Hydrology as a driver of biodiversity: Controls on carrying capacity, nicheformation, and dispersalMegan Konara,, M. Jason Todda, Rachata Muneepeerakulb, Andrea Rinaldoc,d, Ignacio Rodriguez-IturbeaaPrinceton University, Department of Civil and Environmental Engineering, Princeton, NJ, USAbArizona State University, School of Sustainability and Mathematical, Computational, and Modeling Science Center, Tempe, AZ, USAccole Polytechnique Fdrale de Lausanne, Laboratorie of cohydrology, Faculte ENAC, CH-1015, Lausanne, SwitzerlanddUniversita di Padov, Department IMAGE and International Centre for Hydrology, Dino Tonini, Padua, Italyarti cle i nfoArticle history:Available online 3 March 2012Keywords:HydrologyBiodiversityDispersalCarrying capacityNichesClimate changeabstractA synthesis is presented highlighting the importance of hydrologic variables and dynamics to biodiversitypatterns. The focus of this paper is the key hydrologic controls crucial towards quantifying the impacts ofclimate changes on the distribution of species. Specically, we highlight the hydrologic controls operatingon the carrying capacity, niche formation, and dispersal dynamics. This synthesis will facilitate avenues offuture research and is connected to issues of major practical importance, such as the integration of thestructureofrivernetworksintoconservationstrategiesandtheevaluationsoftheimpactsofclimatechange on biodiversity. 2012 Elsevier Ltd. All rights reserved.1. IntroductionMaintenance of biodiversity across multiple scales has been ofbasic interest to the biological and ecological sciences for decades[43,41,31]. The debate over the ultimate controls on biodiversity isa contentious one and indeed the numerous papers investigatingthem rarely reach solid conclusions. However, it is without debatethat patterns exist across taxa, geographical areas, and geologicaleras[67]. Withtheevergrowingbodyofliteraturedetailingthebenetsprovidedbybiodiversity[12,14,21,30], increasingatten-tion is being paid towards its fundamental drivers and howchanges to key components could affect specic patterns of biodi-versity. Additionally, increased global species loss [58] has made itcritical to better understand the processes that govern biodiversity.The principal threats to biodiversity vary widely depending ongeographic location and the complicating effects of differences inspatial and temporal scale. At present, the principal threats to bio-diversity are the effects due to land use change and associated hab-itatlossandfragmentation, astheyactonamuchshortertimescale than other processes [28,19,55,56]. Several global modelingscenarios show that land use changes will continue to be the prin-cipal reasonfor terrestrial biodiversityloss until at least 2050[34,56,71,72]. However, climatechangeislikelytobethemajorreason for biodiversity loss worldwide after 2050 [48,75,76].While the predicted ultimate percentage loss of species due toclimatechangevarieswidelyfromstudytostudy(e.g. [44,77]),the IPCC reports that 2030% of animal and plant species are likelyto be at high risk of extinction with a global mean annual temper-ature rise of 23 C [75]. Indeed, research has shown that despitethe numerous possible explanations for changes in biological pat-terns and communities, climate change effects are already inuenc-ingbiodiversitythroughrangeshiftsandalterationofphenology[19,56]. This loss of biodiversity has the ability to produce a multi-tudeofconsequences, suchasthelossofecosystemfunctioningand reduction or elimination of goods and services [56]. Many ofthese responses may be nonlinear and difcult to predict [5], lead-ing to rapid transitions or sudden shifts in ecosystemstates[56,20,74].One of the major pathways through which climate change willimpact biodiversity patterns is through altered hydrologic patternsand processes [83]. It is well known that climate change will im-pactglobal precipitationpatterns[53,75], resultinginincreasedvariability in rainfall regimes in both time and space [54], which,in turn, change the hydrologic conditions that regulate ecologicalprocesses [13,67,59,46]. This is one of the reasons why it is impor-tant to focus on the specic mechanisms through which hydrologyimpacts biodiversity. In this paper, we present a synthesis of someof the most important hydrologic controls on biodiversity, with aneye towards understanding the potential impacts of climatechange. Note that we make no attempt to summarize the vast lit-erature that relates to the many and varied interactions betweenhydrologyandspeciesdiversity. Instead, wefocusonhydrologic0309-1708/$ - see front matter 2012 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.advwatres.2012.02.009Corresponding author.E-mail address: mkonar@princeton.edu (M. Konar).Advances in Water Resources 51 (2013) 317325ContentslistsavailableatSciVerseScienceDirectAdvances in Water Resourcesj our nal homepage: www. el sevi er. com/ l ocat e/ advwat r escontrols that canbe usedto quantify the impacts of climatechanges on biodiversity.While the role of hydrology on biodiversity within freshwaterecosystemsmayseemselfevident, hydrologiccontrolsalsoplayavitalroleinstructuring andmaintainingterrestrialecosystems[16,65,73]. Hydrology has been shown to play a vital role in struc-turing terrestrial vegetation, particularly in water-limited ecosys-tems. In water-limited ecosystems, soil moisture controls theavailabilityofnutrientsandlimitsplanttranspiration[64,65]. Inhumid ecosystems, the interactions between water and energy cy-cles increases in importance, and the diversity of trees is inuencedbyevapotranspiration[13]. Vegetationcommunities inriverinesystemsareoftenstructuredbyhydrogeomorphological interac-tions [2]. Hydrologic disturbances, including droughts and oods,playanimportantroleinthemaintenanceof bothaquaticandoodplain ecosystems [38]. Likewise, droughts and oods areimportant determinants ofvegetationdiversityinterrestrialsys-tems [78].Hydrology is not the only factor impacting biodiversity patternsand processes. For example, it is well known that the species rich-ness of almost all life forms increases from high to low latitudesandalongelevationgradients. Thislatitudinal diversitygradientisthoughttobegeneratedbyseveral mechanisms, suchastheavailability of energy, historical perturbations, and interactions be-tween species, but may simply be a consequence of more land areain the tropics [66,67]. The biogeography of plants in mid- to highlatitudesmaybest beexplainedbythespacetimepatternsoftheshortwave radiativeux[17]. Determiningtheultimate con-trolsonbiodiversitypatternsiscomplicatedbythefactthatthespecicunderlyingmechanismsandtheirimportancemaydifferacross both spatial and temporal scales. Hydrologic processes actwith considerable variability across multiple spatial and temporalscales, which is one of the reasons why biodiversity is likely inu-encedbysomeaspect of hydrologyat most scales of analysis[13,63,59].In this paper we focus on three specic determinants of diversityfor which hydrology may play an important role and which can beusedtoquantifypotential climatechangeimpacts. InSection2,we describe the hydrologic control of the spatial pattern of carryingcapacityinsomeecosystems. Section3discussesthehydrologiccontrol of niches favoring or restricting the existence of differentspecies. Section4describes thehydrologiccontrol of dispersalmechanisms, with a focus on lotic populations in river networks.We look to the future in Section 5.2. Hydrologic control of carrying capacityCarrying capacity is dened as the maximum number of indi-vidualsorunitsoforganismsthatcanbemaintainedinagivenarea on a long-term basis. Some habitats are far more productivethan others and, in general, more productive areas support moreindividuals and more species [13]. However, this pattern is oftencomplicated and may followa non-monotonic relationship inmany systems [67]. For this reason, we draw examples from natu-ral systems for which an increase in the carrying capacity has beenshowntoleadtomorespecies. Thefocusof thissectionisthehydrologic control of the spatial distribution of carrying capacity.The relationship between hydrology and carrying capacity hasbeenestablishedforquitesometime[81,25,62]. Bothterrestrialandaquaticecosystemswithmorefreshwaterresourcestendtosupport more individuals. Here, we focus on the controlling inu-ence of hydrology inbothaquatic andterrestrial ecosystems:namely, the inuence of precipitation on the carrying capacity oftrees and the impact of ow characteristics on the carrying capac-ity of sh.Recently, mean annual precipitation was found to be the majordeterminantof potential woodycoverinAfricansavannas[73].Sankaran et al. [73] demonstrate that maximum fractional woodycover, which proxies for the carrying capacity of trees, is primarilycontrolledbymoisturelimitation. Theyconduct acontinental-scale analysis of Africa in an effort to determine whether savannasare primarily climatically determined or disturbance driven, nd-ingthat savannasarepredominantlywaterlimitedinlocationswith less than approximately 650 mm yr1, while those locationsthat receive greater than650 mm yr1are disturbance driven.Thus, mean annual precipitation controls the upper bound on woo-dy cover, although disturbance regimes and soil characteristics doimpose signicant controls on woody cover below the bound [73].RecentresearchbuildsupontheworkofSankaranetal. [73]andutilizes meanannual precipitationas a driver of carryingcapacitytomodel distributionsof treespeciesdiversity. Konaret al. [39] demonstratethat aneutral meta-communitymodel,coupled with an appropriate representation of tree carrying capac-ity, effectively reproduces empirical patterns of tree diversity. Themodel was not able to reproduce empirical tree diversity patternswithout a spatial representation of tree carrying capacity based onrainfall [39,11]. ThisanalysiswasconductedfortheMississippiWatershed (shown in grey in Fig. 1A), showing that mean annualprecipitationappropriatelycharacterizesthecarryingcapacityoftreesinhumidecosystems. Notethat forest coverwasusedasaproxyof treecarryingcapacityinKonaretal. [39], ratherthanwoody cover as in Sankaran et al. [73], which accounts for differ-encesinfunctionalform. Additionally, Sankaranetal. [73]focuson savannas, which, by denition are regions with tree-grass co-existence, i.e. treecover never reaches 100%insavannas. Twohydrological variables were considered for use as a driver of forestcoverinKonaretal. [39]: evapotranspirationandmeanannualprecipitation. The relationship between forest cover and mean an-nual precipitation exhibited a more well-dened relationship thanthat betweenforest cover andevapotranspiration. Additionally,projectionsof meanannual precipitationunder climatechangescenarios are readilyavailable, making thisvariable desirable forprojection purposes.Importantly, the modeling approach used in Konar et al. [39] haspredictivepowers, sinceitallowsforthedirectlinkageoflarge-scalebiodiversitypatternstoenvironmentalforcings. Projectionsof meanannual precipitationunder different climatescenarioswereusedtoobtainnewvalues oftreecarrying capacityfortheMississippi Watershed. With these resulting new carrying capaci-ties, Konar et al. [39] determine howvarious climate change scenar-ios are projected to affect tree diversity patterns in the MississippiWatershed. 15climatechangescenariosareimplementedinthemodel. Here, the spatially-explicit impacts under the most dramaticspecies-poor scenario are shown in Fig. 1B. Note that the probabil-ity of any particular outcome in large-scale macrobiodiversity pat-ternsisheavilyreliant ontheprobabilitiesassociatedwiththeprojectedprecipitationpatterns providedbytheglobal climatemodels. For this reason, the patterns should be interpreted as enve-lopes of plausible biodiversity scenarios, rather than as predictionsof biodiversity outcomes. Tree diversity patterns are impacted moreunder the species-poor scenarios than under the species-rich sce-narios, with the exceptions of a few select regions, where impactsare of comparable magnitudes under both scenarios. Additionally,rare species are disproportionately impacted under climate change[39], a nding shared with niche-based models [50].Recent research indicates that the timing and intensity of rain-fall may be a more important driver of carrying capacity than sheerquantitiesof rainfall insomesystems[70,22,1]. Inacontinent-scale analysis of Africa, Good and Caylor [22] build upon the workof Sankaranet al. [73] anddemonstratethat thequantityandintensityof rainfall eventsinuencestheupperlimit of woody318 M. Konar et al. / Advances in Water Resources 51 (2013) 317325vegetation cover. They show that areas with similar seasonal rain-falltotals havehigher fractional woody coverifthelocal rainfallclimatology consists of frequent, less intense precipitation events.This distinction is crucially important since climate change is pro-jectedtoleadtomoreintense, lessfrequentstormevents, withimportant repercussionsfor vegetation. Similarly, Bartholomeusetal. [1] demonstratethatincreasedextremesinsoil moisture,similartothoselikelytobeobservedunderclimatechange, willimpact plant functional groups, though this study does not addressthe impact of hydrologic variation on species diversity directly. Toour knowledge, there has not yet been a study that quanties theimpact of hydrologic variation on species diversity, so this repre-sents an important area for future research. Thus, although hydro-logicaveragesmaydrivethecarryingcapacityanddiversityinsomesystems, spatial andtemporal variabilitymayprovemoreimportant in other ecosystems.The carrying capacity of organisms in river systems is stronglyimpacted by the gradient of a number of physical factors connectedto the drainage network [81]. In fact, the ow of water, which isstronglycorrelatedwithotherphysicalvariables, canbeconsid-ered a master variable that limits the distribution and abundanceof organismsinrivers[63,59]. Recent researchonshbiomassdemonstratestheimportanceof owcharacteristics. Hallsetal.[27] developed a ood index that accounts for the extent and dura-tionof oodingduringtheentireoodperiodof theTonleSapRiver, Cambodia. Fish biomass varied by almost a factor of 5 in re-sponsetotheood index overthelastdecade [27]. TheMekongRiver is subject to rapid development, with the construction of sev-eral large-scale hydropower dams and reservoirs [40]. The relatedow alterations will have signicant impacts on productivity andnumbers of sh in Tonle Sap ecosystem, with signicant repercus-sions for sh diversity.Takingadvantageof therelationshipbetweenthenumberofsh and river ow, Xenopoulos et al. [84] analyze the projected im-pacts of climatechangeandfuturewater withdrawals onshdiversity at a global scale. To address this issue, they present therst coupling of a global hydrological model with a freshwater bio-diversity model. A species-discharge regression model was used toproducescenariosofshspeciesloss. Speciesrichnessgenerallyincreased with increased mean river discharge, in a similar mannerto the species-area curve of terrestrial ecosystems [84].Fig. 2 illustrates the 52 watersheds that Xenopoulos et al. [84]showwill lose more than 10% of their sh species. The combinationof increased evapotranspiration and decreased precipitation as a re-sult of climate change (specically, under the IPCC SRES A2 scenario[75]) is the most important driver of freshwater sh loss. However,theimpactofwaterwithdrawalforhumanconsumptiononshdiversity is particularly important in certain watersheds. For exam-ple, consumptivewateruseintheEuphratesandCauveryRiverswasthemajordriverof shspecieslossesinthesewatersheds.Xenopoulos et al. [84] present an excellent rst approximation tothe projected impacts of climate change and human modicationon sh diversity. However, as with all species-distribution models,there are inherent problems in making projections based on corre-lations with current environmental regimes (see e.g. Pearson andDawson [57]). For this reason, a model based upon ecological pro-cesses would be a most benecial addition to the literature.Inthissection, wehaveshownthathydrologicvariablesareimportant to the carrying capacity of trees and sh. Recent researchdemonstrates the inuence of hydrology on carrying capacity[73,27,22], aswellashowtoutilizethisrelationshiptoquantifythe impact of climate change on biodiversity patterns [84,39,1].3. Hydrologic control of nichesNiches refer to segments of ecosystems that have been parti-tioned in both space and time and that are available for utilizationby different organisms. The number of niches that a given ecosys-tem contains has long been thought to be a major driver of speciesdiversity[32,67]. Thehydrologiccontrol of nichesoccursacrossbothterrestrial andaquaticenvironments, butmaybebestdis-playedwithinwetlands, astheyhavebeenaprincipal locationfor studying the relationship between hydrology and the structur-ing of vegetation communities [49]. Due to their dynamic hydrol-ogy, climatechangeimpactsintheseareascouldbeparticularlysevere, as slight changes in water availability may have profoundinuences on the surrounding biodiversity. The focus of this sec-tion is the inuence of hydrology on niche structure and the roleclimate change may have in altering these relationships.Thehydrology ofawetland environment canbedescribed bymultiplefactors, butismostoftendescribedbyitshydropatternor hydroperiod, acombinationof theperiodicityof inundationevents for a given location along with the length and depth of inun-dation. One such environment where these relationships have beenextensively studied is the Everglades, with multiple studies show-ing a relationship between the hydropattern and vegetation com-munities of a givenarea [15,26,87]. Nevertheless, the role ofhydrology in structuring vegetation communities in wetlands is dif-cult to determine due to a suite of interacting variables [6,69,86].Fig. 1. (A) Map of the Mississippi Watershed within the continental United States. The Mississippi Watershed is highlighted in grey and the network structure in blue. (B)Impact of climate change on region-averaged local species richness (LSR) in sub-regions of the Mississippi Watershed. Shades of green indicate the percentage change in theregion-averaged local species richness under climate change, with dark green indicating a higher percentage lost. The general trend is that a higher percentage of species arelost in the west with a decreasing trend to the east. The change per local community in region-averaged local species richness under climate change is indicated for eachregion by the bold numbers. The species-rich regions east of the 100oW meridian lose more species, though these species represent a smaller percentage of species in theseregions. The mean local species richness in the South is anticipated to decrease by 6.3 species under climate change, the largest loss of all sub-regions. Taken and modiedfrom Konar et al. [39] and Bertuzzo et al. [4]. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)M. Konar et al. / Advances in Water Resources 51 (2013) 317325 319Additionally, the lack of long termtemporal and wide spatial cover-age data presents a further complication to understanding the roleof hydrology in structuring vegetation communities.Todd et al. [79] examine the hydrologic control of niches acrossa long temporal and wide spatial scale of the entire Everglades Na-tional Park (ENP). This analysis quanties the hydrological nichesof the dominant vegetation communities. Mean hydroperiod depthandpercenttimeinundatedbestdescribethevegetationniches.Forinstance, muhlygrassoccursmostoftenatshallowerdepthsand at locations that are not inundated for long period of time. Incontrast, bay-hardwood scrub favors more hydric locations. How-ever, resultsfor sawgrass, byfar themost commonvegetationcommunity across the ENP, supported earlier studies that hydrol-ogyisnottheonlyvariablestructuringsawgrassniches[26,29].Thus, theworkof Toddetal. [79] showedthat, whilemultiplefactors can inuence the landscape distribution of vegetation com-munities, hydrology plays a principal, if not dominant role, in manycases. Most importantly, this work indicates that hydrology struc-tures vegetative niches at large spatial and temporal scales, rein-forcing the results of many studies conducted previously atsmaller scales.In an effort to gauge the impacts of projected changes in precip-itation on plant diversity, Todd et al. [80] used the relationships be-tween vegetation communities and hydrologic variables developedin Todd et al. [79]. Specically, Todd et al. [80] utilized projectedprecipitationvalues forthis region under variousclimate changescenariostomodel futurehydroperiodcharacteristics. Projectedvalues of meandepth and percent time inundated are then usedto determine impacts to vegetation communities, based upon therelationships described by Todd et al. [79]. Under increasing emis-sions scenarios, precipitation decreased across the ENP by as muchas 8% from present scenarios, leading to an associated decrease inmeandepthandpercenttimeinundated(refertoFig. 3). Thesechanges inhydroperiodwere thenusedtoproject changes tovegetation communities. Thus, vegetation communities that favorxericconditions increasedinabundanceandcommunities thatfavor hydric conditionsbecame scarcer (refer to Table 1). Forin-stance, the relative abundance of muhly grass, which favors xericconditions, increasedby15%underthehighemissionsscenario.In contrast, the relative abundance of bay-hardwood scrub, whichfavorshydricenvironments, decreasedby66%. Thus, Toddetal.[80] utilizes the previously established relationships between hyd-roperiodsandspeciesdistributions[79] toprojectthepotentialimpacts of climate change on plant diversity at large spatial scales.Anotherwetlandregionthathasundergoneintenseresearchregarding the inuence of changing climate on vegetation commu-nitystructureandbiodiversityis thePrairiePotholeRegionofNorthAmerica. This regionhas numerous wetlands coveringasuite of inundation regimes from temporary (one to two months)to semipermanent (inundated throughout most years). Associatedwith these varying inundation regimes, the vegetation foundtherein shows complex zonation dynamics. Through the years, var-ious models have been developed to simulate vegetation dynamicsacrossarangeof hydrologicparametersincludingprecipitation,runoff, potential evapotranspiration, snowpack and subsurface in-ow[60,61,37]. Thesemodelsgenerallydidasatisfactoryjobofreproducinghydrologicandvegetationconditionsof thesewet-lands. Whenincorporatingclimatechangescenariosofincreasesintemperature as comparedtohistoric levels, these wetlandsshowed marked hydrologic changes. In general, the wetlands expe-rienced earlier snowpack melting, decreased water depths and vol-ume, diminished hydroperiods, and reduced peaks from snowmeltandrainfall unlessrainfall acrosstheregionincreases57%perdegree of warming [37]. However, the inuence of these changeswas not uniform across wetland type with seasonal wetlands beingthe most affected. Theseseasonal wetlands normally persist intoFig. 2. Impact of change in water discharge from climate change and water consumption on sh diversity at a global scale. Colors display the percentage change in shdiversity, where red indicates approximately 50% loss of diversity and green indicates roughly 50% gain in diversity. The 52 watersheds that are projected to lose more than10%of theirshspeciesarenumbered. Climatechangeisprojectedtodecreaseshdiversitymorethanwaterwithdrawalsinmostwatersheds. However, inafewwatersheds, such as the Euphrates and Cauvery, sh species losses due to water withdrawals will be more severe. Taken and modied from Xenopoulos et al. [84]. (Forinterpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)320 M. Konar et al. / Advances in Water Resources 51 (2013) 317325the summertime months, but suffered shortened hydroperiods dueto the enhanced evaporative demand of a warmer climate.The changes in hydroperiod structure in the Prairie Pothole Re-gionhadamarkedeffectonthehabitatavailableformigratorywaterfowl. Johnson et al. [36] utilized the WETSIM model, a pro-cess-oriented, deterministic model that simulates wetland surfaceprocesses and vegetation dynamics, to quantify suitable waterfowlniche space. Three future scenarios were run in WETSIM to quan-tify the impacts of climate change on wetland hydrology and cor-responding waterfowl habitat availability. These scenarios are (1) a3 Ctemperatureincreasewithnochangeinprecipitation, (2)a3 Ctemperatureincreasewitha20%increaseinprecipitation,and (3) a 3 C temperature increase with a 20% decrease in precip-itation. FromFig. 4, it is clear that all threescenarios leadtomarked geographic shifts in the availability of waterfowl habitat.The scenario with increased temperature and decreased precipita-tionled to the most dramatic contractionin suitable habitat(Fig. 4E).Many of the studies investigating the role of hydrology in struc-turing vegetative niches implement climate change as a long, grad-ual process, witha trend towards elevated temperatures andassociated changes in hydrology. However, it is important to notethatindynamicenvironments, suchaswetlands, changesintheoccurrence of extreme precipitation events may play an importantrole[35]. Additionally, increasedvariabilityinrainfall patterns,leadingtooodinganddroughts, mayhaveprofoundinuencesonvegetation[85,47]. As anexample, Zedler [85] investigatedthe inuences of oods and droughts on wetland vegetation overthirty years, showing that biodiversity richness and evenness wasreduced abruptly due to sequential, catastrophic oods.Inthis section, we have describedthe important role thathydrologyplaysinstructuringnichesinwetlandecosystemsforbothvegetationandmigratorybirds. Recentresearchquantiesthehydrologicnichesofplants[15,26,79,87]andwaterfowl[36]and has recently been used to quantify the potential ecological im-pacts of climate change [36,80,85].4. Hydrologic control of dispersal mechanismsDispersal of organisms refers to their movement from one hab-itat to another and has long been recognized as an important driverof biodiversity patterns. Dispersal tends to increase the diversity oflocalcommunities[43], butsystemswithhighratesofdispersaltendtohavelowerglobal diversity[31]. Thedispersal distance[8] and shape of the dispersal kernel [68] are important determi-nants of diversity patterns as well. Empirical evidence suggest thatthedispersal kernelsof seedshavefat-tails[9], whichindicatesthatdispersalislikelytooccurifthereisanavailablepathway.Dispersal of organisms throughheterogeneous landscapes alsoFig. 3. The joint probability surface of percent time inundated and mean depth for (A) present conditions and under (B) low (B1), (C) middle (A1B), and (D) high (A2) climateemissions scenarios. The value of each pixel represents the relative frequency of all pixels across the Everglades National Park meeting both hydrologic conditions. Thesehydrologic scenarios were used to determine the impact of climate change on plant communities in the Everglades National Park, provided in Table 1. Taken from Todd et al.[79].Table 1Percent coverage of dominant vegetation types within Everglades National Park underthe present and high emissions scenarios. The percent change of dominant vegetationtypes between the present and high emissions scenarios are also provided. Only thosevegetation types constituting more than one percent of the total landscape are listed.Taken from Todd et al. [79].Vegetation type Present High % ChangeSawgrass 60.68 55.21 9.0Mixed Gramminoids 6.55 8.82 34.7Tall Sawgrass 5.80 2.24 61.4Muhly Grass 4.07 10.25 152.0Spike Rush 2.98 1.38 53.5Red Mangrove Scrub 2.16 0.92 57.4Bayhead 1.72 0.83 51.7Pine Savanna 1.59 5.17 224.3Willow Shrublands 1.47 1.36 7.9Dwarf Cypress 1.45 0.69 52.1Bay-Hardwood Scrub 1.44 0.49 66.1Brazilian Pepper 1.22 2.50 104.4Cattail Marsh 1.09 0.29 73.5Slash Pine with Hardwoods 0.88 2.96 237.2Hardwood Scrub 0.71 1.57 121.9Subtropical Hardwood Forest 0.75 1.43 91.1M. Konar et al. / Advances in Water Resources 51 (2013) 317325 321impactsthedistribution oforgansims. Thisisbecauseorganismsadapttheirdispersalstrategiesbasedupontheunderlyingland-scape structure [52].There are a number of hydrologic variables andhydrologicdependent landscapes which may impact dispersal. Here, we focuson systems with hydrologically determined carrying capacities andhabitat connectivity, since organisms adapt their dispersal strate-gies to both factors [52]. River systems present a unique opportu-nity to explore how dispersal is inuenced by the combination ofthe spatial distribution of carrying capacity and the landscape con-nectivitystructure(i.e. riverinetopology). Thisuniquelandscapestructure, inturn, makesitpossibleexplorehowecological dis-persal might be impacted by human modication. The focus of thissection is recent research that quanties how river topology inu-ences dispersal and diversity and, subsequently, how human mod-ication to river topology may impact diversity patterns.Withincreasedvariabilityinrainfall regimes under climatechange and increased demands on freshwater, not only will riverows be altered, but it is likely that human modication of riversystems will intensify,in aneffort to stabilize water supplies forhuman consumption, agriculture, and industry [82]. For example,inter-basinwater transfer projects, inwhichman-madecanalstransform the river connectivity to provide water to areas wheredemand exceeds supply, are anticipated to become more prevalent[42], asarewaterwithdrawalprojects[84]. Rivertopologyisanimportant element of dispersal, particularly for aquatic organisms.In fact, sh and other entirely aquatic organisms are constrained bythelandscapeimposeddispersal constraintsembeddedinrivernetworks [18].River networks impose constraints onpopulationprocesses, suchas spread, growth, and survival [23]. Thus, straight line distances in2-D space may be very different to distances as measured throughthe river network. In other words, as the crow ies may be differ-ent to as the trout swims [18]. For this reason, one of the principalconsequencesof theriverinestructureistoalterthelandscapethrough which organisms disperse. For example, the river networkstructure may inuence population extinction risk, slow the spatialspreadingof exotic organisms, enhancediversityat conuencepoints of the network [23], and impact the persistence of species [3].Muneepeerakul et al. [51] present a neutral model of sh diver-sityintheMississippi Watershedthat isbaseduponecologicalprocesses. In this metacommunity model, the sh disperse basedupon the underlying river network topology. The model producedexcellent ts to the empirical data, once the spatial distribution ofcarrying capacity was appropriate characterized and the dispersalprocesswasconstrainedtofollowtherivernetwork. Themodeleffectively reproduced several key diversity patterns; namely, theproleof local species richness, thelocal species richness fre-quency distribution, the rank-occupancy curve, and the b diversityas characterized by the Jaccards similarity index [51]. Additionally,Bertuzzo et al. [4] demonstrate that the model is able to reproducespatially explicit biodiversity patterns; namely, the environmentalresistance, which quanties the spatial loss of community similar-ity and is useful in identifying biogeographic regions. Remarkably,the model produced a good t to this spatial pattern with no fur-ther calibration [4].Theprocess-basedmodel presentedinMuneepeerakul et al.[51] demonstrates that thedispersal of shintheMississippiWatershed is controlled by the combination of the spatial distribu-tion of carrying capacity and river network structure. Interestingly,the model results indicate that sh do not exhibit biased dispersalandthemajoritytravel locally, thoughanimportantfractiondotravel verylongdistances[51,4]. Therivernetworkstructureisan essential component of the model; sh dispersal is dramaticallyimpactedbythedrainagestructure, sinceshoffspringarecon-strained to follow the ecological corridors dened by the network[51]. Thus, river network structure, a key hydrologically inuencedlandscape, isanimportantcontrollingfeatureofthedispersalofsh.Inter-basin water transfer projects restructure river connectivityand have the potential to alter landscape-scale patterns of speciesdistributionsbyintroducingnewdispersalroutes. Thelandscapeconstraints imposed by river topology are essential in determiningthe dispersal characteristics and diversity of entirely aquatic organ-isms (i.e. those constrained to solely within-network movement).However, the unique topology of river networks does not entirelycontrol the dispersal of aquatic organisms capable of out-of-network movement [23,24]. For example, salamanders utilize bothout-of-network andwithin-network movement, whichactstostabilize their populations [24]. Thus, the network structure is notas important in the dispersal of organisms that are able to travelout-of-network, suchassalamandersortrees[24,39]. Changingthe connectivity structure of river networks with man-made canalsforinter-basinwatertransferprojectsisanextremeexampleofout-of-natural-network movement, most likely to impact entirelyaquatic organisms.Fig. 4. A. The Prairie Pothole region of central North America (inset map) and six eco-regions. Yellow symbols represent weather stations (three per eco-region) used inWETSIM analyses. Simulated waterfowl habitat across the Prairie Pothole region under historic (B) and future scenarios of climate change (C, D, and E). Taken from Johnsonet al. [36]. (For interpretation of the references to colour in this gure legend, the reader is referred to the web version of this article.)322 M. Konar et al. / Advances in Water Resources 51 (2013) 317325A neutral metacommunity model of sh in river networks, sim-ilar to that of Muneepeerakul et al. [51], can be used to determinethe impacts of inter-basin water transfer projects on sh diversity.With this approach, Lynch et al. [42] quantify the potential ecolog-icalimpactsoftheproposedinter-basin watertransfersinIndia.Although the neutral model is not able to predict how the proposedinter-basin transfer project will impact specic species, it is a suit-ableframeworktoevaluatethepotentiallandscape-scaleconse-quences of changes to river connectivity [42].Changes in species richness due to the 11 proposed inter-basinwatertransfersarequantiedandshowninFig. 5. Theirstudydemonstrates the fragility of reaches in close proximity to the pro-posed canals. Note that local diversity increases throughout Indiawithincreasedconnectivity, whichisanticipatedunderthenullscenarioof habitat fragmentation[7]. However, global diversityin the system becomes spatially more homogenous with a notabledecrease in b diversity. Additionally, rare species decrease with theintroduction of the proposed canals [42]. In other words, more spe-cies are present in each location, but this is because certain speciesbecome widespread and dominant when otherwise distant basinsare linked through human modication. This increase in commonspeciescomesattheexpenseof endemicandrarespecies. ThestudybyLynchetal. [42]isthersttoexaminetheimpactofchanging network connectivity on basin-wide patterns of biodiver-sity and species richness.We can look to the past to observe how human introduction ofexoticorganismsintorivernetworkshasexacerbatedbiologicalinvasions. Invasivespeciesareknowntodisplacenativespecies,usuallydecreasingspecies diversityinthe systemas a whole[10]. Mari et al. [45] show that anthropogenic drivers of dispersalFig. 5. (A) Map of the Indian Peninsula. Major rivers are shown in gray, and the 11 proposed links considered in the analysis are shown in black. The four major river basinsinvolved are shaded and labeled. The links are numbered according to enumeration assigned by Indias National Water Development Agency. (B) Impact of the 11 proposedinterbasin water transfers on local species richness (LSR) for the Indian Peninsula river network. The absolute differences (after minus before) in LSR are shown for each link inthe river network. Taken and modied from Lynch et al. [42].Fig. 6. Dynamics of the zebra mussel invasion of the Mississippi Watershed. (A) Empirical data on zebra mussel spread over time. (B) Model snapshots of zebra mussel spread.Note that the model (B) reproduces the data (A) very closely when both the connectivity structure of the river network and human interferences in zebra mussel spread aretaken into account. Human-mediated dispersal impacts the spread of zebra mussels through commercial and recreational boating activities. Taken and modied from Mariet al. [45].M. Konar et al. / Advances in Water Resources 51 (2013) 317325 323(i.e. commercial navigation) in combination with the river networktopologyprovidethebestttothespatio-temporaldynamicsofinvasion of zebra mussels. In Fig. 6, model estimates of zebra mus-sel invasion in space and time provide excellent ts to the data ofzebra mussel invasion. Dispersal in the model was constrained tofollowtherivernetwork, butzebramusselsweretransportedatan accelerated rate and to regions of the network that may havebeenoutsidetheirrangebyships. ThestudybyMarietal. [45]illustratesthathuman-mediationof dispersal isof greatimpor-tance, particularly when temporal dynamics are taken into accountinrivernetworkcontext, wherenatural ecological distancesareconstrained to follow the topology.Inthis section, we have presentedexamples of hownetwork con-nectivity can be used to hind-cast or fore-cast the potential impactsof re-routing the connectivity structure of river systems. Examplesof human-mediateddispersal inbothIndia andthe MississipiWatershed indicate changes to river network topology may havedisastrous consequences for species diversity, largely through theintroduction of invasive species. As increased pressures are placedon global freshwater resources frompopulation growth and climatechange [82], it is likely that human modication of riverine systemswill intensify. It is imperative to understand how water withdraw-als and modications to the connectivity structure of river networkswill impact ecological dispersal, in order to properly evaluate theprojectedbiodiversityimpacts. Thisisparticularlyrelevantsincepolicy makers and resource managers are more able to directly im-pact water withdrawals and river engineering projects than globalclimate.5. Looking forwardItisimperativetounderstandthekeyrelationshipsbetweenhydrologic processes andbiodiversity as global freshwater re-sources are subject toincreasing pressures from both populationgrowth and climate change. There is certainly a vast literature thatrelates to the many and varied interactions between hydrology andspeciesdiversitythatwedonotcoverinthispaper. Instead, wehave outlined some of the major hydrologic drivers of biodiversity:carrying capacity, niche formation, and dispersal. In this synthesis,we have chosen to focus on hydrologic controls that can be used toquantify the impacts of climate changes on biodiversity.Forecasting the potential impacts of climate changes on biodi-versity is of increasing importance. For this reason, the focus of thispaper is research that uses hydrologic processes to quantifychangesinspeciescompositions. Goingforward, thereisaclearneedtodevelopamechanisticmodel linkingrainfall dynamicsandbiodiversity. Whilethemeanannual precipitationmaybethe major driver of the distribution of species in some cases, it isclear that rainfall timing and intensity will play a major role as well[22]. For this reason, the development of a model that integratesspeciesdiversityprocesseswithrainfalldynamics wouldmakeavaluablecontributiontotheliterature. Similarly, recentresearchhas highlighted the importance of soil moisture variability to plantfunctional groups [64,1]. It is important that future research con-tinue to develop the relationship between soil moisture processesand plant diversity.Asglobalwaterresourcesareincreasinglyconstrained, waterwithdrawals, transfer projects and dams will become increasinglypopular. Understandingtheecologicalimpactsofproposedengi-neering projects is particularly important and challenging. The im-pact toshdiversityduetowithdrawals andchanges inrivernetwork connectivity have only recently been examined quantita-tively. In the future, these modeling approaches could be improvedbyincorporatingamorecomprehensivesetof relationshipsbe-tween hydrologic variables and the carrying capacity of sh.Of course, the major hydrologic drivers that we have highlighteddo not act in isolation. For example, changes to global precipitationpatterns will impact the spatial distribution of carrying capacities ofcertain organisms. This, in turn, may lead those organisms to adapttheir optimal dispersal strategies accordingly. Similarly, increasedvariability in rainfall regimes may lead to enhanced niche spaces,which may, in turn, affect the carrying capacity and diversity of cer-tain species. Thus, accounting for the interplay between the majorhydrological drivers of biodiversity is an important and challengingtask going forward.Additionally, the interaction betweenthese hydrologic and otherabioticvariablesshouldbetakenintoaccountwhenappropriate[33]. Inparticular, researchthatmechanisticallylinkshydrologicprocesseswithotherfactors, suchasnutrientsandtemperature,willbeofincreasedimportanceinthefuture, particularlyasthescientic community strives to understandthe multitude of interac-tionsbetweenachangingclimateandthediversityofbiologicalsystems.AcknowledgementsM.K., R.M., and I.R.-I. acknowledge the support of the James S.McDonnell FoundationthroughtheStudyingComplexSystemsGrant(220020138). M.K., M.J.T., andI.R.-I. acknowledgethesup-port of NASA WaterSCAPES (Science of Coupled Aquatic ProcessesinEcosystemsfromSpace; NASANNX08BA43A). A.R. acknowl-edgesfundingfromERCAdvancedGrantRINEC22761andSFNGrant 200021/124930/1.References[1] BartholomeusRP, WitteJM,vanBodegomPM, vanDamJC, AertsR. Climatechange threatens endangered plant species by stronger and interacting water-related stresses. J Geophys Res 2011;11(G04023).[2] BendixJ, HuppCR. Hydrological andgeomorphological impactsonriparianplant communities. Hydrol Process 2000;14(16-17):297790.[3] Bertuzzo E, Suweis S, Mari L, Maritan A, Rodriguez-Iturbe I, Rinaldo A. Spatialeffects on species persistence and implications for biodiversity. Proc Nat AcadSci 2011;108(11):434651.[4] Bertuzzo ER, Muneepeerakul R, Lynch HJ, Fagan WF, Rodriguez-Iturbe I,Rinaldo A. On the geographic range of freshwater sh inriver basins. WaterResour Res 2009;45.[5] Burkett VR, WilcoxDA, StottlemyerR, BarrowaW, FagreD, BaronJ, et al.Nonlineardynamicsinecosystemresponsetoclimaticchange:casestudiesand policy implications. Ecol Complex 2005;2(4):35794.[6] BuschDE, LoftusWF, BassOL. Long-termhydrologiceffectsonmarshplantcommunity structure in the southern everglades. Wetlands1998;18(2):23041.[7] Chave J, Muller-Landau HC, Levin SA. Comparing classical community models:theoretical consequences for patterns of diversity. The Am Natur2002;159(1):123.[8] ChaveJ, NordenN. Changesof speciesdiversityinasimulatedfragmentedneutral landscape. Ecol Model 2007;207(1):310.[9] Clark JS, Silman M, Kern R, Macklin E, HilleRisLambers J. Seed dispersal nearand far: patterns across temperate and tropical forests. Ecology1999;80(5):147594.[10] Collins MD, Vazquez DP, Sanders NJ. Species-area curves, homogenization andthe loss of global diversity. Evol Ecol Res 2002;4:4574647.[11] Convertino M. Neutral metacommunity clustering and sar: river basin vs. 2-dlandscape biodiversity patterns. Ecol Model 2010;222(11):186379.[12] Costanza R, dArge R, de Groot R, Farber S, Grasso M, Hannon B, et al. The valueof the worlds ecosystem services and natural capital. Nature1997;387(6630):25360.[13] Currie DJ. Energy and large-scale patterns of animal and plant species richness.The Am Natur 1991;137(1):2749.[14] Daily GC. Natures services: societal dependence on natural ecosystems. IslandPress; 1997.[15] Davis SM, Gunderson LH, Park WA, Richardson JR, Mattson JE. Everglades: theecosystem and its restoration. St. Lucie Press; 1994.[16] DOdoricoP, LaioF, PorporatoA, RidolL, RinaldoA, Rodriguez-IturbeI.Ecohydrology of terrestrial ecosystems. Bioscience 2006;601(11):898907.[17] Eagleson, PS. Range and richness of vascular land plants: the role of variablelight. 2009.[18] Fagan WF. Connectivity, fragmentation, and extinction risk in dendriticmetapopulations. Ecology 2002;83(12):32439.324 M. Konar et al. / Advances in Water Resources 51 (2013) 317325[19] Fahrig L. Effectsofhabitat fragmentation on biodiversity. AnnRevEcol EvolSyst 2003;34:487515.[20] FolkeC, CarpenterS, WalkerB, SchefferM, ElmqvistT, GundersonL, etal.Regime shifts, resilience, and biodiversity in ecosystem management. Ann RevEcol Evol Syst 2004;35:55781.[21] Chapin III FS, Zavaleta ES, Eviners VT, Naylor RL, Vitousek PM, et al.Consequences of changing biodiversity. Nature 2000;405:23442.[22] Good SP, Caylor KK. Climatological determinants of woody cover in africa. ProcNat Acad Sci 2011.[23] Campbell Grant EH, Lowe WH, Fagan WF. Living in the branches: populationdynamics and ecological processes in dendritic networks. Ecol Lett2007;10:16575.[24] Campbell Grant EH, Nichols JD, Lowe WH, Fagan WF. Use of multiple dispersalpathways facilitates amphibian persistence in stream networks. Proc Nat AcadSci 2010;107(15):693640.[25] Gregory SV, Swanson FJ, McKee WA, Cummins KW. An ecosystem perspectiveof riparian zones. BioScience 1991;41(8):54051.[26] GundersonLH. Vegetationof theEverglades: determinants of communitycomposition. In: Davis SM, Ogden JC, editors. Everglades: the ecosystem andits restoration. St. Lucie Press; 1994.[27] Halls AS, Sopha L, Pengby N, Phalla T. New research reveals ecological insightsinto dai shery. Catch Culture 2008;14(1):812.[28] HeywoodVH, editor. Global biodiversityassessment. CambridgeUniversityPress; 1995.[29] Hofstetter RH, Parsons F. The ecology of sawgrass in the Everglades of southernFlorida. US Department of the Interior; 1979.[30] Hooper DU, Chapin III FS, Ewel JJ, Hector A, Inchausti P, Lavorel S, et al. Effectsof biodiversity on ecosystem functioning: a consensus of current knowledge.Ecol Monogr 2005;75(1):335.[31] Hubbell SP. The unied neutral theory of biodiversity andbiogeography. Princeton University Press; 2001.[32] Hutchinson GE. Homage tosanta rosaliaor why arethere somanykindsofanimals? Am Natur 1959;93(870):14559.[33] HwangT, BandL, HalesTC. Tecosystemprocessesat thewatershedscale:extending optimality theory fromplot to catchment. Water Resour Res2009;45:W11425.[34] Jenkins M. Prospects for biodiversity. Science 2003;302(5648):11757.[35] JentschA, KreylingJ, BeierkuhnleinC. Anewgenerationof climate-changeexperiments: events, not trends. Front Ecol Environ 2007;5(7):36574.[36] Johnson WC, Millett BV, Gilmanov T, Voldsteth RA, Guntenspergen GR, NaugleDE. Vulnerability of northern prairie wetlands to climate change. BioScience2005;55(10):86372.[37] Johnson WC, Werner B, Guntenspergen GR, Voldseth RA, Millett B, Naugle DE,et al. Prairie wetland complexes as landscape functional units in a changingclimate. BioScience 2010;60(2):12840.[38] JunkWJ, BayleyPB, SparksRE. Theoodpulseconcept inriver-oodplainsystems, In: D.P. Dodge[ed.] Proceedings of theInternational LargeRiverSymposium, vol. 106. 1989.[39] Konar M, Muneepeerakul R, Azaele S, Bertuzzo E, Rinaldo A, Rodriguez-Iturbe I.Potential impacts of precipitation change onlarge-scale patterns of treediversity. Water Resour Res 2010;46:W11515.[40] Kummu M, Sarkkula J. Impact of the mekong river ow alteration on the tonlesap ood pulse. Ambio 2008;37(3):18592.[41] Levin SA. The problem of pattern and scale in ecology. Ecology1992;73(6):194367.[42] Lynch HJ, Campbell Grant EH, Muneepeerakul R, Arunachalam M, Rodriguez-Iturbe I, Fagan WF. How restructuring river connectivity changes freshwatersh biodiversity and biogeography. Water Resour Res 2011;47:W05531.[43] MacArthur RH, WilsonEO. The theory of islandbiogeography. PrincetonUniversity Press; 1967.[44] MalcolmJR, LiuCR, NeilsonRP, HansenL, HannahL. Global warmingandextinctions of endemic species frombiodiversity hotspots. Conserv Biol2006;20(2):53848.[45] Mari L, Bertuzzo E, Casagrandi R, Gatto M, Levin SA, Rodriguez-Iturbe I, et al.Hydrologic controls and anthropogenic drivers of the zebra mussel invasion ofthe mississippimissouri river system. Water Resour Res 2011;47:W03523.[46] McGill BJ. Matters of scale. Ecology 2010;328:5756.[47] Miao SL, Zou CB. Vegetation responses to extreme hydrological events:sequence matters. Am Nat 2009;173(1):1138.[48] Millenium Ecosystem Assessment. Ecosystems and human well-being:Current state and trends, vol. 1. Island Press, 2005.[49] Mitsch WJ, Gosselink JG. Wetlands. 4th ed. Wiley; 2007.[50] Morin X,Thuiller W. Comparing niche- and process-based models to reduceprediction uncertainty in species range shifts under climate change. Ecology2009;90(5):130113.[51] Muneepeerakul R, Azaele S, Levin S, Rinaldo A, Rodriguez-Iturbe I. Evolution ofdispersal in explicitly spatial metacommunities. J Theor Biol2011;269:25665.[52] Muneepeerakul R, Bertuzzo E, Lynch HJ, Fagan WF, Rinaldo A, Rodriguez-IturbeI. Neutral metacommunity models predict sh diversity patterns inMississippi-Missouri basin. Nature 2008;453:2202.[53] NakicenovicN, Swart R, editors. Emissionsscenarios. Special report of theintergovernmental panel on climate change. Cambridge University Press; 2000.[54] OGormanPA, SchniderT. Thephysical basisforincreasesinprecipitationextremesinsimulationsof21st-centuryclimatechange. ProcNatlAcadSci2009;106:147737.[55] Parmesan C, Yohe G. A globally coherent ngerprint of climate change impactsacross natural systems. Nature 2003;421(6918):3742.[56] ParryML, CanzianiOF, Palutikof, JP, vanderLinden, PJ, Hanson, CE. editors.Climate Change 2007: Impacts, Adaptation, and Vulnerability. Contribution ofWorking Group II to the Fourth Assessment Report of the IntergovernmentalPanel on Climate Change. Cambridge University Press, 2007.[57] PearsonRG, DawsonTP. Predictingtheimpacts of climatechangeonthedistribution of species: are bioclimate envelope models useful? Global Ecol Bio2003;12:36171.[58] PimmSL, Russell GJ, GittlemanJL, Brooks TM. Thefutureof biodiversity.Science 1995;269(5222):34750.[59] Poff NL, Allan JD, Bain MB, Karr JR, Prestegaard KL, Richter BD, et al. The naturalow regime. Bioscience 1997;47(11):76984.[60] Poiani KA, Johnson WC. A spatial simulation-model of hydrology andvegetation dynamics in semi-permanent prairie wetlands. Ecol Appl1993;60(2):12840.[61] Poiani KA, Johnson WC, Swanson GA, Winter TC. Climate change and northernprairie wetlands: simulations of long-termdynamics. Limnol Oceanogr1996;41(5):87181.[62] Polis GA, Anderson WB, Holt RD. Toward an integration of landscape and foodwebecology:thedynamicsofspatiallysubsidizedfoodwebs. AnnRevEcolSyst 1997;28:289316.[63] PowerME, SunA, ParkerM, DietrichWE, WoottonJT. Hydraulicfood-chainmodels: anapproachto the study of foodweb dynamics inlarge rivers.BioScience 1995;45:15967.[64] Rodriguez-Iturbe I, Porporato A. Ecohydrology of water-controlledecosystems: soil moisture and plant dynamics. 1st ed. Cambridge UniversityPress; 2004.[65] Rodriguez-Iturbe I, Rinaldo A. Fractal river basins: chance and self-organization. 1st ed. Cambridge University Press; 1997.[66] Rosenzweig ML. Species diversity gradients: we know more and less than wethought. J Mammal 1992;73:71530.[67] RosenzweigML. Speciesdiversityinspaceandtime. CambridgeUniversityPress; 1995.[68] Rosindell J, Cornell SJ. Species-area curves, neutral models, and long-distancedispersal. Ecology 2009;90(7):174350.[69] Ross MS, Reed DL, Sah JP, Ruiz PL, Lewin MT. Vegetation: environmentrelationships andwater management insharkslough, everglades nationalpark. Wetlands Ecol Manag 2003;11:291303.[70] SaboJL, FinlayJC, KennedyT, Post DM. Theroleof dischargevariationinscaling of drainage area and food chain length in rivers. Science2010;330:9657.[71] Sala OE. Biodiversity across scenarios. Ecosystems and human well-being, vol.2. Scenarios: Island Press; 2005.[72] Sala OE, Chapin III FS, Armesto JJ, BerlowE, Bloomeld J, et al. Globalbiodiversity scenarios for the year 2100. Science 2000;287:17704.[73] SankaranM, HananNP, ScholesRJ, RatnamJ, AugustineDJ, CadeBS, etal.Determinants of woody cover in african savannas. Nature 2005;438(8):8469.[74] SchefferM, CarpenterS, FoleyJA, FolkeC, WalkerB. Catastrophicshiftsinecosystems. Nature 2006;413(6856):5916.[75] SolomonS, QinD, ManningM, Marquis M, AverytK, TignorMMB,Jr. LeRoyMiller H, Chen Z, editors. Climate Change 2007: the Physical Basis. Contributionof Working Group I to the Fourth Assessment Report of the IntergovernmentalPanel on Climate Change. Cambridge University Press, 2007.[76] StrengersB, LeemansR, EickhoutBJ, deVriesB, BouwmanAF. The land-useprojections and resulting emissions in the ipcc sres scenarios as simulated bythe image 2.2 model. Geo J 2004;61:38193.[77] Thomas CD, Cameron A, Green RE, Bakkenes M, Beaumont LJ, Collingham YC,et al. Extinction risk from climate change. Nature 2004;427(6970):1458.[78] TilmanD, El Haddi A. Drought andbiodiversityingrasslands. Oecologica1992;89:25764.[79] Todd MJ, Muneepeerakul R, Miralles-Wilhelm F, Rinaldo A, Rodriguez-Iturbe I.Possible climate change impacts on the hydrological and vegetative characterof Everglades National Park, Florida. Ecohydrology 2011;4.[80] ToddMJ, Muneepeerakul R, PumoD, AzaeleS, Miralles-WilhelmF, RinaldoA, etal. Hydrologicaldriversofwetlandvegetationcommunitydistributionwithin everglades national park, orida. Adv Water Resour2010;33:127989.[81] VannoteRL, Minshall GW, CumminsKW, Sedell JR, CushingCE. Therivercontinuum concept. Canad J Fish Aqua Sci 1980;37(1):1307.[82] VorosmartyCJ, GreenP, SalisburyJ, LammersRB. Global water resources:Vulnerability from climate change and population growth. Science2000;289(5477):2848.[83] WeltzinJF, LoikME, SchwinningS, WilliamsDG, FayPA, HaddadBM, etal.Assessing the response of terrestrial ecosystems to potential changes inprecipitation. Am Inst Biol Sci 2003;53(10):94152.[84] Xenopoulos MA, Lodge DM, Alcamo J, Marker M, Schulze K, Van Vuurens DP.Scenarios of freshwater sh extinctions fromclimate change and waterwithdrawal. Global Change Biol 2005;11:155764.[85] Zedler JB. Howfrequent storms affect wetlandvegetation: a previewofclimate-change impacts. Front Ecol Environ 2010;8(10):5407.[86] Zweig CL, Kitchens WM. Effects of landscape gradients on wetland vegetationcommunities: information for large-scale restoration. Wetlands2008;28:108696.[87] ZweigCL, KitchensWM. Multi-statesuccessioninwetlands:anovel useofstate and transition models. Ecology 2009;90:19009.M. Konar et al. / Advances in Water Resources 51 (2013) 317325 325