Citation
Characterizing Precipitation Variability and Trends in the World's Mediterranean-Climate Areas

Material Information

Title:
Characterizing Precipitation Variability and Trends in the World's Mediterranean-Climate Areas
Series Title:
Deitch, M.J., Sapundjieff, M.J. and Feirer, S.T., 2017. Characterizing precipitation variability and trends in the world’s Mediterranean-Climate areas. Water, 9(4), p.259.
Creator:
Deitch, Matthew
Publisher:
MDPI
Publication Date:
Language:
English
Physical Description:
Journal Article

Notes

Abstract:
The Mediterranean climate is principally characterized by warm, dry summers and cool, wet winters. However, there are large variations in precipitation dynamics in regions with this climate type. We examined the variability of precipitation within and among Mediterranean-climate areas, and classified the Mediterranean climate as wet, moderate, or dry based on annual precipitation; and strongly, moderately, or weakly seasonal based on percentage of precipitation during summer. Mediterranean biomes are mostly dry (<700 mm annually) but some areas are wet (>1300 mm annually); and many areas are weakly seasonal (>12% of annual precipitation during summer). We also used NOAA NCDC climate records to characterize interannual variability of annual and dry-season precipitation, as well as trends in annual, winter, and dry-season precipitation for 337 sites that met the data quality criteria from 1975 to 2015. Most significantly, sites in many Mediterranean-climate regions show downward trends in annual precipitation (southern California, Spain, Australia, Chile, and Northern Italy); and most of North America, the Mediterranean basin, and Chile showed downward trends in summer precipitation. Variations in annual and summer precipitation likely contribute to the high biodiversity and endemism characteristic of Mediterranean-climate biomes; the data indicate trends toward harsher conditions over the past 40 years.
Acquisition:
Collected for University of Florida's Institutional Repository by the UFIR Self-Submittal tool. Submitted by Matthew Deitch.
General Note:
This paper was published with support from the University of Florida's Open Access Publishing Fund.

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
All rights reserved by the submitter.

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

Article CharacterizingPrecipitationVariabilityandTrendsin theWorld'sMediterranean-ClimateAreas MatthewJ.Deitch 1, *,MicheleJ.Sapundjieff 2 andShaneT.Feirer 3 1 SoilandWaterSciencesDepartment,UniversityofFlorida, IFASWestFloridaResearchandEducationCenter,Milton,FL32583,USA 2 UniversityofFlorida,IFASWestFloridaResearchandEducationCenter,Milton,FL32583,USA; mgoodfellow@u.edu 3 UniversityofCaliforniaANRInformaticsandGISStatewideProgram,Hopland,CA95445,USA; stfeirer@ucanr.edu Correspondence:mdeitch@u.edu AcademicEditors:MattKondolfandBenPorter Received:1November2016;Accepted:29March2017;Published:6April2017 Abstract:TheMediterraneanclimateisprincipallycharacterizedbywarm,drysummersandcool,wetwinters.However,therearelargevariationsinprecipitationdynamicsinregionswiththisclimatetype.WeexaminedthevariabilityofprecipitationwithinandamongMediterranean-climateareas,andclassiedtheMediterraneanclimateaswet,moderate,ordrybasedonannualprecipitation;andstrongly,moderately,orweaklyseasonalbasedonpercentageofprecipitationduringsummer.Mediterraneanbiomesaremostlydry<700mmannuallybutsomeareasarewet>1300mmannually;andmanyareasareweaklyseasonal>12%ofannualprecipitationduringsummer.WealsousedNOAANCDCclimaterecordstocharacterizeinterannualvariabilityofannualanddry-seasonprecipitation,aswellastrendsinannual,winter,anddry-seasonprecipitationfor337sitesthatmetthedataqualitycriteriafrom1975to2015.Mostsignicantly,sitesinmanyMediterranean-climateregionsshowdownwardtrendsinannualprecipitationsouthernCalifornia,Spain,Australia,Chile,andNorthernItaly;andmostofNorthAmerica,theMediterraneanbasin,andChileshoweddownwardtrendsinsummerprecipitation.VariationsinannualandsummerprecipitationlikelycontributetothehighbiodiversityandendemismcharacteristicofMediterranean-climatebiomes;thedataindicatetrendstowardharsherconditionsoverthepast40years. Keywords:Mediterraneanclimate;spatialvariability;interannualvariability;MannKendalltrendsanalysis;precipitationseasonality;dryseason 1.IntroductionClimateisamongthemostfundamentalfeaturesthatdenethehumannaturalsystem[1].Patternsoftemperatureandprecipitationinuencebioticcommunitycompositionandstructure[25],aswellastheextentoffeasiblehumandevelopmentandthewaterresourcedevelopmentnecessarytomeethumanneeds[68].Precipitationvariabilityisalsoimportantfordeninghumannaturalsystems:differencesinannualprecipitationandseasonalvariabilityfromoneyeartothenextcharacterizethecapacitytosustainecosystemsandassociatedprovisioning,regulating,andsupportingservicesacrosstheyearsandinthelongterm[6].Further,changesinthesepatternscancausefundamentalshiftsinthecapacityforecosystemstore-organizetopreexistingsteadystatesresilience[9],andcanalsoinuencewaterresourceplanningforhumandevelopment[10].Precipitationandtemperaturearefactorscommonlyutilizedtodeneclimatezones[11,12].Forexample,ecosystemsoftheMediterraneanbiomecommonlyexperiencecoolwetwintersandWater 2017 9 ,259;doi:10.3390/w9040259www.mdpi.com/journal/water

PAGE 2

Water 2017 9 ,259 2of21warmdrysummers,meaningthatprecipitationisdistributedduringthecoolerhalfoftheyear,withgenerallymildratherthancoldwintertemperatures;summersaretypicallydry[13].Theseregionsalsotypicallyreceivemorethan12inchesmmofprecipitationannually,distinguishingthemfromaridregions[14].TheMediterraneanbiomeanditsassociatedtemperatedry-summerclimatetypeislocatedoneveryinhabitedcontinent,includingtheportionsofEurope,Africa,andAsiasurroundingtheMediterraneanSeaandextendingeastwardthroughsouthernTurkeyandnorthernSyria;centralcoastalChileonthewestsideofSouthAmerica;thewestcoastofNorthAmericastretchingfromnorthernMexicointoCanada;thewest-facingsoutherncoastofAustralia;andthesouthwesternendofSouthAfrica[15].Theseregionsaretypicallylocatedatlatitudesbetween30and40degrees,onthewesternedgeofthecontinent[16].Precipitationisstronglyinuencedbyorographiceffectsthroughmostareas[1719]TheseasonalityofprecipitationisconsideredthemainfactorinuencingterrestrialandaquaticecosystemsoftheMediterraneanbiome[16].Thisdistinctclimatepatterntendstoresultinhighspeciesrichnessandendemism[20,21].Terrestrialplantshaveadaptedtothelongsummerdryseasonbydevelopingsclerophyllousfoliageanddrought-adaptedrootgrowth[22].Inaquaticecosystems,theprolongeddryseasonleadstoreducedsummerstreamow,whichcausesashiftinecosystemcompositionfromorganismsthatarewell-suitedtolotici.e.,owingenvironmentstothosewellsuitedtolenticnon-owingones[23,24].Interannualvariabilityofclimatecanproducevaryingowconditionsbytheconclusionofthedryseasoneitherowingorintermittent,whichcanalsocausedifferencesincommunitycompositionandstructurefromoneyeartothenext[25,26].SpatialanalyseshavesuggestedthatthedistributionofareashavingMediterraneanclimatecharacteristicswillchangewithchangingglobalclimatedynamics[15],andthepersistenceofMediterranean-climateecosystemsareatriskfromtheincreasedvariabilityandshiftsinprecipitationandtemperaturepatternsexpectedwithclimatechange[27,28].Seasonalityofprecipitationalsoinuencesthepatternsofhumandevelopment.Becauseoftheseasonaldivergencebetweenwhenwaterismostneededandwhenitismostavailableespeciallyrelatedtoagriculturalwaterneeds,Mediterranean-climateareasoftenneedhighlevelsofwaterinfrastructuretomeethumanwaterdemand.SpainhasmorewaterstoredinlargedamspervolumerunoffthananyothercountryinEurope[29].LargedamsarealsocommonamongtheriversofthewestcoastoftheUnitedStates,especiallyamongtheSacramentoSanJoaquinandtheColumbiaRiverdrainagenetworks[30,31]thoughbothoftheseriversystemsobtainmuchoftheirdischargefrommountainousareasthatreceiveheavywintersnowfall.LargereservoirsarealsofoundintheMurrayDarlingbasin,withitsoutletonthesoutherncoastofAustraliasoutheastofAdelaide[32].Insomecases,smallreservoirsareubiquitousacrossthelandscape[33].Conceptually,thecharacteristicsofthedry-summerclimateandtheMediterraneanbiomearesimilar.Thetemperatedry-summerclimateKoppentypeCsischaracterizedbyacombinationofmoderatetemperatureandadrysummerseasone.g.,theamountinthewettestmonthcomparedtoamountinthedriestmonth[11,12];plantsandanimalsintheMediterraneanbiomehavedevelopedphysiological,morphological,orbehavioraladaptationstowithstanddry-seasonconditions[23].However,despitethisrelativelywell-acceptedconceptoftheMediterraneanclimateanditseffectsonbiota,thereisadiscrepancybetweenregionswithdrysummersandtheMediterraneanbiome.Forexample,onsomecontinents,theCsmoderatelydrysummerclimatetypeextendsfarbeyondareasdemarcatedastheMediterraneanbiome;andonothers,theMediterraneanbiomeextendsbeyondtheCsclimatetype.ResearchersmayusethetermMediterraneanclimateinterchangeablywiththeKoppentypeCs[10,13],butdifferencesinbiomeclassicationacrosstheCsclimatetypesuggestthathydrologicvariationswithintheCsclimatetypemayleadtodissimilarecologicalcommunities.Further,highbiodiversityandendemismwithinandamongMediterraneanbiomessuggestthatclimaticconditionsmayalsobehighlyvariablewithinandamongtheseregions.AdditionalanalysistoarticulatethespecicclimaticcharacteristicsoftheMediterraneanbiomecanhelpresearcherstobettercharacterizetheextentofdrysummerfacedbyterrestrialandaquaticorganismsadapted

PAGE 3

Water 2017 9 ,259 3of21totheseregions,aswellasthewatermanagementchallengesnecessarytosupporturbanandagriculturaldevelopment.Changesinclimateconditionsthroughthe21stCenturyareexpectedtoaltertheamountanddistributionofMediterranean-typeecosystems.Researchhasindicatedthatclimateandassociatedhydrologicprocessessuchasstreamowproductionhavealreadybeguntochangeinsomepartsoftheworld,includingsignicantdownwardtrendsinprecipitationinpartsofIndia[34]andNewZealand[35],aswellasdownwardtrendsindry-seasonstreamowandupwardtrendsinpeakstreamowinpartsoftheUnitedStates[3640].Iftemperatedry-summerclimatezonesareundergoingchangesinclimatevariablessuchasincreasedordecreasedannualprecipitationthesechangescouldalreadybecausingshiftsinecologicalcommunitycompositionandstructure.ThepurposeofthispaperistoexamineinmoredetailthetemperatedryclimaterelativetotheMediterraneanbiometoaddresstwobroadquestions:canwebetterdescribetheclimatecharacteristicsofplaceswithMediterraneanbiomebeyondthetemperatedry-summerCsclassication,throughpatternsofannualprecipitation,seasonalprecipitation,andvariability;andaretemperatedry-summerregionsincluding,butnotexclusively,theMediterraneanbiomeshiftingtowardwetterordrierconditions?ThesequestionsarefundamentallyinterrelatedbecausearticulationoftheMediterraneanbiomeclimatecharacteristicsaswellassurroundingareaswithdrysummerclimatecoupledwithanalysisoftrendsinannualanddry-seasonprecipitationcanhelppredictthoseregionswhereecologicalchangesmaybeunderway,causingdeviationsfromoroutsideofthebiome,deviationstowardtheadaptationsoforganismsintheMediterraneanbiome.Theseanalysesmayalsobeusefulforwatermanagerslookingtounderstandwhetherrecentmulti-decadalprecipitationtrendsresembleexpectednear-termclimatepredictions. 2.Methods 2.1.ClimateoftheMediterraneanBiomeTomoreclearlyidentifytheclimatecharacteristicsoftheMediterraneanbiomeandidentifydifferencesinbiomeswithinplaceswithatemperatedry-summerclimate,werstidentiedplaceswithtemperatedry-summerclimatethatmeetanannualprecipitationthresholdusingtheWorldClimdataset.TheWorldClimdatasetisahigh-resolutionspatialclimatedatasetwithdetailedinformationcharacterizingtheprecipitationandtemperatureatmonthlyintervalsfortheglobeatthe30arc-secondapproximately1km2scale,basedon30yearsofdatafromclimatestationsacrosstheglobe[41].Usingaveragemonthlyprecipitationandtemperaturedatafromthe30-yearWorldClimdataset,weidentiedregionswithtemperatedry-summercharacteristicsasthosehaving: 1.Meanannualprecipitationgreaterthan300mmsuggestingatleastsemiaridconditions[14]. 2.Morethan50%ofthemeanannualprecipitationoccurringduringthecoolhalfoftheyear,frommid-fallthroughwintertomid-springwhichweabbreviateFWS.Fordry-summerclimateareas,wedenedtheFWSperiodasthewetterofeitherOctoberMarchorNovemberAprilinthenorthernhemisphere;andeitherAprilSeptemberorMayOctoberinthesouthernhemisphere. 3.Lessthan20%ofthemeanannualprecipitationoccurringduringthesummerthreemonths.ForeachoftheveMediterranean-climateareas,wedenedthesummerperiodasthedrierofeitherJuneAugustorJulySeptemberinthenorthernhemisphere;andeitherDecemberFebruaryorJanuaryMarchinthesouthernhemisphere. 4.Meancoldest-monthtemperatureisgreaterthan0Celsiusindicatingalikelihoodofprecipitationpredominantlyoccurringasrainfallratherthansnowfall.Thesecriteriaidentifythoseareasthatarenotarid,havemildwintertemperatures,receivethemajorityoftheirannualprecipitationduringFWS,andhaveatleastslightlydrysummers.Thecriteriaweredesignedtobegenerousinidentifyingareasashavingadry-summerclimate:thetwoconditionsofseasonality%oftheaverageannualprecipitationoccurringduringFWSandlessthan20%of

PAGE 4

Water 2017 9 ,259 4of21precipitationduringsummerarenotparticularlystrong,andwedidnotidentifyamaximumannualprecipitation.Thisallowedustoexaminesimilaritiesanddifferencesoverabroaderrangeofareasthanmaytypicallybeconsideredashavingadry-summerclimateforexample,theareaincludingBarcelona,whichisintheMediterraneanbiomebutdoesnotttheKoppenCscriteria.Forthoseregionsthatmeetthefourclimatecriterialistedabove,wemappeddifferencesinthreecharacteristics:meanannualprecipitation;thepercentageofprecipitationoccurringduringtheFWShalfoftheyear;andthepercentageofprecipitationoccurringduringthethree-monthdry-summerseason.Mapsofthesefeaturesforthevetemperatedry-summerregionsappearintheSupplementaryMaterials,FiguresS1S5.TodescribeclimatecharacteristicsassociatedwiththeMediterraneanbiome,weclassiedtwovariablesmeanannualprecipitationandpercentageofannualprecipitationduringsummerintothreecategoriesTable1.Thisresultedinnineregionsidentiedaseitherwet,moderate,ordryintermsofmeanannualprecipitation,andeitherstrong,moderate,orweakseasonalitywithrespecttothepercentageofannualprecipitationduringsummer.WeoverlaidtheMediterraneanandsurroundingbiomeboundariesasdenedbytheWorldWildlifeFederationandtheNatureConservancy[42,43]ontomoderatedry-summerareasidentiedasoneofninecategoriesinTable1;andusingspatialanalyticaltoolsinArcMap,wequantiedthepercentageoftheMediterraneanbiomefallingintoeachclimatecategory.WealsoquantiedthoseareaswithintheMediterraneanbiomeidentiedasnotmeetingthethresholdofannualprecipitationi.e.,lessthan300mm,andnotmeetingthethresholdofseasonalityi.e.,greaterthan20%duringthedrysummerorlessthan50%duringFWS. Table1.Divisionofaverageannualprecipitationandpercentageofannualprecipitationinsummerseasonalityintothreegroups,givingninecategoriesofprecipitationandseasonality.ColorcodesinTable1correspondtocombinationofannualprecipitationandsummerseasonalityinFigures1,2,4,6. Precipitation:Seasonality:PercentageofPrecipitationinSummer MeanAnnualRainfallStrong:<5%Moderate:5%%Weak:12%% Dryclimate:<700mm Dryclimate/Strong seasonality Dryclimate/Moderate seasonality Dryclimate/Weakseasonality Moderateclimate: 700mm Moderateclimate/Strong seasonality Moderateclimate/moderate seasonality Moderateclimate/ weakseasonality Wetclimate:>1300mm Wetclimate/Strong seasonality Wetclimate/Moderate seasonality Wetclimate/ Weakseasonality 2.2.InterannualPrecipitationVariabilityandTrendsWeusedsite-specichistoricalprecipitationrecordsmeasuredatlocationswithinourmoderatedry-summerareasbasedontheWorldClimdatacalculationsdescribedabovetoevaluatevariabilityandtrendsofprecipitationcharacteristics.ThemajorityofprecipitationdatawereobtainedthroughNOAA'sNationalClimaticDataCenterNCDC,eitherthroughitsGlobalHistoricalClimatologyNetworkorClimateDataOnline,whichhashistoricalprecipitationdataavailableforapproximately29,000sitesacrosstheglobe[44,45].Allsiteswithinourdenedclimatecriteriawerereviewedfordurationofgaugeoperation;thosesiteswithmorethan20yearsofrecordwerefurtherreviewedtoevaluatethequalityofthedata.Siteswerefurtherdividedbydurationofmoreorlessthan30yearsinanalyticalproceduresasdescribedbelow.NCDCprecipitationdatasetsfrequentlyhaveperiodswhendataareomitted;theseomissionperiodsmaybeasbriefasoneorafewdays,oraslongasseveralyearsorevendecades.Additionally,datasetscommonlyhavedayswhenthedataareerroneousenteredas99.99inNCDCdatasets.Precipitationvaluesotherthan99.99wereassumedtobeaccurate.Amongthosesiteswithmorethan20yearsofrecord,weevaluatedthesumofthesetwodataerrorsi.e.,omittedanderroneousandselectedthosedatasetswithnomorethan5%combinederroroveratleast20continuousyearsofoperationbetween1975and2015TablesS1S5.Becausethisresultedina

PAGE 5

Water 2017 9 ,259 5of21lownumberofsitesamonglocationsinthesouthernhemisphere,wesupplementedtheNCDCdatawithdataavailableonlinefromtheAustralianBureauofMeteorologyindicatedinTableS1andtheDireccinMeteorolgicadeChileindicatedinTableS5.PrecipitationdatafromthesesourceswereselectedforfurtheranalysisiftheymetthesamequalitycontrolcriteriaastheNCDCdata. 2.2.1.InterannualVariabilityWeusedsite-specicdatatocomparetheinterannualvariabilityofmeanannualprecipitationandsummerprecipitationpercentagerelativetoournineclimateregions.Foreachsite,wecalculatedtheannualprecipitationandprecipitationduringthedry-summerseasonforeachyear;andthenthepercentageofprecipitationoccurringduringthedrysummer.Wethencalculatedthecoefcientofvariationoftendescribedastherelativestandarddeviationofannualprecipitationandannualpercentageofsummerdischargeforeachsite.ThecoefcientofvariationCV,denedastheratioofthestandarddeviationofthedatasettoitsmean,isparticularlyinformativebecauseitdepictsthedistributionofdatarelativetoitsmagnitude,thusmakingitausefultooltocomparedistributionsamongdatasetswithdifferentmagnitudes[46].WemappedtheCVofannualprecipitationandpercentageofsummerprecipitationontoourmapsofnineclimateregionsfromSection2.1toexaminewhetherthemagnitudeofannualorsummerprecipitationCVcorrespondedwithcharacteristicsofannualprecipitationorstrengthofsummerprecipitationseasonality. 2.2.2.InterannualTrendsinAnnualPrecipitation,WinterPrecipitation,andSummerPrecipitation; andVariabilityWeusedtheMannKendallnon-parametrictrendstesttodeterminewhetherannualprecipitation,percentageofwinterprecipitation,andpercentageofsummerprecipitationhavebeenundergoingastatisticallysignicantupwardordownwardtrendateachsiteoverrecentdecades.TheMannKendalltrendstestisparticularlyadvantageousforhydrologicdatasetsbecauseitisnotdependentonnormallydistributeddata;ittestswhetherdatashowanupwardtrendaswellasadownwardtrendthroughasequentialdatasete.g.,overaperiodofyears[47,48].Anupwardordownwardtrendinmeanannualprecipitationindicatesthatthesiteisreceivingmoreorlessprecipitation,respectively,overtime.Anupwardtrendinpercentageofprecipitationinsummerindicatesweakerseasonalitywithsummerbecomingmorewetovertime;andadownwardtrendinsummerpercentageindicatesstrongerseasonality.AnupwardtrendinpercentageofFWSprecipitationisanotherindicatorofstrongerseasonality,whereasadownwardtrendindicatesweakerseasonalitythisvariablehasimplicationsforaquaticecosystempersistencediscussedfurtherbelow.Additionally,weconductedaMannKendallanalysisofvariabilityinannual,percentageFWS,andpercentagesummerprecipitationastheabsolutevalueofthedifferenceofeachyear'svaluefromtheoverallmean.Anupwardtrendinvariabilityindicatesagreaterdifferencebetweenthevariableandthemeanovertime.Trendswereidentiedassignicanteitherat=0.05or=0.1.Becausemultiannualcyclese.g.,ENSOandeventse.g.,long-termdroughthavethepotentialtoinuenceprecipitationoverperiodsgreaterthanadecade[38,49],wedistinguishedthosesiteswithbetween30and40yearsofquality-checkedrecordfromthosewithbetween20and30yearsofquality-checkedrecord. 3.Results 3.1.ClimateoftheMediterraneanBiomeTheMediterraneanbiomegenerallytswithintheclimatecharacteristicsdescribedaboveprecipitationmorethan300mmannually,lessthan20%duringthesummerdryseason,lessmorethan50%duringthewinterhalfoftheyear,thoughportionsofeachregiondonottthesecriteriaFigure1.Manyoftheareasthatdonotmeetourcriteriaarearid,receivinglessthan300mmannually,includingtheRegindeAtacamainnorthernChile,theCentralValleyofCaliforniaandMexico'sBajaPeninsula,andtheSaharanAtlasRangeaswellaspartsofAustraliaandSouthAfrica;

PAGE 6

Water 2017 9 ,259 6of21Figure1.OtherportionsoftheMediterraneanbiomearenotadequatelyseasonaltomeetourcriteriai.e.,morethan20%ofprecipitationduringthethreesummermonths,includingpartsofSouthAfricatheeasternpartofWestCapeandwesternpartofEastCape,AustraliasouthernNewSouthWales,andEuropeNortheasternSpainandItaly.TheMediterraneanbiomealsodoesnotalwaysmatchtheKoppenGeigerdry-summerCsclimatetype. Figure1.Areaswithadry-summerclimatecharacterizedaseitherdry,moderate,orwetwithrespecttomeanannualprecipitationandeitherweakly,moderately,orstronglyseasonalintermsofpercentageofsummerprecipitation;alongwiththeMediterraneanbiomeandKoppenGeigerKGdry-summerclimatetypeCs.Areaswithoutcolorreceivelessthan300mmannually.AreaswithaMediterraneanbiomevarywithrespecttomeanannualprecipitationandpercentageofprecipitationduringsummer.InNorthAmerica,theMediterraneanbiomeisalmostentirelywithinourdrycategorylessthan700mm,mostlywithstrongseasonalityFigure1,Table2.TheMediterraneanbiomeofNorthAmericaalsoincludessmallerareasthatreceivemoreprecipitationmorethan700mmannuallyaroundtheSanFranciscoandMontereyBaysandthefoothillsoftheSierraNevadaRange;aswellasareasthataredryandmoderatelyseasonalthemountainseastofLosAngelesandSanDiegoand,farthersouth,dryandweaklyseasonal.InSouthAmerica,mostoftheMediterraneanbiome%receiveslessthan300mmannuallyTable2.Thoseareaswithinthebiomethatreceivemorethan300mmvarywithlatitude,includingdrierandmorestronglyseasonalareastothenorthandwettermoderatelyseasonalareastothesouthFigure1.ThemajorityoftheMediterraneanbiomeinAustraliaisdryandeitherweaklyormoderatelyseasonalTable2;onlyasmallportionofthebiomeinAustraliaalongthesouthwesterncoastisstronglyseasonalFigure1.AlargeportionoftheSouthAfricanMediterraneanbiomedoesnotmeetourcriteriaforseasonalityindicatingthatmorethan20%ofmeanannualprecipitationoccursduringthesummerseason;nopartsofSouthAfricaarestronglyseasonal.ThemajorityoftheMediterraneanbiomearoundtheMediterraneanSeaisdrymmannually;Table2;17%ismoderatewith

PAGE 7

Water 2017 9 ,259 7of21respecttoprecipitationmmandasmallportion.5%iswetalongtheAdriaticSea.Thoseareasthatarestronglyseasonaltendtobenorth-facingpartsofcoastalnorthernAfricaorwest-facingsouthernSpain,Greece,Turkey,southwestAsia;Figure1.BasedontheWorldClimspatialdataset,theMediterraneanbiomearoundtheMediterraneanSeaisthemostdiversewithrespecttoannualprecipitationandseasonality,comparedtootherregions:includingsmallareasontheeastcoastoftheAdriaticSeathatarewetwhereprecipitationexceeds1800mmannually,itincludesallninecategoriesofannualprecipitationandseasonalitywedeneinthisstudy. Table2.Percentageofbiomeineachregioncategorizedasdry,moderate,orwetwithrespecttoaverageannualprecipitation;andweakly,moderately,orstronglyseasonalwithrespecttopercentageofannualprecipitationinsummer;alongwithpercentagethatisverydrylessthan300mmannuallyandpercentagethatisnotseasonalmorethan20%orannualprecipitationduringsummer. ClimateCategory Precip,Seasonality AustraliaMedRegionNorthAmericaSouthAmericaSouthAfrica Biomearea,km 2 779.6002,051,900192,800148,00096,600 Dry,weak,%38.6%11.1%2.4%none14.9% Dry,moderate,%17.4%20.3%4.4%0.2%19.4% Dry,strong,%0.8%19.4%52.4%13.2%0.0% Moderate,weak,%0.5%2.3%0.1%none0.7% Moderate,moderate,%4.3%8.2%0.2%5.3%5.6% Moderate,strong,%1.7%6.1%10.5%10.9%0.0% Wet,weak,%None0.1%NoneNoneNone Wet,moderate,%None0.3%None7.9%None Wet,strong,%None0.05%NoneNoneNone Verydry<300mm,%29%19%26%63%21% Notseasonal,%8%13%4%0%38% Onmostcontinents,largeportionsofthedry-summerclimateareasextendbeyondtheMediterraneanbiome.PortionsofthewestcoastofNorthAmericathatreceivelessthan5%oftheannualprecipitationduringthesummerextendnorthintoOregon;westernWashingtonandVancouverIslandreceivelessthan12%ofannualprecipitationduringsummerFigure1.SpatialpatternsofannualprecipitationandseasonalityaresimilarinChile,whereareasfartherfromtheequatorandbeyondtheMediterraneanbiomearerelativelydryinsummerandreceivemoreprecipitationannuallythanclosertotheequator.PartsofTasmaniaalsoaredryinsummerandreceivemoreprecipitationthanintheMediterraneanbiomeofAustralia.IntheregionsurroundingtheMediterraneanSea,areaswithadrysummerandmorethan300mmprecipitationannuallyextendeastintonorthernIraqandwesternIran;andaroundthesouthernandsoutheasternCaspianSea. 3.2.InterannualVariabilityTheCoefcientofVariationCVforannualprecipitation,whichcharacterizesvariabilityoftotalprecipitationoveradenedperiod,isalsodifferentwithinandamongMediterranean-climateareas.WithinthegreaterMediterraneanregion,areaswithstrongerseasonalitysouthernSpainandPortugal,Morocco,Gibraltar,Crete,southwestAsiatendtohavehigherannualprecipitationCVthanareaswithweakerseasonalitythoughnorthernItalyandSicilyhavehighCVandweakormoderateseasonality;Figure2.AnnualprecipitationCVinNorthAmericaisgenerallygreaterinthesouththanthenorth,evenwithinareasofsimilarclimatecategoryforexample,annualprecipitationCVisgreaterinthedrystronglyseasonalsouthernCaliforniacoastthanthedrystronglyseasonalcentralcoastorSacramentoValley.CVislessinareasthatarewetormoderatelywet.CVofannualprecipitationisconsistentlylowerinAustraliaandSouthAfricacomparedtootherregions;andloweramongsitesalongthecoastofSouthAfricathaninland.AmongthefewsitesinSouthAmerica,annualprecipitationCVisvariableandshowsnodistinctrelationshipwithprecipitationcategoryorothergeographicalgradientse.g.,coastal/inland,north/south.Inallregions,annualprecipitation

PAGE 8

Water 2017 9 ,259 8of21variabilityisgreatestamongsitesthataredryandstronglyseasonal,andarelowinplacesthatarewetandinplacesthatareweaklyseasonalFigure3.CVdataalsoappearinTablesS1S5. Figure2.AnnualprecipitationCoefcientofVariationCVbasedonsite-specicdatainregionswithamoderatedry-summerclimatewhereclimaterecordsmetthestandardsdescribedinSection2.2,above. Figure3.MagnitudeofannualprecipitationCoefcientofVariationamongsitessortedbyclimatecategorydry,moderate,orwetwithrespecttoannualprecipitation;strongly,moderately,orweaklyseasonalwithrespecttopercentageofsummerrainfallandregion.TheinterannualvariabilityofsummerseasonprecipitationisalsohighlyvariablewithinandamongMediterranean-climateareas.SummerprecipitationCVislowamongallsitesinSouthAfricaandAustralia,comparedtoNorthAmericaandtheMediterraneanregion;andlowamongmostofthe

PAGE 9

Water 2017 9 ,259 9of21SouthAmericasitesFigure4.InthegreaterMediterraneanregion,summerprecipitationvariabilityisgreatestthroughthesouthernportion,fromMoroccoandSpain,toSardiniaandSicily,towesternTurkey.WithinNorthAmerica,summerCVishighertothesouththantothenorth.Amongallregions,stronglyseasonalareasindicatingverylowsummerprecipitationonaveragehavegreatersummerprecipitationCVthanareaswithweakerseasonalityFigure5. Figure4.PercentageofannualprecipitationduringsummerCoefcientofVariationCVbasedonsite-specicdatainregionswithamoderatedry-summerclimatewhereclimaterecordsmetthestandardsdescribedinSection2.2,above. Figure5.MagnitudeofsummerprecipitationpercentageCoefcientofVariationamongsitessortedbyclimatecategorydry,moderate,orwetwithrespecttoannualprecipitation;strongly,moderately,orweaklyseasonalwithrespecttopercentageofsummerrainfallandregion.

PAGE 10

Water 2017 9 ,259 10of21 3.3.Trends 3.3.1.AnnualTrendsTrendsinannualprecipitationsince1975varybetweenareaswithdrysummerclimates.MostAustraliasitesshowdecliningprecipitationtrends,althoughtwositesinSouthAustraliawestofSpencerGulfareanomalousindicatingupwardtrendsFigure6.AnnualprecipitationtrendsarealsospatiallyvariableinNorthAmericaandSouthAmerica.Amongsitesshowingtrends,precipitationhasdeclinedinrecentdecadesinthesouthernportionofdrystronglyseasonalNorthAmericacentralandsouthernCalifornia;andinSouthAmerica,sitesclosertothecoastshowdownwardtrendscomparedtothoseinland.AmongsitesinSouthAfrica,mostthatshowtrendsinannualprecipitationindicatelessprecipitationovertimeespeciallyinland.TrendsinannualprecipitationintheMediterraneanregionarealsospatiallyvariable.SitesingeneralshowdownwardtrendsinSpain,southernFrance,andnorthernItaly;andincreasingtrendsinnorthernAfrica.SitesshowingtrendsincentralandsouthernItalyandeastoftheAegeanSeaarelessspatiallysorted,withsitesindicatingupwardtrendsnearsitesindicatingdownwardtrends.Almostallsitesinmoderatelywetandwetareasthosewithmorethan700mmannually,withmorethan30yearsofrecentdata,showdownwardtrendswithexceptionsinNorthAfricaandCrete.Mostsiteswithupwardtrendsinannualprecipitationarelocatedindryareasmmannually. Figure6.Annualprecipitationtrends,1975sitesdistinguishedbetweenthosewith20and30yearsofrecord,inareaswithadrysummerclimate;signicanceat 0.05and0.1.Trendsinannualvariability,indicatedbychangeinthemagnitudeofdeviationfromthemeanovertheperiodofrecord,arealsogenerallysortedspatially.AmongthosesitesthatindicatedtrendsinAustralia,variabilityinWesternAustralia,NewSouthWales,andVictoriaisincreasing;anddecreasingamongsitesinSouthAustraliaFigure7.TrendsinannualvariabilityamongsitesinNorthandSouthAmericawerealsospatiallysortedbutlessconsistent.TwositesinwetweaklyseasonalSouthAmericaindicatedadownwardtrendinannualprecipitationvariability,whileothersitesshowednotrend.InNorthAmerica,onlythosesitesinCaliforniashowtrendsinannualprecipitationvariabilityat =0.05or0.1;mostbutnotalltrendsaredownward,indicatinglessvariationfromthe

PAGE 11

Water 2017 9 ,259 11of21meanovertime.MostMediterraneanregionsitesshownotrendsinannualprecipitationvariability;signicantupwardtrendsinannualvariabilityarelocatedinItaly,whilemostshowingdownwardtrendsinannualvariabilityareeastoftheAegeanSea.AlltrendsresultsalsoappearinTablesS1S5. Figure7.Annualvariabilitytrends,1975sitesdistinguishedbetweenthosewith20and30yearsofrecord,inareaswithadrysummerclimate;signicanceat 0.05and0.1. 3.3.2.FWSTrendsTrendsinFWSprecipitationinAustraliamirrorannualprecipitation:siteswithstatisticallysignicanttrendsshowdecreasingtrendssince1975Figure8.However,afewsitesthatindicatenotrendsinannualprecipitationsuchasthoseininlandWesternAustraliaindicatedownwardtrendsinFWSprecipitation.FWSprecipitationtrendsinNorthAmericaandSouthAmericaalsomostlyfollowtrendsinannualprecipitation.FWSprecipitationincentralandsouthernCaliforniaisgenerallydecreasingwithtwositesintheSacramentoValleyshowingupwardtrends;siteswithtrendsnorthofCaliforniavarythoughmostshownotrends.FWSprecipitationtrendsamongsitesinSouthAmericaarealsosimilartoannualtrends,althoughtwositesthatindicatedownwardtrendsinannualprecipitationindicatenotrendinFWSprecipitation.FWSprecipitationtrendsintheMediterraneanregionalsovaryoverspace.Similartoannualtrends,FWSprecipitationtrendsinnorthernAfricaaregenerallyupward;andtrendsinItalyandeastoftheAegeanSeadifferbylocation.Overall,fewersitesacrosstheregionindicateadownwardtrendinFWSprecipitationcomparedtoannualprecipitationtrends;moreshownotrend.TrendsinFWSprecipitationvariabilityweresimilartotrendsinannualprecipitationvariabilityinsomeregions,butdifferentinotherregions:mostnotably,thosesitesinVictoriaandNewSouthWalesthatshowedincreasingannualprecipitationvariabilityshowednotrendsinFWSprecipitationvariabilityFigure9.TrendsinFWSvariabilityamongsitesinNorthAmericaandSouthAmericawerealmostalldownwardthoughmanysitesshowednotrends.LikeNorthAmericaandSouthAmerica,trendsinFWSprecipitationandFWSvariabilityamongsitesinSouthAfricaaresimilartoannualtrends,withprecipitationamonginlandsitesdecreasingandcoastalsitesincreasing.MoresitesshowadecreasingtrendinFWSprecipitationvariability;onlyoneinNorthAmericawithmorethan30yearsofrecordshowsanincreasingtrendinFWSvariability.FWSprecipitationvariability

PAGE 12

Water 2017 9 ,259 12of21issimilartoannualvariabilitytrends:increasingthroughmostofItalyandMediterraneanislandsSicily,Corsica,Crete,anddecreasingthroughwesternTurkey. Figure8.FWSprecipitationtrends,1975sitesdistinguishedbetweenthosewith20and30yearsofrecord,inareaswithadrysummerclimate;signicanceat 0.05and0.1. Figure9.FWSvariabilitytrends,1975sitesdistinguishedbetweenthosewith20and30yearsofrecord,inareaswithadrysummerclimate;signicanceat 0.05and0.1.

PAGE 13

Water 2017 9 ,259 13of21 3.3.3.SummerTrendsTrendsinsummerprecipitationvaryamongtheveregions.InAustralia,summerprecipitationshowsanupwardtrendamongmostsitesinSouthAustralia;almostallsitesinWesternAustralia,NewSouthWales,andVictoriashownotrendsFigure10.SummerprecipitationtrendsinNorthAmericaarealmostalldownward,fromsouthernCaliforniathroughWashington.AlltrendsinsummerprecipitationinSouthAmericaaredownward.TrendsinsummerprecipitationthroughtheMediterraneanregionarealsomostlydownward,especiallyinSpain,Italy,andtheislandssouthoftheAegeanSea.SitesinplacesincludingnorthernAfrica,Sicily,Crete,andtheBalkans,aswellasSouthAfrica,donotshowthesamedecliningtrendsinsummerprecipitationaselsewhereintheregion.Siteswithtrendsinsummerprecipitationvariabilityarenotascommonasthosewithtrendsinsummerprecipitation.InAustralia,onlyafewsitesinWesternAustraliaincreasingandtwositesinSouthAustraliaandVictoriadecreasingshowtrendsinsummerprecipitationvariabilityFigure11.InSouthAmericaandSouthAfrica,onlyafewsitesineachregionshowsummervariabilitytrends.InNorthAmericaandtheMediterraneanregion,severalsitesshowtrendsinsummerprecipitationvariability;butthosetrendsdonotshowthesamegeographicdistributionoftrenddirectionassummerprecipitationpatterns. Figure10.Summerprecipitationtrends,1975sitesdistinguishedbetweenthosewith20and30yearsofrecord,inareaswithadrysummerclimate;signicanceat 0.05and0.1.

PAGE 14

Water 2017 9 ,259 14of21 Figure11.Summervariabilitytrends,1975sitesdistinguishedbetweenthosewith20and30yearsofrecord,inareaswithadrysummerclimate;signicanceat 0.05and0.1. 4.Discussion 4.1.TheMediterraneanClimateasaFunctionofAnnualPrecipitationandSeasonalityTheresultspresentedaboveindicateawiderangeofannualandseasonalprecipitationconditionsamongthevelocationsassociatedwithaMediterraneanbiome.ThoughthemajorityoftheMediterraneanbiomeintheMediterraneanregionfallsintoourcategoryofdrybetween300and700mmprecipitationannually,partsaremoderateandasmallfractiontheeastcoastoftheAdriaticSeaiswet.TheMediterraneanbiomeofChileistheonlyotherareatohavethisdiversityofannualprecipitation;theMediterraneanbiomesofNorthAmerica,Australia,andSouthAfricareceivelessthan1300mmannuallythroughout.TheMediterraneanregionalsohasthemostevendistributionofseasonality,withatleast10%strongly,moderately,andweaklyseasonal.MostoftheMediterraneanbiomesofNorthAmericaandSouthAmericaareeitherstronglyormoderatelyseasonal;inAustraliaandSouthAfrica,mostaremoderatelyandweaklyseasonal.WhilemanyfeatureshavebeenproposedasattributingtothehighspeciesdiversityinMediterraneanbiomes,includinglandscapeheterogeneity,generaltrendsinseasonality,juxtapositionrelativetootherclimatetypes,andgeologicalhistory[16],thewiderangeofclimateconditionswithinthebiomelikelyalsoplaysaroleinbiodiversity.AreaswithintheMediterraneanbiomesoftheMediterraneanregionandChileexhibitthegreatestrangeofannualprecipitation,fromaslittleas100mmincludingtheAtlasMountainsandAtacamaregionwithinintheMediterraneanbiometomorethan1800mm.Further,ourresultsindicatethatsomepartsofMediterraneanbiomeareasareweaklyseasonal;othersevenreceivemoreprecipitationinsummerthaninotherseasons.AttemptstocharacterizeMediterraneanbiomesbyclimaticorgeophysicalfeaturesarefew,typicallyhavingbeeneitherverycomplexe.g.,agro-bioclimaticclassicationsincludingseveralvariablesandsubcategories[50]orsimplisticwithrespecttogeographye.g.,dividingthe

PAGE 15

Water 2017 9 ,259 15of21Mediterraneanregionintofourquadrants[51]orclimatee.g.,asKoppen-basedsubcategoriesonwarm,moderate,orcoolsummertemperatures[12].Othershaveclassiedecosystemswithinsubregions[52].HomogenizationoftheMediterraneanbiomeasoneclimatetypemayoverlooktheopportunitytoidentifypotentialregion-wide,climate-drivencausesofobservedvariationsinaquaticorterrestrialecosystemsamongMediterraneanbiomes.Terrestrialecosystemsaregenerallyadaptedtoawarmdryseason[20]andaquaticecosystemsadapttotheoodinganddryingcharacteristicofstreamsthathaveastorm-drivenrainyseasonandthenalongdryseason[24];butdifferencesinecologicalcommunitieswithinandamongMediterraneanbiomesmaybedrivenbynuanceddifferencesinclimateandtopography[53].Equallyimportant,thethreatstoconservationwithintheecosystemsoftheMediterraneanbiome[54,55]maybeoversimpliedifthebiomeisassumedtobeuniform.Climatedifferencesinuencethemagnitudeofstressplacedonecosystemsbyhumandevelopment[56];forexample,conictsbetweenhumanwaterdemandsandaquaticecosystemswilllikelybedifferentinplacesthatreceiveover1000mmofprecipitationannuallythaninplacesthatreceivelessthan500mm.OurresultsindicatethattheclimateoftheMediterraneanbiomeandthustheMediterraneanclimatecanbeanycombinationofourannualprecipitationandseasonalitycategories,aswellastwoadditionalcategorieswedidnotinitiallyclassify:verydryconditionslessthan300mmannuallyandnon-seasonalmorethan20%ofannualprecipitationduringsummer.Basedonclimaticvariablesofaverageannualprecipitationandsummerseasonality,weproposeaclassicationoftheMediterraneanclimateaccordingtotheninecategoriesoutlinedinTable1,plustheadditionalcategoriesofverydryandnon-seasonal.ThesetwoadditionalcategoriescompriselargeportionsofsomeMediterraneanbiomes,includingChile'sRegiondeAtacama,thesouthernCentralValleyofCaliforniaandsouthernAtlasMountainsverydry,andwesternpartofEastCape,SouthAfrica,andnortheasternSpainnon-seasonal.TheareaswithinMediterraneanbiomesidentiedasnon-seasonaltintoourdrycategorymmannually:despitereceivingmorethanone-quarteroftheannualprecipitationinsummer,theymayhaveMediterranean-typeecosystemsbecauseannualprecipitationislow.ThisclassicationsystemmayhelptoexplaindifferencesinbiodiversityamongregionswithaMediterraneanbiome:forexample,despitebeingmuchsmallerinarea,ChilemayhavehigherbiodiversitythanAustraliabecauseaverageannualprecipitationrangesfrom100mminthenorth,tomorethan2000mminthesouth.Seasonalitymayalsohavebiodiversityimplications:biotainareaswithmoderateprecipitationandstrongseasonalitymaybebetteradaptedtoverydrysummerconditionsthanareaswithmoderateprecipitationandweakseasonality.ThisclassicationsystemmayalsobeusefulinapplyingtootherareaswithsimilarclimaticcharacteristicsbeyondtheMediterraneanbiome.Forexample,thevalleysofOregonandcoastalCalifornianorthoftheMediterraneanbiomereceivesimilarprecipitationannuallyasthewetportionsofChileandtheMediterraneanregion;andhavestrongerseasonalityreceivinglessprecipitationinsummerthanChile.Theseareashaveawet/stronglyseasonalormoderate/stronglyseasonalMediterraneanclimatetype;otherfactorssuchaslatitude,topography,andgeologymaydeterthepresenceofMediterranean-typeecosystems. 4.2.InterannualVariabilityTheinterannualvariabilityofannualprecipitationindicatesthatsomeareashavemoreconsistentprecipitationthanothers.Amongareasorientedeastwest,annualprecipitationinEasternAustraliaandtheeasternportionoftheMediterraneanregionismoreconsistentfromoneyeartothenexti.e.,lowerCVthaninwesternpartsofeacharea.AnnualprecipitationisespeciallyvariableinthecoastalareasofsouthernSpain,southernFrance,andwesternItaly,whereinfrequenttorrentialrainfalllikelyplaysaroleinlong-termannualprecipitationpatterns[57].Amongareasorientednorthsouth,variabilityinNorthAmericaisespeciallyhighinsouthernCaliforniaandisgenerallygreateralongthecoastthaninland;whileinChile,siteswithhighannualvariabilityarelocatedthroughoutthearea.ThisvariabilityinNorthAmericamaybearesultofelNio/SummerOscillationcycles,whichcanresultineitherveryheavywinterprecipitationorverylowwinterprecipitation[58];occasionalhigh

PAGE 16

Water 2017 9 ,259 16of21precipitationinsouthernCaliforniamaydisproportionallyaffectthetemporaldistributionofannualprecipitation.LowvariabilityinAustraliaandSouthAfricasuggeststhattheseareasarenotsubjecttotheclimatepatternsthatcreateoccasionalverywetyearsinotherMediterranean-climateregions.LowervariabilityamongMediterraneanbiomesinthesouthernhemispheremayleadtoadifferentcompositionoforawithrespecttospeciestraitscomparedtoareaswithhighervariability[59].Patternsofsummerprecipitationvariabilitymirrorthoseofannualvariability.MediterraneanbiomesofthesouthernhemispheregenerallyhavelowvariabilitycomparedtothoseinNorthAmericaandtheMediterraneanregion,indicatingthattheclimatepatternsthatcauseoccasionalsummerprecipitationinnorthernhemisphereMediterraneanbiomesdonothavesimilareffectsinthesouthernhemisphere.SummerprecipitationvariabilityisespeciallyhighinnorthwesternAfrica,southernEurope,andcoastalIsrael;andincentralandsouthernCaliforniaandasiteatSantiago,innorth-centralChile.Mostoftheseareasarestronglyseasonal,typicallyhavingverylittlesummerprecipitationlessthan5%ofannualprecipitation;summerprecipitationintheseregionsistypicallyveryclosetozeroinmostyears,butonoccasionisrelativelyhigh.TheserareeventsmaybeduetomigrationoftropicalmoisturefrommonsoonsinAsiaandAfrica[60].Interannualprecipitationvariabilityhasimportantimplicationsforwatermanagementtomeethumandemands.Lowervariabilityovertimeindicatesamoreconsistentsupplythanhighervariability:evenifaverageprecipitationisless,lessvariabilitytranslatestoincreasedcertaintyinavailablewaterfromoneyeartothenext[61],whichcanaffectecosystemservicessuchasirrigationforagriculture[62].Long-termprecipitationrecordsmostlyovertheperiod1975indicateawiderangeofvariabilityinMediterranean-climateareas:overall,theMediterraneanregionandNorthAmericahavemanylocationswithveryhighvariation,includingsomesiteswithcoefcientsofvariationgreaterthan0.5.Bydenition,aCVof0.5indicatesadistributionofannualprecipitationsuchthatonestandarddeviationisequalto50%ofthemean.Forexample,ifmeanannualprecipitationis500mmandtheCVis0.5,then68%ofthevaluesonestandarddeviationgreaterandonesmallerthanthemeanarewithin250mmofthe500mmaverage.Saidanotherway,approximately68%ofthevaluesarebetween250and750mm.IftheCVforasitewithsimilarmeanannualprecipitationmmis0.3,thentherangeofapproximately68%ofthevaluesis350to650mm.Watermanagersinplaceswithhighervariabilityofmeanannualprecipitationhavetoincorporateagreateroccurrencefrequencyoflow-precipitationconditionsintomanagementframeworks.Incorporationoflikelydroughtconditionsintowaterpoliciesiscriticalformaintaininghumanwell-beingduringperiodsofwaterscarcity[63];thisisespeciallyimportantinplaceswhereannualprecipitationCVishighbecauseofthegreaterfrequencyoflow-precipitationconditions. 4.3.TrendsTheresultsaboveindicatevaryingtrendsinannualprecipitationandproportionofsummerandFWSprecipitationamongandwithinMediterranean-climateregions.Datafrommanyareasindicateadownwardtrendinannualprecipitation.InAustralia,thiscouldbeexpected:theperiodbeginningthelatterhalfofthestudyperiod,years20ofthepast40yearscomprisedAustralia'sMillenniumDrought,thelongestperiodofconsecutivedryyearsinrecordedhistory[64].Basedonourresults,theeffectofthisdroughton30-to40-yearprecipitationtrendsdidnotleadtoasignicantdownwardtrendeverywhere;partsofSouthAustraliawithmorethan30yearsofdatarecordedanincreasingtrendinannualprecipitation.TheAustraliaresultsalsohighlightthelimitationsoftrendsanalysiswithlimiteddata:somesiteswithlessthan30yearsofrecordindicateincreasingtrendsovertimebecausedatafromseveralofthosesitesbeganduringtheMillenniumDrought.Long-termclimatevariationscanleadtodifferenthydrologictrendswhendataareevaluatedoverdifferentperiodsofrecord[37].BeyondAustralia,someportionsofMediterranean-climateareasindicatetrendsinmeanannualprecipitationwhileothersdonot.AllsiteswithtrendsinCaliforniaindicatedownwardtrends,similartomostsiteswithtrendsinSpain,SouthAmerica,SouthAfrica.NorthernAfricaistheonlyidentied

PAGE 17

Water 2017 9 ,259 17of21regionwhereannualprecipitationtrendsaremostlyupward.Inotherareas,suchascentralandsouthernItalyandTurkey,somesitesindicateupwardtrendswhileothersindicatedownwardtrends.Insuchareaswheretrendsresultsaremixed,directionoftrendsisnotassociatedwithourproposedclimatecategories:forexample,themixedupwardanddownwardtrendsineasternItalyandwesternTurkeyoccurwithinsimilarconditionsofdryormoderateannualprecipitationcoupledwithweakormoderateseasonality.Thesevariationsmaybeattributedtomicroclimaticconditions,whichcanbeinuencedbyatmosphericcirculationpatterns,orographicvariations,andproximitytocoastalmoisture[65].Summerprecipitationpatternsoverthepast30yearsalsoindicateadownwardtrendinmanylocations,suggestingthatsummersinmanyareasincludingSpainandItaly,California,andSouthAfricaarebecomingdrier.SomeareasbeyondtheMediterraneanbiomearealsobecomingdrierinsummer.OregonandWashingtoninNorthAmerica,andChilesouthofSantiago,allaremostlywet-stronglyseasonal;withthesummersbecomingdrier,ecosystemsintheseregionsmayshifttowardmorecloselyresemblingnearbyMediterraneanbiomes.ResearchhassuggestedthattheMediterraneanbiomesofsomeregionscouldexpandthroughthe21stcentury[15];ourresultsindicatethatchangesinclimaticdriversmayalreadybeunderway.TheseresultsalsohaveimportantimplicationsforunderstandingthelimitationsofspatialdatasetssuchasWorldClimversion1usedhereandelsewhere,whicharebasedonmodeled30-yearprecipitationdatasetsendingin1990.Thesedatasetsarenowalmost30yearsold,andourresultssuggestthataverageannualandseasonalprecipitationinmanylocationsmaynotbethesametodayastheywere30yearsago.FundingtosupportthedevelopmentofdatasetssuchasWorldClimversion2whichwouldincorporatedatafrom1970to2000iscriticalforunderstandinghowhumanandecologicaldemandsforwatercanbemetinthecomingcentury.Themajorityofsitesshownotrendsatalpha0.05or0.01amongannual,FWS,andsummerprecipitation.AndingofnosignicanttrendisacommonresultamongstudiesthatusetheMannKendalltesttoidentifytrendsamongseveralsites[39,40],andismorecommonamongprecipitationthandischarge[36].Thismaybeduetoabreakinclassication.Wechosetoshowtrendsatalphavaluesandnotoutsideofthosethresholds;asaresult,weconcludeasignicanttrendat =0.09butnotat0.11.SomestudiesreporttheZscorecalculatedbytheMannKendalltest[39,40],whichdescribesthemagnitudeofthetrendandisthuscontinuousfromverystrongtrendtoverylittleornotrend;butwefocusedonstatisticalsignicanceratherthanagraduatedscaleforalessnuancedclassication.Additionally,ourstudyuseseasilyobtaineddataoverlargeareasareaswithadrysummerclimateacrosstheglobeandcoverageovertheseareasisnotcomprehensive.Whereprecipitationdatacanbeobtainedandveried,focusedtemporaltrendsevaluationsmayprovideadditionalimportantinsightstomorethoroughlyconrmandbettercharacterizethespatialvariabilityofclimatetrends;andexploretheinuenceoftopographicfeaturesathigherresolution[65].Trendsthatindicatedecliningprecipitationhaveimportantimplicationsforaquaticandterrestrialecosystems.Forexample,decliningsummerprecipitationindicatesatrendtowardincreasedseasonality:manyplacesthatweremoderatelyorweaklyseasonalshoweddownwardtrendsinsummerprecipitation,withlessprecipitationtodaythaninpreviousdecades.Thissuggeststhatsummersarebecomingmoreharsh,whichmaycauseashifttowardaquaticandterrestrialecosystemsbettersuitedtostronglyseasonalconditions[66,67].Equallysignicant,downwardtrendsinannualprecipitationbutnotFWSprecipitationsuggestthatprecipitationisnotdecliningduringtherainyseason,butratherduringothertimesofyearpossiblyduringearlyfallandlatespring.Areductioninprecipitationduringlatespringastherainyseasonendscanhavemajorimplicationsforterrestrialandaquaticecosystems.Interrestrialecosystems,lessrainduringspringcanleadtoreducedvegetationproductivity[68]anddifferentcomposition[69].Instreamecosystems,lessprecipitationinspringcanleadtolessstreamowinsummer[70],whichcandepletehabitatforshcommunities[71].

PAGE 18

Water 2017 9 ,259 18of21 5.ConclusionsTheanalysesabovepresenttwoimportantndingsregardingstatusandtrendsofprecipitationinMediterraneanbiomes.First,climatecharacteristicsofannualprecipitationandseasonalityareremarkablyvariableinareascategorizedaswithinaMediterraneanbiome.InSouthAmericaandtheMediterraneanregion,averageannualprecipitationintheMediterraneanbiomevariesfromlessthan100tomorethan1800mm.Thepercentageofprecipitationduringsummeralsoishighlyvariable,fromlessthan5%inpartsofNorthAmerica,SouthAmerica,theMediterraneanregion,andAustralia,tomorethan20%inpartsofAustralia,SouthAfrica,andtheMediterraneanregion.ThesevariationsarelikelymajordriverscontributingtothehighbiodiversityofMediterranean-climateregions:someareasareverydrywhileothersareverywet,andsomeareashaveaverydrysummerseasonwhileothersdonot.OurproposedmethodofcategorizingareaswithaMediterraneanclimateaccordingtoprecipitationcharacteristicscanhelpscientistsandresourcemanagerstobetterunderstandthevaryingdegreesofpressureplacedbyprecipitationpatternsonecosystemsandwatermanagementinstitutions.ThesecondndingisregardingtrendsinprecipitationwithinMediterranean-climateregions.Thoughnotuniformthroughoutallregions,precipitationdatahavefollowedgeneraldownwardtrendsinannualprecipitationindicatinglessannualprecipitationovertimeanddownwardtrendsinsummerprecipitationindicatingadecliningproportionofprecipitationduringsummer.Theseshiftsarelikelytohaveconsequencesforaquaticandecosystems,aswellasforwatermanagementinstitutionsresponsibleformaintainingadequatewatersuppliestomeethumandemandswhileprotectingenvironmentalresources.Reducedsummerprecipitationcanleadtoincreasedstressonecosystemsandgreaterwaterdemandforagricultureandmunicipalities;reducedannualprecipitationsuggeststhatoverallavailabilityislessthanintherecentpast.Waterconservation,throughpracticessuchasmoreefcientuseandreduceddemand,islikelytobecomeevenmoreimportanttowatermanagementinMediterranean-climateregionsthroughthetwenty-rstcentury. SupplementaryMaterials:Thesupplementarymaterialsareavailableonlineatwww.mdpi.com/2073-4441/9/4/259/s1. Acknowledgments:WethanktheUniversityofFloridaIFASResearchandtheWestFloridaResearchandEducationCenterforsupportofthisproject.PublicationofthisarticlewasfundedinpartbytheUniversityofFloridaOpenAccessPublishingFund.Wealsothankthreeanonymousreviewerswhosefeedbackonearliermanuscriptdraftsgreatlyimprovedthequalityofthismanuscript. AuthorContributions:MatthewJ.Deitchconceivedanddesignedtheexperiments;MatthewJ.DeitchandMicheleJ.Sapundjieffperformedtheexperiments;MatthewJ.Deitch,MicheleJ.SapundjieffandShaneT.Feireranalyzedthedata;ShaneT.Feirercontributedreagents/materials/analysistools;MatthewJ.Deitchwrotethepaper.Authorshipmustbelimitedtothosewhohavecontributedsubstantiallytotheworkreported. ConictsofInterest:Theauthorshavenointerestorrelationship,nancialorotherwise,thatmightbeperceivedasinuencinganauthor'sobjectivityoranyotherconictofinterest. References 1.Carpenter,S.R.;Mooney,H.A.;Agard,J.;Capistrano,D.;DeFries,R.S.;Daz,S.;Dietz,T.;Duraiappah,A.K.;Oteng-Yeboah,A.;Pereira,H.M.;etal.Scienceformanagingecosystemservices:BeyondtheMillenniumEcosystemAssessment. Proc.Natl.Acad.Sci.USA 2009 106 ,1305.[CrossRef][PubMed] 2.Parmesan,C.Ecologicalandevolutionaryresponsestorecentclimatechange.Annu.Rev.Ecol.Evol.Syst.2006 37 ,637.[CrossRef] 3.Westerling,A.L.;Hidalgo,H.G.;Cayan,D.R.;Swetnam,T.W.WarmingandearlierspringincreasewesternUSforestwildreactivity. Science 2006 313 ,940.[CrossRef][PubMed] 4.Breshears,D.D.;Cobb,N.S.;Rich,P.M.;Price,K.P.;Allen,C.D.;Balice,R.G.;Romme,W.H.;Kastens,J.H.;Floyd,M.L.;Belnap,J.;etal.Regionalvegetationdie-offinresponsetoglobal-change-typedrought.Proc.Natl.Acad.Sci.USA 2005 102 ,15144.[CrossRef][PubMed] 5.Prentice,I.C.;Cramer,W.;Harrison,S.P.;Leemans,R.;Monserud,R.A.;Solomon,A.M.Specialpaper:Aglobalbiomemodelbasedonplantphysiologyanddominance,soilpropertiesandclimate.J.Biogeogr.1992 19 ,117.[CrossRef]

PAGE 19

Water 2017 9 ,259 19of21 6.MillenniumEcosystemAssessment.EcosystemsandHumanWell-Being:Synthesis;IslandPress:Washington,DC,USA,2005. 7.Piao,S.;Ciais,P.;Huang,Y.;Shen,Z.;Peng,S.;Li,J.;Zhou,L.;Liu,H.;Ma,Y.;Ding,Y.;etal.TheimpactsofclimatechangeonwaterresourcesandagricultureinChina. Nature 2010 467 ,43.[CrossRef][PubMed] 8.Patz,J.A.;Campbell-Lendrum,D.;Holloway,T.;Foley,J.A.Impactofregionalclimatechangeonhumanhealth. Nature 2005 438 ,310.[CrossRef][PubMed] 9.Ruh,A.;Holmes,E.E.;Rinne,J.N.;Sabo,J.L.Anomalousdroughts,notinvasion,decreasepersistenceofnativeshesinadesertriver. Glob.Chang.Biol. 2015 21 ,1482.[CrossRef][PubMed] 10.Deitch,M.J.;Dolman,B.RestoringSummerBaseFlowunderaDecentralizedWaterManagementRegime:Constraints,Opportunities,andOutcomesinMediterranean-ClimateCalifornia.Water2017,9,29.[CrossRef]11.Kottek,M.;Grieser,J.;Beck,C.;Rudolf,B.;Rubel,F.WorldmapoftheKppen-Geigerclimateclassicationupdated. Meteorol.Z. 2006 15 ,259. 12.Peel,M.C.;Finlayson,B.L.;McMahon,T.A.UpdatedworldmapoftheKppen-Geigerclimateclassication.Hydrol.EarthSyst.Sci.Discuss. 2007 4 ,439.[CrossRef] 13.Conacher,A.J.;Conacher,J.Introduction.InLandDegradationinMediterraneanEnvironmentsoftheWorld;Conacher,A.J.,Sala,M.,Eds.;Wiley:NewYork,NY,USA,1998. 14.Grove,A.T.;Miles,M.R.;Worthington,E.B.;Doggett,H.;Dasgupta,B.;Farmer,B.H.Thegeographyofsemi-aridlands[anddiscussion]. Philos.Trans.R.Soc.Lond.BBiol.Sci. 1977 278 ,457.[CrossRef] 15.Klausmeyer,K.R.;Shaw,M.R.Climatechange,habitatloss,protectedareasandtheclimateadaptationpotentialofspeciesinMediterraneanecosystemsworldwide.PLoSONE2009,4,e6392.[CrossRef][PubMed]16.Bonada,N.;Resh,V.H.Mediterranean-climatestreamsandrivers:Geographicallyseparatedbutecologicallycomparablefreshwatersystems. Hydrobiologia 2013 719 ,1.[CrossRef] 17.Montecinos,S.;Gutirrez,J.R.;Lpez-Corts,F.;Lpez,D.Climaticcharacteristicsofthesemi-aridCoquimboRegioninChile. J.AridEnviron. 2016 126 ,7.[CrossRef] 18.Abatzoglou,J.T.;Redmond,K.T.;Edwards,L.M.ClassicationofregionalclimatevariabilityinthestateofCalifornia. J.Appl.Meteorol.Climatol. 2009 48 ,1527.[CrossRef] 19.Ducic,V.;Lukovic,J.;Buric,D.;Stanojevic,G.;Mustac,S.PrecipitationextremesinthewettestMediterraneanregionKrivoijeandassociatedatmosphericcirculationtypes.Nat.HazardsEarthSyst.Sci.2012 12 ,687.[CrossRef] 20.Cowling,R.M.;Rundel,P.W.;Lamont,B.B.;Arroyo,M.K.;Arianoutsou,M.PlantdiversityinMediterranean-climateregions. TrendsEcol.Evolut. 1996 11 ,362.[CrossRef] 21.Olson,D.M.;Dinerstein,E.TheGlobal200:Prioritybiomesforglobalconservation.Ann.Mo.Bot.Gard.2002 89 ,199.[CrossRef] 22.Dallman,P.R.PlantLifeintheWorld'sMediterraneanClimates:California,Chile,SouthAfrica,Australia,andtheMediterraneanBasin ;UnivofCaliforniaPress:Berkeley,CA,USA,1998. 23.Gasith,A.;Resh,V.H.Streamsinmediterraneanclimateregions:Abioticinuencesandbioticresponsestopredictableseasonalevents. Annu.Rev.Ecol.Syst. 1999 30 ,51.[CrossRef] 24.Beche,L.A.;Mcelravy,E.P.;Resh,V.H.Long-termseasonalvariationinthebiologicaltraitsofbenthic-macroinvertebratesintwoMediterranean-climatestreamsinCalifornia,USA.Freshw.Biol.2006,51,56.[CrossRef] 25.Resh,V.H.;Bche,L.A.;Lawrence,J.E.;Mazor,R.D.;McElravy,E.P.;O'Dowd,A.P.;Rudnick,D.;Carlson,S.M.Long-termpopulationandcommunitypatternsofbenthicmacroinvertebratesandshesinNorthernCaliforniaMediterranean-climatestreams. Hydrobiologia 2013 719 ,93.[CrossRef] 26.Garc a-Roger,E.M.;delMarS nchez-Montoya,M.;G mez,R.;Su rez,M.L.;Vidal-Abarca,M.R.;Latron,J.;Rieradevall,M.;Prat,N.DoseasonalchangesinhabitatfeaturesinuenceaquaticmacroinvertebrateassemblagesinperennialversustemporaryMediterraneanstreams?Aquat.Sci.2011,73,567.[CrossRef]27.Gritti,E.S.;Smith,B.;Sykes,M.T.VulnerabilityofMediterraneanBasinecosystemstoclimatechangeandinvasionbyexoticplantspecies. J.Biogeogr. 2006 33 ,145.[CrossRef] 28.Filipe,A.F.;Lawrence,J.E.;Bonada,N.Vulnerabilityofstreambiotatoclimatechangeinmediterraneanclimateregions:Asynthesisofecologicalresponsesandconservationchallenges.Hydrobiologia2013,719,331.[CrossRef] 29.Batalla,R.J.;Gomez,C.M.;Kondolf,G.M.Reservoir-inducedhydrologicalchangesintheEbroRiverbasinNESpain. J.Hydrol. 2004 290 ,117.[CrossRef]

PAGE 20

Water 2017 9 ,259 20of21 30.Kondolf,G.M.;Batalla,R.J.HydrologicaleffectsofdamsandwaterdiversionsonriversofMediterranean-climateregions:ExamplesfromCalifornia. Dev.EarthSurf.Process. 2005 7 ,197. 31.Deitch,M.J.;Kondolf,G.M.SalmoninaMediterraneanClimate:California'sIncendiaryMix.InSustainableWater:ChallengesandSolutionsfromCalifornia;Lassiter,A.,Ed.;UniversityofCaliforniaPress:Berkeley,CA,USA,2015. 32.McClure,M.M.;Holmes,E.E.;Sanderson,B.L.;Jordan,C.E.Alarge-scale,multispeciesstatusassessment:AnadromoussalmonidsintheColumbiaRiverbasin. Ecol.Appl. 2003 13 ,964.[CrossRef] 33.Deitch,M.J.;Merenlender,A.M.;Feirer,S.CumulativeeffectsofsmallreservoirsonstreamowinNorthernCoastalCaliforniacatchments. WaterResour.Manag. 2013 27 ,5101.[CrossRef] 34.Mondal,A.;Khare,D.;Kundu,S.SpatialandtemporalanalysisofrainfallandtemperaturetrendofIndia.Theor.Appl.Climatol. 2015 122 ,143.[CrossRef] 35.Caloiero,T.AnalysisofrainfalltrendinNewZealand. Environ.EarthSci. 2015 73 ,6297.[CrossRef] 36.Asarian,J.E.;Walker,J.D.Long-TermTrendsinStreamowandPrecipitationinNorthwestCaliforniaandSouthwestOregon,1953-2012. JAWRAJ.Am.WaterResour.Assoc. 2016 52 ,241.[CrossRef] 37.Poshtiri,M.P.;Pal,I.PatternsofhydrologicaldroughtindicatorsinmajorUSRiverbasins.Clim.Chang.2016,134 ,549.[CrossRef] 38.Feng,D.;Beighley,E.;Hughes,R.;Kimbro,D.SpatialandTemporalVariationsinEasternUSHydrology:ResponsestoGlobalClimateVariability.JAWRAJ.Am.WaterResour.Assoc.2016,52,1089.[CrossRef]39.Rice,J.S.;Emanuel,R.E.;Vose,J.M.TheinuenceofwatershedcharacteristicsonspatialpatternsoftrendsinannualscalestreamowvariabilityinthecontinentalUS. J.Hydrol. 2016 540 ,850.[CrossRef] 40.Ficklin,D.L.;Robeson,S.M.;Knouft,J.H.ImpactsofrecentclimatechangeontrendsinbaseowandstormowinUnitedStateswatersheds. Geophys.Res.Lett. 2016 43 ,5079.[CrossRef] 41.Hijmans,R.J.;Cameron,S.E.;Parra,J.L.;Jones,P.G.;Jarvis,A.Veryhighresolutioninterpolatedclimatesurfacesforgloballandareas. Int.J.Climatol. 2005 25 ,1965.[CrossRef] 42.Olson,D.M.;Dinerstein,E.;Wikramanayake,E.D.;Burgess,N.D.;Powell,G.V.;Underwood,E.C.;D'amico,J.A.;Itoua,I.;Strand,H.E.;Morrison,J.C.V.;etal.TerrestrialEcoregionsoftheWorld:ANewMapofLifeonEarth:Anewglobalmapofterrestrialecoregionsprovidesaninnovativetoolforconservingbiodiversity. BioScience 2001 51 ,933.[CrossRef] 43.WorldWildlifeFundandNatureConservancyTerrestrialEcoregionLayer,2011.SourceDataObtainedon1February2017.Availableonline:http://maps.tnc.org/gis_data.htmlaccessedon20January2017. 44.ClimateDataOnlineDataPortal.Availableonline:http://www7.ncdc.noaa.gov/CDO/countryaccessedon17October2016. 45.GlobalClimatologyDataNetworkDataPortal.Availableonline:https://www.ncdc.noaa.gov/cdo-web/search?datasetid=GHCNDaccessedon1November2016. 46.Zar,J.H. BiostatisticalAnalysis ,4thed.;PrenticeHall:SaddleRiver,NJ,USA,1999. 47.Esterby,S.R.Reviewofmethodsforthedetectionandestimationoftrendswithemphasisonwaterqualityapplications. Hydrol.Process. 1996 10 ,127.[CrossRef] 48.Helsel,D.R.;Hirsch,R.M.StatisticalMethodsinWaterResources;USGeologicalSurvey:Reston,VA,USA,2002;Volume323. 49.Mochizuki,T.;Ishii,M.;Kimoto,M.;Chikamoto,Y.;Watanabe,M.;Nozawa,T.;Sakamoto,T.T.;Shiogama,H.;Awaji,T.;Sugiura,N.;etal.Pacicdecadaloscillationhindcastsrelevanttonear-termclimateprediction.Proc.Natl.Acad.Sci.USA 2010 107 ,1833.[CrossRef][PubMed] 50.LeHouerou,H.AnAgro-BioclimaticClassicationofAridandSemiaridLandsintheIsoclimaticMediterraneanZones. AridLandRes.Manag. 2004 18 ,301.[CrossRef] 51.DeFigueroa,J.M.T.;Lpez-Rodrguez,M.J.;Fenoglio,S.;Snchez-Castillo,P.;Fochetti,R.FreshwaterbiodiversityintheriversoftheMediterraneanBasin. Hydrobiologia 2013 719 ,137.[CrossRef] 52.Fernndez,N.;Paruelo,J.M.;Delibes,M.EcosystemfunctioningofprotectedandalteredMediterraneanenvironments:AremotesensingclassicationinDoana,Spain.RemoteSens.Environ.2010,114,211.[CrossRef] 53.Cid,N.;Bonada,N.;Carlson,S.M.;Grantham,T.E.;Gasith,A.;Resh,V.H.HighVariabilityIsaDeningComponentofMediterranean-ClimateRiversandTheirBiota. Water 2017 9 ,52.[CrossRef] 54.Underwood,E.C.;Viers,J.H.;Klausmeyer,K.R.;Cox,R.L.;Shaw,M.R.Threatsandbiodiversityinthemediterraneanbiome. Divers.Distrib. 2009 15 ,188.[CrossRef]

PAGE 21

Water 2017 9 ,259 21of21 55.Cox,R.L.;Underwood,E.C.TheimportanceofconservingbiodiversityoutsideofprotectedareasinMediterraneanecosystems. PLoSONE 2011 6 ,e14508.[CrossRef][PubMed] 56.Klve,B.;Ala-Aho,P.;Bertrand,G.;Gurdak,J.J.;Kupfersberger,H.;Kvrner,J.;Muotka,T.;Mykr,H.;Preda,E.;Rossi,P.;etal.Climatechangeimpactsongroundwateranddependentecosystems.J.Hydrol.2014 518 ,250.[CrossRef] 57.Romero,R.;Sumner,G.;Ramis,C.;Genovs,A.AclassicationoftheatmosphericcirculationpatternsproducingsignicantdailyrainfallintheSpanishMediterraneanarea.Int.J.Climatol.1999,19,765.[CrossRef] 58.Piechota,T.C.;Dracup,J.A.;Fovell,R.G.WesternUSstreamowandatmosphericcirculationpatternsduringElNio-SouthernOscillation. J.Hydrol. 1997 201 ,249.[CrossRef] 59.Cowling,R.M.;Ojeda,F.;Lamont,B.B.;Rundel,P.W.;Lechmere-Oertel,R.Rainfallreliability,aneglectedfactorinexplainingconvergenceanddivergenceofplanttraitsinre-pronemediterranean-climateecosystems. Glob.Ecol.Biogeogr. 2005 14 ,509.[CrossRef] 60.Alpert,P.;Baldi,M.;Ilani,R.;Krichak,S.;Price,C.;Rodo,X.;Saaroni,H.;Ziv,B.;Kishcha,P.;Barkan,J.;etal.RelationsbetweenclimatevariabilityintheMediterraneanregionandthetropics:ENSO,SouthAsianandAfricanmonsoons,hurricanesandSaharandust. Dev.EarthEnviron.Sci. 2006 4 ,149. 61.Padowski,J.C.;Jawitz,J.W.Wateravailabilityandvulnerabilityof225largecitiesintheUnitedStates.WaterResour.Res. 2012 48 .[CrossRef] 62.Ramos,M.C.RainfalldistributionpatternsandtheirchangeovertimeinaMediterraneanarea.Theor.Appl.Climatol. 2001 69 ,163.[CrossRef] 63.Wilhite,D.A.Preparingfordrought:Amethodology.InDrought:AGlobalAssessment;Wilhite,D.A.,Ed.;Routledge:London,UK,2000;VolumeII,Chapture35;pp.89. 64.Dijk,A.I.;Beck,H.E.;Crosbie,R.S.;Jeu,R.A.;Liu,Y.Y.;Podger,G.M.;Timbal,B.;Viney,N.R.TheMillenniumDroughtinsoutheastAustralia:Naturalandhumancausesandimplicationsforwaterresources,ecosystems,economy,andsociety. WaterResour.Res. 2013 49 ,1040.[CrossRef] 65.Lana,X.;Martnez,M.D.;Serra,C.;Burgueno,A.SpatialandtemporalvariabilityofthedailyrainfallregimeinCatalonianortheasternSpain,1950. Int.J.Climatol. 2004 24 ,613.[CrossRef] 66.Henne,P.D.;Elkin,C.;Colombaroli,D.;Samartin,S.;Bugmann,H.;Heiri,O.;Tinner,W.ImpactsofchangingclimateandlanduseonvegetationdynamicsinaMediterraneanecosystem:Insightsfrompaleoecologyanddynamicmodeling. Landsc.Ecol. 2013 28 ,819.[CrossRef] 67.Hershkovitz,Y.;Gasith,A.Resistance,resilience,andcommunitydynamicsinmediterranean-climatestreams. Hydrobiologia 2013 719 ,59.[CrossRef] 68.Chelli,S.;Canullo,R.;Campetella,G.;Schmitt,A.O.;Bartha,S.;Cervellini,M.;Wellstein,C.Theresponseofsub-Mediterraneangrasslandstorainfallvariationisinuencedbyearlyseasonprecipitation.Appl.VegSci.2016 19 ,611.[CrossRef] 69.Zwicke,M.;Picon-Cochard,C.;Morvan-Bertrand,A.;Prud'homme,M.P.;Volaire,F.Whatfunctionalstrategiesdrivedroughtsurvivalandrecoveryofperennialspeciesfromuplandgrassland?Ann.Bot.2015,116 ,1001.[CrossRef][PubMed] 70.Deitch,M.J.;Kondolf,G.M.;Merenlender,A.M.Hydrologicimpactsofsmall-scaleinstreamdiversionsforfrostandheatprotectionintheCaliforniawinecountry. RiverRes.Appl. 2009 25 ,118.[CrossRef] 71.Hwan,J.L.;Carlson,S.M.FragmentationofanIntermittentStreamduringSeasonalDrought:Intra-annualandInterannualPatternsandBiologicalConsequences. RiverRes.Appl. 2015 32 ,856.[CrossRef] 2017bytheauthors.LicenseeMDPI,Basel,Switzerland.ThisarticleisanopenaccessarticledistributedunderthetermsandconditionsoftheCreativeCommonsAttributionCCBYlicensehttp://creativecommons.org/licenses/by/4.0/.