocolforAlaska:Minimum StandardstoGenerateDataUsefulforRegional-ScaleAnalyes.HomerandAnchorage,AK.Availableat

<http://aknhp.uaa.alaska.edu/wp-content/uploads/2015/01/StreamTemperatureStandardsandProtocolfor-Alaska.pdf/>. Mayer,T.D.,2012.ControlsofsummerstreamtemperatureinthePaciﬁcNorthwest.J.Hydrol.475,323–335.

Nelitz,M.A.,MacIsaac,E.A.,Peterman,R.M.,2007.Ascience-basedapproachforidentifyingtemperature-sensitivestreamsforrainbowtrout.NorthAm.J. Fish.Manage.27,405–424.

Poole,G.C.,Risley,J.,Hicks,M.,2001.IssuePaper3SpatialandTemporalPatternsofStreamTemperature(Revised).

Rich,H.B.,Quinn,T.P.,Scheuerell,M.D.,Schindler,D.E.,2009.Climateandintraspeciﬁccompetitioncontrolthegrowthandlifehistoryofjuvenilesockeye salmon(Oncorhynchusnerka)inIliamnaLake,Alaska.Can.J.Fish.Aquat.Sci.66,238–246.

Rieman,B.E.,Isaak,D.,Adams,S.,Horan,D.,Nagel,D.,Luce,C.,Myers,D.,2007.Anticipatedclimatewarmingeffectsonbulltrouthabitatsandpopulations acrosstheinteriorColumbiaRiverBasin.Trans.Am.Fish.Soc.136,1552–1565.

Rieman,B.E.,Isaak,D.J.,2010.ClimateChange,AquaticEcosystems,andFishesintheRockyMountainWest:ImplicationsandAlternativesfor Management.U.S.DepartmentofAgriculture,ForestService,RockyMountainResearchStation,FortCollins,CO.

RueschA.S.,Torgersen,C.E.,Lawler,J.J.,Olden,J.D.,Peterson,E.E.,Volk,C.J.,Lawrence,D.J.,2012.Projectedclimate-inducedhabitatlossforsalmonidsin theJohnDayRivernetwork,ConservationBiology.Oregon,U.S.A.,26,873–882.

Sergeant,C.J.,Johnson,W.F.,Nagorski,S.,2013.FreshwaterWaterQualityMonitoringProtocol:VersionFQ-2013.1.NationalParkService,FortCollins,CO. Shearer,J.,Moore,C.,2011.SouthwestAlaskaFreshwaterFlowSystemMonitoringProtocolNarrative.SouthwestAlaskaNetwork,FortCollins,CO.

Som,N.A.,Monestiez,P.,VerHoef,J.M.,Zimmerman,D.L.,Peterson,E.E.,2014.Spatialsamplingonstreams:principlesforinferenceonaquaticnetworks. Environmetrics25,306–323.

Sowder,C.,Steel,E.A.,2012.Anoteonthecollectionandcleaningofwatertemperaturedata.Water4,597–606.

Stevens,D.L.,Olsen,A.R.,2004.Spatiallybalancedsamplingofnaturalresources.J.Am.Stat.Assoc.99,262–278.

Toohey,R.C.,Neal,E.G.,Solin,G.L.,2014.GuidelinesfortheCollectionofContinuousStreamWater-TemperatureDatainAlaska.Reston,Virginia. WatershedSciences,2010.AirborneThermalInfraredRemoteSensingAnchorRiverBasin,Alaska.GeologicalSurvey,Corvallis,OR.

Wehrly,K.E.,Brenden,T.O.,Wang,L.,2009.Acomparisonofstatisticalapproachesforpredictingstreamtemperaturesacrossheterogeneouslandscapes.J. Am.WaterResour.Assoc.45,986–997.

Wenger,S.J.,Isaak,D.J.,Luce,C.H.,Neville,H.M.,Fausch,K.D.,Dunham,J.B.,Dauwalter,D.C.,Young,M.K.,Elsner,M.M.,Rieman,B.E.,Hamlet,A.F.,Williams, J.E.,2011.Flowregime,temperature,andbioticinteractionsdrivedifferentialdeclinesoftroutspeciesunderclimatechange.Proc.Natl.Acad.Sci.U.S. A.108,14175–14180.