Charles Darwin University

Riverscape recruitment

A conceptual synthesis of drivers of fish recruitment in rivers

Humphries, Paul; King, Alison; McCasker, Nicole; Kopf, R. Keller; Stoffels, Rick; Zampatti,Brenton; Price, AminaPublished in:Canadian Journal of Fisheries and Aquatic Sciences

DOI:10.1139/cjfas-2018-0138

Published: 01/02/2020

Document VersionPublisher's PDF, also known as Version of record

Link to publication

Citation for published version (APA):Humphries, P., King, A., McCasker, N., Kopf, R. K., Stoffels, R., Zampatti, B., & Price, A. (2020). Riverscaperecruitment: A conceptual synthesis of drivers of fish recruitment in rivers. Canadian Journal of Fisheries andAquatic Sciences, 77(2), 213-225. https://doi.org/10.1139/cjfas-2018-0138

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal

Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Download date: 08. Nov. 2020

ARTICLE

Riverscape recruitment: a conceptual synthesis of drivers offish recruitment in riversPaul Humphries, Alison King, Nicole McCasker, R. Keller Kopf, Rick Stoffels, Brenton Zampatti,and Amina Price

Abstract: Most fish recruitment models consider only one or a few drivers in isolation, rarely include species’ traits, and havelimited relevance to riverine environments. Despite their diversity, riverine fishes share sufficient characteristics that predictionof recruitment should be possible. Here we synthesize the essential components of fish recruitment hypotheses and the keyfeatures of rivers to develop a model that predicts relative recruitment strength, for all fishes, in rivers under all flow conditions.The model proposes that interactions between flow and physical complexity will create locations in rivers, at mesoscales, whereenergy and nutrients are enriched. The resultant production of small prey will be concentrated and prey and fish larvae located(through dispersal or retention) so that the larvae can feed, grow, and recruit. Our synthesis explains how flow and physicalcomplexity affect fish recruitment and provides a conceptual basis to better conserve and manage riverine fishes globally.

Résumé : La plupart des modèles de recrutement de poissons n’intègrent qu’un seul ou quelques facteurs de manière isolée,incluent rarement des caractères des espèces et sont peu pertinents en ce qui concerne les milieux fluviaux. Malgré la diversitédes poissons de rivière, le nombre de caractéristiques qu’ils ont en commun devrait être suffisant pour permettre des prédictionsdu recrutement. Nous présentons une synthèse des éléments essentiels d’hypothèses relatives au recrutement de poissons et deséléments clés des rivières pour élaborer un modèle qui prédit la force relative du recrutement, pour tous les poissons et dans desrivières représentant toutes les conditions d’écoulement. Le modèle propose que les interactions entre l’écoulement et lacomplexité physique produisent des endroits dans les rivières, à des échelles intermédiaires, où il y a enrichissement en énergieet en nutriments, où la production en résultant de petites proies est concentrée et où les proies et larves de poisson sont situées(en raison de la dispersion ou de la rétention), de sorte que les larves peuvent s’y nourrir, croître et y être recrutées. Notresynthèse fournit une justification de l’influence de l’écoulement et de la complexité physique sur le recrutement de poissons,ainsi qu’un fondement conceptuel pour la conservation et la gestion améliorées des poissons de rivière en général. [Traduit parla Rédaction]

IntroductionUnderstanding what drives fluctuations in the size of fish pop-

ulations has been one of the dominant pursuits of fisheries scien-tists for more than a century (Hjort 1914; Houde and Hoyt 1987;Matthews 1998). This interest arose largely because of the desire tomanage stocks for sustainable commercial and recreational har-vesting in highly variable, unpredictable, and complex environ-ments (Ricker 1954; Lewin et al. 2006). Despite some progress, thegeneral consensus is that this goal has not been realized — mostmarine fish stocks have been fully exploited or degraded, and theexistence of sustainable fisheries is a rare exception (Caddy andSeijo 2005). The status of nonexploited fishes is no less dire(Hilborn et al. 2003; Jelks et al. 2008). Knowledge of fish popula-tion dynamics in fresh water generally lags far behind that inmarine environments (Schlosser and Angermeier 1995; Ziv et al.2012; Lintermans 2013; Cohen et al. 2016; McIntyre et al. 2016), the

exception being recreational freshwater sport fisheries (Lewinet al. 2006).

The increasing imperilment of freshwater fishes worldwide(Jelks et al. 2008; Froese and Pauly 2018) emphasizes the need tounderstand the fundamental processes influencing the size andcomposition of fish populations. Knowledge of what drives fishrecruitment is a key piece of the population dynamics puzzle, butis poorly understood (Chambers and Trippel 2012). Recruitmentcan be broadly defined as the number of individuals entering thepopulation at some well-defined age or stage class in a given year(e.g., Caley et al. 1996; Maceina and Pereira 2007; King et al. 2013).We define recruitment as the number of 0+ individuals enteringthe population each year, as measured at a time following hatch-ing, such that individuals have passed through the period whenthe majority of larval mortality has occurred (see King et al. 2013).This definition is quantitative and applies well to fishes of con-trasting life histories.

Received 3 April 2018. Accepted 1 May 2019.

P. Humphries, N. McCasker, and R.K. Kopf.† Institute for Land, Water, and Society, Charles Sturt University, Albury, NSW 2640, Australia.A. King.* Research Institute for the Environment and Livelihoods, School of Environment, Charles Darwin University, Darwin, NT 0909, Australia.R. Stoffels.† Commonwealth Scientific and Industrial Research Organisation (CSIRO), P.O. Box 821, Wodonga, VIC 3689, Australia.B. Zampatti.‡ Inland Waters and Catchment Ecology Program, South Australian Research and Development Institute (SARDI), Aquatic Sciences, P.O.Box 120, Henley Beach, SA 5022, Australia.A. Price. School of Life Sciences, La Trobe University, P.O. Box 821, Wodonga, VIC 3689, Australia.Corresponding author: Paul Humphries (e-mail: phumphries@csu.edu.au).*Present address: Centre for Freshwater Ecosystems, La Trobe University, P.O. Box 821, Wodonga, VIC 3689, Australia.†Present address: NIWA, P.O. Box 8602, Riccarton, Christchurch 8440, New Zealand.‡Present address: Commonwealth Scientific and Industrial Research Organisation (CSIRO), Locked bag No. 2, Glen Osmond, SA 5064, Australia.Copyright remains with the author(s) or their institution(s). This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0),which permits unrestricted use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.

213

Can. J. Fish. Aquat. Sci. 77: 213–225 (2020) dx.doi.org/10.1139/cjfas-2018-0138 Published at www.nrcresearchpress.com/cjfas on 18 June 2019.

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

It is axiomatic that recruitment is higher when the early lifestages of a fish coincide with conditions favorable for survival(Houde and Hoyt 1987; Houde 1989a; Cushing 1990). However,what constitutes favorable recruitment conditions is the subjectof much debate (see Chambers and Trippel 2012; King et al. 2013).The huge diversity of fish traits (e.g., morphology, life history,physiology, behavior) also makes generalizations concerning re-cruitment problematic (Winemiller and Rose 1993; Wootton1998). Despite this diversity of traits and environments in whichfishes live, the majority of fishes share enough characteristics tosuggest the potential for some common drivers that influencerecruitment strength (Chambers and Trippel 2012). The profusionof fish recruitment hypotheses and models and the doggedness offisheries scientists are a testament to this potential (Table 1).

For a young fish to develop, survive, and therefore recruit suc-cessfully, it must be located in an area of suitable water quality,find enough food of the appropriate size and type that meets itsnutritional requirements, and avoid being eaten. Importantly, thetraits of the progeny of different species (e.g., egg and larval size,number, and dispersal capability) will influence interactions withthese drivers (Winemiller and Rose 1993). However, most recruit-ment hypotheses have their origins in marine systems (Table 1;Chambers and Trippel 2012). Moreover, because many marinebony fishes are broadcast spawning, “periodic” species (sensuWinemiller and Rose 1992), comparative studies of recruitmentdrivers of species of different life history strategies are relativelyrare (Winemiller 2005).

The physical, chemical, and biological characteristics of riversare unlike those of marine ecosystems. Furthermore, riverinefishes from all life history strategies (opportunistic, periodic, andequilibrium) are common (Winemiller 1989; Winemiller and Rose1992; Humphries et al. 1999; Mims and Olden 2012). Thus, al-though the same requirements of all young fish must be met forthem to recruit, the manner in which riverine fish accomplishthis necessarily depends on ecosystem type and life history strat-egy (see Hoagstrom and Turner 2015). Several fish recruitmenthypotheses, mostly for marine environments, have attempted toexplain how young riverine fish coincide with appropriate condi-tions for recruitment in space and time (Humphries et al. 1999,2013; Hoagstrom and Turner 2015). These hypotheses, however,tend to be limited geographically, climatically, or phylogeneti-cally; to date, no integrated model has predicted recruitment forall types of riverine fishes under all flow conditions. Furthermore,despite the contribution that river ecosystem concepts haveplayed in shaping our current understanding of riverine patternsand processes (Vannote et al. 1980; Junk et al. 1989; Poff et al. 1997;Walker et al. 1995), they are rarely invoked when attempts have

been made to understand recruitment processes in riverine fish(but see Zeug and Winemiller 2008; Falke et al. 2010).

We propose that to understand the drivers of fish recruitmentin rivers, we need to consider how species traits during the earlylife stage interact with the physical, chemical, and biological fea-tures of rivers, and how these in turn are affected by flow andclimate. We develop the Riverscape Recruitment Synthesis Model(RRSM) to describe how flow and physical complexity of riversaffect fish recruitment, and provide a conceptual basis for gener-ating predictions that are relevant and useful for managementand research. This paper aims to (i) assess which existing fishrecruitment hypotheses are most relevant to riverine fishes; (ii)determine how flow mediates macroscale and mesoscale pro-cesses in rivers that are relevant to fish recruitment; (iii) describehow life history and other species traits influence interactions ofspecies with riverine patterns and processes relevant to recruit-ment, such as temperature, feeding, and predation; (iv) describehow processes associated with the “fundamental triad” (Bakun1998) — the recruitment model that we consider best encapsu-lates the spatial and temporal factors responsible for fish recruit-ment — likely operate in rivers; and (v) propose the RRSM, a modelthat conceptualizes our understanding of what drives recruit-ment in rivers.

Fish recruitment hypothesesFish recruitment hypotheses can be broadly grouped as those

that (i) are associated with matching fish larvae with a suitableprey environment, (ii) emphasize growth-mediated predatoravoidance, or (iii) highlight the importance of movement to, orretention in, nursery areas (see Table 1 and references therein).Most hypotheses assume that high densities of appropriatelysized prey are required for successful recruitment, and some sug-gest that high levels of predation can be avoided by rapid growth(i.e., “bigger is better”) or by dispersing to relatively benign nurs-ery areas (“coastal conveyer belt”). Recruitment hypotheses canalso be grouped into those suggesting that the interactionbetween larvae and environmental factors (largely food and pred-ators) are spatially related (e.g., “plankton contact”) versus tempo-rally related (e.g., ‘“window of opportunity”). Most food- andpredator-related hypotheses consider temporal coincidence of lar-vae with prey. A more limited subset of hypotheses (e.g., “aberrantdrift”) consider spatial coincidence of larvae with prey, withmovement being a central theme in several of these hypotheses.

We contend that Bakun’s fundamental triad hypothesis (Bakun1996, 1998, 2010) is the most comprehensive owing to its inclusionof several drivers (food, predator avoidance, movement–retention)

Table 1. Alignment of major fish recruitment hypotheses with key factors influencing survival, with the first columnindicating whether the interaction between the early stages of fish and these factors are spatially or temporallyrelated.

Matching suitable prey environmentwith critical early life stages

Size and growth-mediatedpredator avoidance

Movement (dispersaland retention)

Temporal Match–mismatcha Bigger is betterh

Critical periodb Stage durationi

Flood recruitment modelc Growth–selection predationg

Stable oceand Growth–predationg

Parental conditione Loophole strategiesg

Low-flow recruitmente

Window of opportunityf

Fundamental triadg

Spatial Plankton contacth Growth–mortalityj Coast conveyer beltj

Fundamental triadg Aberrant driftj

Member/vagrantk

Fundamental triadg

Note: Sources: a: Cushing 1990; b: Hjort 1914; c: Harris and Gehrke 1994; d: Lasker 1978; e: Humphries et al. 1999; f: Humphries et al.2013; g: Bakun 1998; h: Kerr and Dickie 2001; i: Francis 1994; j: Cushing 1975; k: Iles and Sinclair 1982; Sinclair and Iles 1988.

214 Can. J. Fish. Aquat. Sci. Vol. 77, 2020

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

thought to influence recruitment variability and its considerationof the spatial and temporal influence of these drivers across dif-ferent environments. The fundamental triad arose in an attemptto describe the drivers of recruitment for broadcast-spawning pe-lagic marine fishes (Bakun 1996, 1998, 2010). It incorporates threephysical processes that are likely to enhance recruitment: nutri-ent enrichment, food concentration, and processes that result inlarvae and their food being retained in, or moved to, suitablehabitat. It also connects these processes to their sources and at thescales at which they operate. For example, macroscale patterns(e.g., climate) influence the overall amounts of available energy(i.e., carbon) and nutrients (e.g., nitrogen and phosphorus), whilemesoscale patterns (e.g., water movement) influence where en-ergy and nutrients are incorporated into primary and secondaryproduction, and in turn how this translates into food for youngfish. Hoagstrom and Turner (2015) extended the concept into freshwater, illustrating the potential relevance of the fundamentaltriad for braided lowland rivers of the North American GreatPlains. In this ecoregion, broadcast-spawning fishes breed duringfloods, while recruitment is enhanced through the nutrient enrich-ment and concentration, as well as propagule retention, that oc-cur in connection with flood recession and slack-water habitats.Indeed, we propose that, with some modifications, Bakun’s fun-damental triad can be further developed into a more generic riv-erine fish recruitment model suitable for most river types andflow conditions (termed the Riverscape Recruitment SynthesisModel (RRSM); see below). But first, we explore the macroscalepatterns that drive production in rivers, the mesoscale patterns inrivers that provide the template on which fish recruitment acts,and the overriding role of flow in linking these two. We alsoconsider other key drivers of fish recruitment in rivers that arerarely incorporated into previous models (species traits, temper-ature, nutrient and food retention, and dispersal) before present-ing the new model.

Flow-mediated macroscale and mesoscale patternsand fish recruitment in rivers

Macroscale patternsRivers comprise a dendritic network of channels connected

by downstream flow; in some rivers, this network connects to amuch larger floodplain during flooding. Climate and geologylargely govern the predictability, variation, and shape of the flowregimes (Fig. 1). Flow in turn influences the physical (e.g., temper-ature) and chemical (e.g., salinity) characteristics of rivers; the

sources, storage, transformation, and transport of energy and nu-trients; and the transport of sediment, amongst other processes(Vannote et al. 1980; Junk et al. 1989; Walker et al. 1995; Thorpet al. 2008; Humphries et al. 2014). At macroscales (i.e., within ariver catchment), carbon (energy) and nutrient inputs are primar-ily influenced by flow and can be derived from headwaters(Vannote et al. 1980), the floodplain (Junk et al. 1989), or locallywithin the river channel (Thorp and Delong 1994) as a result ofnatural or anthropogenic processes (Kobayashi et al. 2009;Binckley et al. 2010; Hale et al. 2015). Regardless of the source ofcarbon, it is the magnitude, duration, and recurrence interval offlow events and the overall flow regime — interacting with thegeology and geomorphology of rivers — that determine the mag-nitude, timing and location of energy and nutrient inputs(Humphries et al. 2014). These basal resources drive primary andsecondary production, and ultimately, fish recruitment.

Interactions between macroscale and mesoscale patternsFlow regimes and flow events can be conceived of as waves that

progress from upstream to downstream. These waves interactwith geomorphic features and other types of structures in riversto create a diversity of flow-retention zones (such as eddies, slackwaters, and confluences; Fig. 1), influencing the transport andretention of energy and nutrients in rivers derived at macrospa-tial scales (Thorp et al. 2008; Humphries et al. 2014). For example,at the extreme trough of a wave, when a river ceases to flow, thereis very limited transport, resulting in high retention of energy andnutrients along small scales within the river; the vast majority ofenergy inputs and production occur locally, and flow-retentionzones are in effect disconnected lentic pools. At the other ex-treme, at the crest of a wave, during overbank flooding, theretends to be substantial transport of energy and nutrients longitu-dinally and laterally; for those rivers with substantial floodplains,flooding has a major input to river ecosystem metabolism, andthere is typically a diversity and large extent of floodplain flow-retention zones. For these rivers, during and after flooding, thefloodplain will act as the zone of greatest retention. At interme-diate positions on the river wave (e.g., within channel flows),longitudinal transport will tend to predominate over lateral trans-port of energy and nutrients, and there will be an intermediatediversity of flow-retention zones.

It is these flow-retention zones — whose type, features, andspatial location and distribution will change as flow changes —that we propose are equivalent to Bakun’s mesoscale (1–100 m2)

Fig. 1. Macroscale and mesoscale environmental patterns and the interactions between them along with how these environmental patternsare filtered by life history and other species traits, influence individual or population processes, and are mediated by dispersal–retention tocreate fish recruitment in rivers.

Humphries et al. 215

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

habitats within which recruitment processes of many riverinefishes take place. They are therefore a key component in ourRRSM.

Mesoscale patternsLarval fish interact with their prey, predators, and the physico-

chemical characteristics of their environment at the mesoscale(Fig. 1; Winemiller and Rose 1993; Schiemer et al. 2001; Fauschet al. 2002; King 2004b; Bakun 2010). This is, unsurprisingly, alsothe scale in rivers at which nutrients, algae, and zooplanktonproduction are related (Reynolds and Descey 1996; Reynolds 2000;Ning et al. 2010). The more physically complex (structural (e.g.,woody debris, vegetation), geomorphic, and hydraulic) a reach is(O’Neill and Thorp 2011), the longer the water residence time, andthe more eddies, dead zones, and slack waters that exist (Reynolds2000; Schiemer et al. 2001; Vietz et al. 2013). This, in turn, leads togreater accumulation of inorganic and organic matter (Pringleet al. 1988; Sheldon and Thoms 2006) and storage and transforma-tion of nutrients (Newbold 1992; Hein et al. 2005). It also leads togreater production of phytoplankton and zooplankton (Reckendorferet al. 1999; Reynolds 2000), finally resulting in an increased poten-tial for young fish to find enough food. Zones of high water resi-dence time also allow young fish that swim poorly an opportunityto be retained in this relatively benign rearing environment orfind faster currents for dispersal, if avoiding predators is neces-sary (Humphries et al. 1999; Schiemer et al. 2001; Schludermannet al. 2012; Lechner et al. 2014, 2018). Further discussion of the roleof physical complexity in retaining nutrients and other basal re-sources — such as phytoplankton and algae, the prey of fish larvae —as well as the fish larvae themselves is presented below.

These high-retention zones are fundamental to the survival oflarval fish that have little yolk reserve at hatch and must beginfeeding exogenously within hours to days (Fig. 2; Table 2;Schiemer et al. 2001). High concentrations of prey may be lessimportant for other fish that have large stores of yolk at hatch andthus theirownintrinsic sourceofnutrientenrichment (WinemillerandRose 1993; King 2005; Kaminskas and Humphries 2009), allowingthem to effectively bypass the larval stage (sensu Balon 1999).Therefore, because fishes differ enormously in their life historyand other species traits, the nature of, and requirements associ-ated with, enrichment, concentration, and retention–dispersalprocesses will vary among species. To appreciate the implicationsof this for recruitment, we will take a short detour and considerpatterns in life history and other species traits.

Species traits and recruitmentWinemiller and Rose (1992) developed a triangular life history

model, elaborating on Pianka’s r and K-selection continuum(Pianka 1970), which seeks to explain how traits such as longevity,age at maturity, fecundity, and size of offspring interact withenvironmental conditions to ensure the best chance for the sur-vival of young. This model applies widely to marine and freshwa-ter fishes (Winemiller 1989; Humphries et al. 1999; Winemiller2005; King et al. 2013; Mims and Olden 2013). Winemiller andRose’s model proposes three conceptual end points or groups offishes: opportunistic, periodic, and equilibrium (Table 2). The lar-vae of opportunistic and periodic species typically have smallamounts of yolk (consumed within hours to days) and are small,undeveloped, and tend to swim poorly when first feeding exoge-nously (Wolter and Arlinghaus 2004; Kopf et al. 2014). For exam-ple, the early-stage 5–6 mm larvae of silver perch (Bidyanusbidyanus), a periodic species, have critical swimming speeds lessthan 2 cm·s–1 when they first start feeding at only 1–2 days old(Kopf et al. 2014). Larvae of opportunistic species (e.g., Australiansmelt (Retropinna semoni) and Murray River rainbowfish (Melanotae-nia fluviatilis)) often occupy shallow habitats (King 2004b) and relyon dense zooplankton or meiofauna (e.g., rotifers, microcrusta-ceans) to get them through the transition phase from endogenousto exogenous feeding (King 2005; McCasker 2009). Larvae of peri-odic species, such as golden perch (Macquaria ambigua), feed ontiny zooplankton (e.g., planktonic rotifers and microcrustaceans)because they are themselves planktonic and are broadcast intothe water column, typically in one batch (Arumugam and Geddes1987). By contrast, the larvae of equilibrium species typically havelarge amounts of yolk, which they consume over many days; theyare large, well developed, and swim well when first feeding exog-enously. The larvae of Murray cod (Maccullochella peelii), for exam-ple, are 8–10 mm and typically 7–10 days old when first feeding(Kaminskas and Humphries 2009) and have critical swimmingspeeds at this time of 15–30 cm·s–1 (Kopf et al. 2014). First-feedingequilibrium species typically feed on larger meiofauna and earlyinstars of insects, such as benthic cladocerans and chironomids(King 2005; Kaminskas and Humphries 2009; Sauvanet et al. 2013).

For the early life stages of fishes in rivers, movement and thecapacity to disperse cannot be divorced from their life history andother species traits. This is because traits such as development offins, body size, and the amount of yolk will affect the capacity ofa species to move to, and (or) maintain its position in, a preferred

Fig. 2. Examples in rivers of sources of nutrient enrichment, locations for concentration of food, and habitats likely to enhance retention offood and propagules. All photographs and drawings are by P. Humphries, except for fish larvae (by Tim Kaminskas) and salmon carcass (byPeter Sircom Bromley, CC BY-SA 4.0; https://creativecommons.org/licenses/by-sa/4.0, via Wikimedia Commons).

216 Can. J. Fish. Aquat. Sci. Vol. 77, 2020

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

location (Kopf et al. 2014). Such traits also affect a species’ abilityto find food that may be patchily distributed (Winemiller andRose 1993; Nishimura and Hoshino 2009; Jørgensen et al. 2014).There is also variation within life history strategies in the amountand type of movement of the early life stages (Wolter and Arlinghaus2003). Most broadcast spawners are periodic species (Winemillerand Rose 1992), and so the drift of eggs and larvae of these speciesin river currents is the norm (Lechner et al. 2016). But drift oflarvae occurs for only a subset of opportunistic and equilibriumspecies (Humphries and King 2004; Lechner et al. 2016). Thesedifferences in life history strategy, together with the importanceof dispersal and retention processes as part of the fundamentaltriad, mean that our RRSM needs to include not only an axis of lifehistory strategy but also of movement and dispersal during earlylife.

TemperatureTemperature can have a strong influence on all processes affect-

ing the dynamics of fish populations, but it exerts a particularlystrong effect on recruitment. The breadth of a fish species’ funda-mental thermal niche (sensu Allen-Ankins and Stoffels 2017) isgenerally narrower during early life history than at the juvenilestage and early adulthood (Fuiman and Werner 2002; Portner andFarrell 2008). Furthermore, the relatively poor swimming capacityof larval fish leaves them less able to regulate body temperaturethrough active selection of thermal habitats (Wolter and Arlinghaus2004). Indeed, of all physical variables, temperature is thought tohave the greatest effect on survival of the early life stages of fishes(Houde 1989b). Significant interspecific variation exists in the fun-damental thermal niche of riverine fishes within the same assem-blage, including the larval stages of those fishes. For example, theeggs, alevins, parr, and smolt of Arctic char (Salvelinus alpinus) haveboth a narrower and “cooler” thermal niche than those of thebrown trout (Salmo trutta) (Elliott and Elliott 2010; Turschwell et al.2017). Within the Danube River, the fundamental thermal niches

for embryogenesis and larval development differ among browntrout, common nase (Chondrostoma nasus), and common roach(Rutilus rutilus), resulting in their being recruited in river seg-ments with different thermal regimes (Schiemer et al. 2004). Itfollows, therefore, that spatial and temporal variation in the ther-mal properties of the riverscape will likely have an effect on(i) intraspecific recruitment, and hence population dynamics, and(ii) determining interspecific variation in recruitment, and hencedynamics at the community level.

The temperature of aquatic habitat varies at multiple spatialand temporal scales in riverscapes (Caissie 2006). Variation ingeomorphology, topography, solar radiation, streambed compo-sition, groundwater exchange, and hydraulics, among other fac-tors, causes spatial heterogeneity in temperature along both thelateral and longitudinal dimensions of the riverscape. Laterally,thermal regimes of habitats vary among the lotic (flowing chan-nels) and lentic (e.g., oxbow lakes) habitats of the riverscape(Tonolla et al. 2010); among lentic water bodies on a floodplain(Tonolla et al. 2010; Stoffels et al. 2017); and among habitats withinthe channel itself, even on a scale of metres (Wawrzyniak et al.2013; Dugdale et al. 2015; Baldock et al. 2016). Longitudinally,strong thermal gradients occur on a scale of kilometres withinrivers (Caissie 2006; Fullerton et al. 2015; Stoffels et al. 2016;Allen-Ankins and Stoffels 2017). At yet larger spatial scales, varia-tion in both geomorphology and orientation (hence solar radia-tion) and the influence of snowmelt and groundwater generatesspatial heterogeneity in thermal regimes across rivers withincatchments (Lisi et al. 2015; Snyder et al. 2015; Eschbach et al.2017). Variation in climatic conditions among seasons and yearsalso generates temporal variation in the thermal states of aquatichabitats (Caissie 2006; Vatland et al. 2015). Temporal variation inriver flows within and across years further alters the thermalproperties of the riverscape. Temporal changes in flow can causechanges in hydrological connectivity as well as the thermal massof the water bodies in, and hydraulics of, the riverscape; in turn,

Table 2. Species traits and the most significant potential sources and location of processes associated with the fundamental triad relevant for fishrecruitment in rivers.

Life history strategy

Opportunistic Periodic Equilibrium

Species traitsSize at maturity Small Moderate to large Moderate to largeFecundity Low Very high LowParental care Possibly None OftenSize at hatching Tiny Small LargeYolk size Small Small LargeTime from hatching to first

feedingHours to days Hours to days Days to weeks

Size at first feeding Tiny Small LargeSwimming ability Poor Very poor Good

Fundamental triad processesNutrient enrichment Flooding (soil nutrients), vegetation

decomposition; snowmelt;litterfall; fish carcasses; eggs;terrestrial runoff; point sources;groundwater

Flooding (soil nutrients), vegetationdecomposition; snowmelt;litterfall; fish carcasses; eggs;terrestrial runoff; point sources;groundwater

Maternal sources, yolk

Concentration of nutrientsand food

Summer slack waters; hyporheicupwellings; retentive mid-channel and edge habitats;eddies; floodplain habitats;isolated in-channel water holes;downstream of dams;macrophyte beds; debris dams

Floodplains; slack waters duringflood recession

Yolk

Retention–dispersal of foodand fish larvae

Remain or disperse among slackwaters; disperse onto floodplains

Disperse onto floodplains; disperseto slack waters during floodrecession

Disperse downstream to variouslocations

Humphries et al. 217

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

this altered connectivity affects the thermal dynamics of that riv-erscape (Caissie 2006; van Vliet et al. 2011).

Thus the riverscape is a dynamic mosaic of thermal habitatsthat should interact with the narrow, species-specific thermalniches of larval fish to influence spatial and temporal variation inrecruitment. It follows that any conceptual synthesis of riverinefish recruitment must include the effects of temperature.

Enrichment, concentration, and retentionprocesses in rivers

Bakun’s fundamental triad identifies three physical processesthat are likely to enhance recruitment: nutrient enrichment, nu-trient and food concentration, and processes that result in larvaeand their food being retained in, or dispersed to, the same loca-tion (Bakun 1996, 1998, 2010). This concept and related processeswere developed for fishes in the marine environment; however,below we describe how, when, and where these processes may beapplicable to the riverine environment for inclusion in the RRSM.

Nutrient enrichmentPotential sources of nutrient enrichment in rivers are unlike

those in marine systems, and include flood-induced releases ofsoil nutrients (Baldwin and Mitchell 2000) and vegetation decom-position (Naiman and Décamps 1997; Watkins et al. 2011); runofffrom snowmelt (Townsend-Small et al. 2011); seasonal litterfall(Vannote et al. 1980); deposition of fish carcasses (Johnston et al.2004; Verspoor et al. 2011; Swain and Reynolds 2015); mass spawn-ing of vertebrate and invertebrate eggs (Näslund et al. 2015); ter-restrial runoff (Hale et al. 2015); point-source nutrient-enrichedwastewater effluent (Aratani et al. 2007); and groundwater intru-sions (Briody et al. 2016; Table 2; Fig. 2). For opportunistic andperiodic species that are small and need to feed soon after hatch-ing, these sources of enrichment would provide enhanced pri-mary production, which would in turn enhance the production ofbenthic and pelagic zooplankton (e.g., rotifers, copepods, clado-cerans), their principal food source (King 2005; Pease et al. 2006;Nunn et al. 2007). For equilibrium species, we contend that thenutrient enrichment process has largely been supplied to theyoung fish by their mother through the provision of a relativelylarge yolk sac. Of course, if the yolk is consumed before the youngfish reach the juvenile stage, then the same aforementioned po-tential sources of nutrient enrichment may also be important forequilibrium species.

ConcentrationNutrient and larval fish food concentration can occur in a range

of riverine mesoscale hydrogeomorphic patches and also as yolkfor some species (Fig. 2; Table 2), but life history strategy willinfluence the significance of each of these as a potential site ofconcentration. For opportunistic fish species — which can spawnover extended periods and need to find dense aggregations ofsmall prey on which to start feeding in the first few days of life —some of the most significant sites for nursery rearing or recruit-ment include summer slack waters and hyporheic or groundwa-ter upwellings (Humphries et al. 1999; King 2004a; Thorp et al.2006); retentive mid-channel (Reynolds and Descey 1996) or edgehabitats (Schiemer et al. 2001), especially when associated witheddies; floodplain habitats, such as disconnected oxbow lakes(Zeug and Winemiller 2008); and isolated in-channel water holes(Kerezsy et al. 2011). All these habitats occur where currents areslowed or still, living and nonliving organic material is deposited,and water temperatures are warm. For periodic species, whichtypically spawn over a shorter period than opportunistic species,and often in response to rises in flow (King et al. 2003; Hoagstromand Turner 2015), floodplains and off-channel habitats are themost likely locations for successful rearing owing to the higherconcentrations of nutrients and food in these habitats (Lake 1967;Winemiller and Jepsen 1998; De Lima and Araujo-Lima 2004;

Feyrer et al. 2006). But this would depend on the longevity ofwater in floodplain habitats and connection (or reconnection)with the main river channel to facilitate later dispersal. Slackwaters may also concentrate prey and provide good conditions forfeeding, growth, and survival for periodic species in specific situ-ations, such as following flood recession (Hoagstrom and Turner2015). Although larvae of both life history strategies need smallprey, opportunistic species tend to disperse more actively overshort distances to find food, whereas periodic species typicallydisperse more passively in the drift (Humphries et al. 1999; Kopfet al. 2014). Based on Winemiller and Rose’s (1993) model, tinylarvae of periodic species that are largely passive dispersing wouldrequire relatively denser and larger patches of food than oppor-tunistic species for successful feeding. As suggested above,recently inundated floodplains, even when food is patchily dis-tributed, would meet this requirement in many cases. For equilib-rium species, the mothers of the young fish have largely performedthe concentration process for them: relatively large yolk reserves,which are typical of such species, may be able to sustain the fishfor many days after hatching. It will only be once the yolk is usedup that suitable concentrations of prey will be required. But bythen, many of the food-related challenges associated with recruit-ment have been overcome, and the comparatively large juveniles,with wide mouth gapes, have access to a large range of inverte-brates as prey (King 2005; Kaminskas and Humphries 2009). Forthese species, mortality may be highest at the juvenile stage,when other demands such as finding a territory, competition forresources, and predator avoidance become critical (Houde 1994).

Retention and dispersalFish larvae being retained in, or dispersing to, prey-rich, me-

soscale areas are also likely to be critical to successful recruitmentin rivers (Fig. 2; Table 2). Poorly swimming fish larvae and theirzooplankton prey will be affected by similar physical forces inrivers and oceans. Specifically, the more retentive a river reach(defined as approximately two meander bends, or about 20 timesthe width of the channel; Gordon et al. 2004), the greater theopportunity for zooplankton prey to feed, grow, and reproduce,and the greater the opportunity for young fish to feed on thisconcentration of prey (Reynolds et al. 1991; Humphries et al. 1999;Reckendorfer et al. 1999; Schiemer et al. 2001). If food is concen-trated where fish hatch, then, all other things being equal, therewill be an advantage in those young fish being retained locally(King 2004a, 2004b); this would be especially advantageous foropportunistic species. Alternatively, if food is concentrated atsome distance from where fish hatch, there will be an advantagein dispersing to these more prey-rich locations and remainingthere (Zeug and Winemiller 2008); this would be especially advan-tageous for periodic species. This does not mean that all types ofhighly retentive reaches are conducive for recruitment (Mallen-Cooperand Zampatti 2018): weir pools, for example, create other problems,such as uniform depth and velocity, while also providing goodconditions for the predators of young fish (Bice et al. 2017). Infree-flowing rivers, retentiveness of reaches is a function of theinteraction between discharge and physical complexity and is rel-evant to all river types (Reynolds et al. 1991; Reckendorfer et al.1999; Schiemer et al. 2001; Vietz et al. 2013). For example, a riverwith multiple channels may have high overall physical complex-ity during moderate to high within-channel flows, because themultiple channels create a diversity of hydraulic and structuralconditions. But this complexity may be lost when the river re-cedes to only one channel as flows decline. Determining the rela-tionship between discharge and physical complexity of a reach iscritical when assessing reach retentiveness. This relationship willalso affect the extent and behaviour of hyporheic zones, and thusnutrient concentration and transformation. As a consequence ofthis retentiveness, the more physically complex a reach is, thegreater the hydraulic diversity, instream production, retardation

218 Can. J. Fish. Aquat. Sci. Vol. 77, 2020

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

and accumulation of organic and inorganic particles, and enrich-ment and concentration of nutrients and the shorter the nutrientspiral length, all which contribute to enhanced conditions for fishrecruitment (Pringle et al. 1988; Reynolds et al. 1991; Newbold1992; Vranovsky 1995; King 2004b; Sheldon and Thoms 2006). Ifthe larvae of opportunistic species are retained in summer slackwaters and the larvae of periodic species can disperse onto flood-plains, where high concentrations of prey and slow or still cur-rents occur, then little energy needs to be expended by larvae tomaintain their position and find food. For equilibrium species thathave a high concentration of food already available (as yolk that willlast for many days), dispersing downstream or upstream to suit-able juvenile rearing habitats — while avoiding concentrations ofconspecifics and predators — may instead be paramount.

We predict that the influence of physical complexity on thethree processes of enrichment, concentration, and retention inrivers is likely to be least at extreme low and high flows andgreatest at intermediate flows. At extreme low flows, and cease-to-flow conditions, regardless of the physical complexity, nutrientenrichment and concentration of potential prey can be enhancedthrough shrinking water volumes (Ning et al. 2010). During flood-ing, when water moves out onto a previously dry floodplain, nu-trient enrichment and concentration processes can be enhancedthrough the breakdown of vegetation and ponding, regardless ofphysical complexity (Naiman and Décamps 1997; Watkins et al.2011). At intermediate flows in a free-flowing river, the greatestcontrast in retentiveness would likely be between physically com-plex and simple reaches. Diverse hydraulic habitats, high nutrientenrichment and concentration, and high retention of prey andfish larvae are ideal conditions for fish recruitment, assumingpredator density does not overwhelm everything else.

The fundamental triad (Bakun 2010) does not directly addressthe effects of predation on fish recruitment, although there ismuch literature to suggest that predation is a significant factor(see Houde 2009). It does, however, add a caveat or “loophole

strategy,” which allows for good recruitment under relativelypoor feeding conditions if predators are few or absent; alterna-tively, it allows for poor recruitment under good feeding condi-tions if predators are abundant (also see Gido and Propst 2012). Wetherefore recognize the need to include predation as a factor inour RRSM.

The Riverscape Recruitment Synthesis Model(RRSM)

We developed the RRSM based on a synthesis of previous think-ing about fish recruitment drivers (largely from marine systems),riverine productivity and riverscape scaling considerations, andconcepts related to life history and other species traits for fish.The RRSM can be used to predict relative recruitment strengthunder various conditions for fish from the three life history strat-egies (Fig. 3, reading horizontally through the model). Climateand flow are assumed to be the most influential drivers of many ofthe processes that result in fish recruitment, leading the RRSM todescribe how, where, and when these processes might take placein riverscapes and how the different species traits (particularly lifehistory) interact with these processes to predict recruitmentstrength through the flow wave (reading vertically in the model).The main hypothesis of the RSSM proposes that interactions be-tween flow and physical complexity will create locations in rivers,at mesoscales, where energy and nutrients are enriched, resultingin the production of small prey being concentrated, and prey andfish larvae located (either through dispersal or retention) so thatthe larvae can feed, grow, and recruit.

The RRSM includes the following inputs as predictors of relativerecruitment strength (Fig. 3e): flow-related amount and sources ofenergy (Fig. 3a); processes associated with the fundamental triadand how they are influenced by flow and physical complexity(Fig. 3b); flow-related movement and dispersal of the young stagesof fish from different life history strategies (Fig. 3c); and temperature,

Fig. 3. Riverscape Recruitment Synthesis Model. The relationship between flow and (a) the amount and sources of energy in rivers; (b) thefundamental triad of nutrient enrichment and the concentration and retention of young fish and their food mediated by physical complexityof the river reach (solid line, low physical complexity; broken line, high physical complexity); (c) the potential for movement and dispersal ofeggs and young fish (note that no periodic species are included during low flow because of the uncertainty of spawning under theseconditions); and how this combines with (d) hypothesized relative levels of temperature (T), food (F), and predation (Pr) considered relevant forfish from opportunistic (O, blue), periodic (P, green), and equilibrium (E, orange) life history strategies. Using these inputs, the model aims topredict (e) relative recruitment strength for fish in low complexity (solid fish) and high complexity (hatched fish) reaches. A question mark (?)indicates uncertainty of breeding.

Humphries et al. 219

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

food, and predation (Fig. 3d). Predictions are made for equilibrium,periodic, and opportunistic life history strategies. Descriptions of theinputs are brief, as they have previously been discussed.

The model includes the following:

(a) Amount and sources of energy. As discharge increases frombase levels to overbank flooding, sources and amounts of en-ergy inputs to rivers are likely to change (Fig. 3a): at low flow,sources may be local, whereas as discharge increases, sourcesof energy from upstream and from low-lying floodplain hab-itats, such as benches and anabranches, may increase in im-portance (see Humphries et al. 2014). As discharge increases,the amount of energy in perennial river systems increases, asa greater area of a river is inundated and so more carbonbecomes available, and more material is transported longitu-dinally through the system (Vannote et al. 1980; Junk et al.1989). Once discharge is overbank, the floodplain proper willcontribute a large percentage of the energy into rivers (Junket al. 1989). In ephemeral systems, the amount of energy maybe high at low discharge, moderate at moderate discharge,and high again at high discharge, depending on antecedentconditions, sources of carbon, and the amount of shading andturbidity (Pringle et al. 1988).

(b) Fundamental triad processes and physical complexity. Nutri-ent enrichment, concentration of nutrients and the prey oflarval fish, and the retention of the same will vary with flowand reach physical complexity: they are greatest at low andflood flows and intermediate at within-channel discharge.Reach physical complexity will mediate the effects of dischargeon these processes, because it has a substantial influence onhydraulics, instream production, retardation of organic and in-organic particles, and nutrient spiral length (Fig. 3b).

(c) Movement and dispersal. For some riverine fishes, adults moveto locations conducive to the rearing of their young (De Lima andAraujo-Lima 2004). For others, adults spawn in one location andthe eggs and (or) larvae disperse, presumably to increase theirchances of encountering good conditions for recruitment(Lechner et al. 2016). Both types of movement will be mediatedby flow: low flow limits dispersal for species of all life historystrategies, whereas high flows and flooding provides the op-portunity for larger-scale dispersal. The scale of movementwill thus be influenced by the life history and life stage of thespecies, but also by how that movement is facilitated or im-peded by the prevailing hydrological environment (Lechneret al. 2018) (Fig. 3c).

(d) Temperature, food, and predation. Ideal recruitment con-ditions should occur when temperatures are optimal foregg and larval development and growth, food is abundant,and predation is low (Fig. 3d). Because of size, ability tomove, and how much parental care larvae receive, the rel-ative importance of biotic and abiotic factors in influenc-ing recruitment of species from the three life historystrategies will differ. Thus, we make separate predictionsof recruitment strength based on hypothesized tempera-ture, food, and predation conditions for each life historystrategy. The types and densities of food available will alsobe influenced by temperature and flow (Ning et al. 2010).We also differentiate between reaches with low and highphysical complexity because of the substantial influencewe suggest that this has on the fundamental triad pro-cesses.

(e) Relative recruitment strength. Using our understanding ofthe key processes influencing recruitment during differentriver flow conditions, we make predictions of the relativerecruitment strength of fish species from the three lifehistory strategies under three broad flows and for reacheswith low (solid fish) and high (hatched fish) physical com-plexity (Fig. 3e).

Our model predictions assume that

(i) at extreme low flows and extreme high flows, physicalcomplexity will have little effect on the components of thefundamental triad; however, available energy would likelybe lower in reaches with low complexity, because of lesspotential decomposable material and less surface area forprimary and secondary production;

(ii) at low flows, dispersal is limited but becomes increasinglyeasier as flows increase;

(iii) recruitment strength is maximized when temperature isoptimal for the eggs and larvae of a given species to growand develop;

(iv) recruitment strength is maximized when larvae coincidein time and space with high food density and low predatordensity; and

(v) based on the definitions of opportunistic, periodic, andequilibrium end-point life history strategies, we assumethat populations of opportunistic and periodic species willvary annually more than those of equilibrium species, andthat populations of the first two will rarely be limited bythe carrying capacity of the environment, whereas popu-lations of the latter will be limited by, and often at, thecarrying capacity of the environment (Winemiller 2005).

The RRSM allows the relative recruitment strength of fish spe-cies belonging to the three life history strategies to be predicted,regardless of climate and river type (e.g., temperate stochastic,temperate seasonal, tropical seasonal, arid stochastic), althoughwe need to take into account how low flows or flooding are cou-pled with temperatures conducive to enhanced productivity(Winemiller 2004). Antecedent conditions of flow and climate arealso known to influence fish recruitment strength (Stewart-Kosteret al. 2011; Balcombe et al. 2012; Beesley et al. 2014), but are notexplicitly included in the RRSM. We acknowledge that longer-termflow and more general climatic conditions, such as multiyeardrought, are likely to be important influences on recruitment dy-namics, but this level of complexity is beyond the scope of thecurrent model and should be developed as the model is tested andrefined. Nevertheless, we propose that the RRSM predictions aregeneralizable, since they are based on processes and conditionsassociated with levels of flow (and interactions with physical com-plexity of the river) and do not relate to particular flow regimes orreturn times of floods or low flows. Reliance or focus on flowregime has, in our opinion, allowed development of ecologicalconcepts in rivers (e.g., Vannote et al. 1980; Junk et al. 1989), butalso ultimately limited their generalizability. This is discussedlater in relation to predictions with the three broad-flow levelsconsidered in the RRSM.

Low-flow predictions (lower arrow, Fig. 3)In Mediterranean and wet–dry tropical and arid-zone rivers,

low-flow conditions are typically coupled with warmer tempera-tures that are conducive to food production and optimal fishgrowth (Humphries et al. 1999; Balcombe et al. 2012; King et al.2015). Irrespective of the physical complexity of reaches, theamount of energy available to fuel production is likely to be highduring lowflows,asnutrientsarebeingconcentratedbyshrinkinghab-itats. This will result in enhanced growth rates of planktonic al-gae, followed by rapid growth and turnover of zooplankton, andthus an increasing concentration of food for fish larvae. Low flowsand the abundance of shallow slack waters, isolated water holes,and highly retentive habitats more generally will mean that themovement of larger fish will be more restricted than under higher-flow conditions, but concentration of all organisms will increase asrivers contract, and thus predation should be moderate–high. Underthese conditions, we predict that the recruitment strength of equi-librium and opportunistic species would be moderate, because manyof the conditions (i.e., optimal temperatures, high food availability)are conducive to larval survival (Fig. 3e). Predation intensity should,

220 Can. J. Fish. Aquat. Sci. Vol. 77, 2020

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

however, be lower for equilibrium species, as parental care should pro-videadegreeofprotection.Thepresenceofpredatorsatmoderate–highdensities, density-dependent competition, or water-quality issuesother than temperature would reduce survival. Structural refugefrom predation may also influence recruitment strength duringlow-flow conditions. Under the same conditions, we predict thatrecruitment strength of periodic species would be zero–low because,in many cases, these species will not breed under such conditions(e.g., King et al. 2009; Zampatti and Leigh 2013). However, if fish dobreed, the lentic or low-flow conditions would mean that the eggsand larvae — which normally disperse either wholly or largelypassively — would tend to remain mostly near where they werespawned and thus likely end up in a narrower range of habitats thanif they had dispersed widely. Combining this with a poor swimmingability and small size — and therefore a need to find a lot of foodquickly — may create conditions that make survival extremely un-certain for the larvae of periodic species. For all life history strategies,the ability to disperse to new locations will be limited (Fig. 3c).

Intermediate-flow predictions (middle arrow, Fig. 3)Under intermediate, within-channel flows, different conditions

for recruitment strength are likely to prevail in river reaches withlow versus high physical complexity (Fig. 3b). The coupling (e.g.,tropical rivers) or otherwise (e.g., Mediterranean rivers) of inter-mediate flows with warm temperatures will likely be an impor-tant factor driving relative recruitment strength among the threeexemplar river types. In all river types, reaches with low complex-ity and therefore low retentive capacity may have high energy andnutrient loads, but there would be few places where enrichmentcould take place naturally, except perhaps during initial rises inflow. The poor retention of water, energy, and nutrients wouldalso translate to poor production, growth, and retention of fishlarvae and their food. Under these conditions, we predict thatthere would be relatively poor recruitment for periodic and op-portunistic species (Fig. 3e). However, those equilibrium speciesthat have parental care, and those with larvae that are relativelylarge and swim well, may be buffered against the overall relativelypoor conditions, and thus may experience moderate recruitment.

In contrast, for fish of all life history strategies, recruitmentstrength should be proportionately greater in reaches with highcomplexity than in structurally simple reaches. This is becausegreater physical complexity should increase the potential for nu-trient enrichment through downstream transport of nutrientsfrom areas higher in the catchment; increase runoff from previ-ously dry areas; increase the upwelling of groundwater andhyporheic water; and, equally if not more importantly, create agreater number and diversity of hydrogeomorphic zones (e.g.,slack waters, eddies, and dead zones) that would act as sites ofconcentration and retention of water, nutrients, fish larvae, andtheir food. Of the three life history strategies, recruitment shouldbe lowest for periodic species at intermediate flows. Althoughperiodic species may spawn during these conditions, unless thereare downstream lentic areas where small larvae that swim poorlycan passively find dense aggregations of appropriately sized prey,the likelihood is that mortality will be very high. In the case ofintermediate flows, opportunities for predators to move will begreater, and thus predation may play a significant role in affectingrelative recruitment strength. But such conditions also providethe potential for dispersal of actively swimming larval fish toavoid predators and competitors and to recruit to alternative lo-cations (Fig. 3c). Recruitment conditions at intermediate flows forthe early stages of fishes will vary depending on whether the riveris rising or falling and the rate of change. Free-flowing rivers willgenerally rise rapidly but fall more slowly and spend relativelylittle time at any one intermediate flow, whereas regulated riversmay remain at an intermediate flow for considerable periods.

Flood-flow predictions (upper arrow, Fig. 3)During flooding in reaches of both low and high physical com-

plexity for all river types, drifting eggs and larvae will be disperseddownstream and laterally. On floodplains, enrichment throughdecomposition of terrestrial plants and other organic matter andrelease of soil nutrients will occur, fueling the production of zoo-plankton (Watkins et al. 2011). As floodplain water bodies contractafter flooding, the concentration of nutrients and zooplanktonenhances feeding conditions for larval fish (Lake 1967; Winemillerand Jepsen 1998; De Lima and Araujo-Lima 2004; Feyrer et al.2006). Flooding in tropical rivers is predictably coupled withwarmer temperatures that are optimal for the spawning, growth,and development of early life stages (King et al. 2013); however,this coupling is less predictable in Mediterranean rivers (whereflooding commonly occurs from late winter to early summer) orarid-zone rivers (where flooding occurs stochastically, but can oc-cur throughout the year depending on prevailing weather condi-tions). Assuming there is a coupling of high flows and warmtemperatures, we predict that periodic species would have thestrongest relative recruitment strength of all life history strate-gies under these conditions (Fig. 3e). Large numbers of small eggsand larvae would be dispersed widely among a diversity of envi-ronments, a subset of which would provide ideal conditions forsurvival and growth (McGinley et al. 1987; Winemiller and Rose1993). Flood flow conditions will affect the recruitment of equilib-rium species differently from that of periodic species, as manyequilibrium species tend to breed predictably in the main chan-nel, regardless of flow (Humphries 2005). The eggs and larvae ofthose species receiving parental care generally require flowingwater or active intervention by the guarding male to provideenough oxygen, conditions that are more challenging on flood-plains than in the main channel. Recruitment of equilibrium spe-cies may also be lower than during low flows because the greaterrisk of predation and high current speeds in the main channelincrease the risk of premature washout from nests or disruptionto other male guarding practices (e.g., mouth brooding). Underflood flow conditions, we predict that recruitment strength foropportunistic species will be relatively high, for reasons that aresimilar to those for periodic species. The differences in relativerecruitment strength would probably be due to the lower fecun-dity of opportunistic species, which increases the risk of larvaeending up in unfavorable conditions. Flooding does not typicallycue opportunists to spawn (King et al. 2003), but it does allow thedispersal of larvae amongst floodplain wetlands (Tonkin et al.2008), which is favorable for recruitment overall.

Use of the RSSM for conservation andmanagement of riverine fishes

Our synthesis highlights some key factors and processes that wehope will contribute to not only an understanding of flow–ecology relationships, but also the conservation and manage-ment of riverine fishes. In recent decades, environmental flows ofvarious types have been proposed to enhance or preserve riverinefish populations, particularly in flow-altered rivers (Arthington2012). Although flow is a key driver for recruitment, our modelintegrates other important factors, such as temperature, physicalcomplexity, movement, predation, and species traits that interactwith flow to influence fish recruitment strength. The model cantherefore be used to explore the elements of a river (e.g., temper-ature, physical complexity, flow) that can potentially be manipu-lated or ameliorated by managers in regulated rivers or protectedin at-risk rivers. Predictions can readily be made and tested aboutwhich species, based on life history and knowledge of the physicalcomplexity of river reaches, would benefit and which could beimpaired by flow alteration.

Our synthesis also proposes the likely zones in rivers that arecritical for producing food for fish larvae, and suggests flow-

Humphries et al. 221

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

related mechanisms and pathways that are responsible for thegrowth and survival of young fishes. The synthesis highlights theimportance of physical complexity (hydraulic, geomorphic, andstructural) in rivers and its role in the retentive capacity ofreaches, which we predict will have major influence on recruit-ment, especially at intermediate, in-channel flows. It must be em-phasized, however, that higher retention alone is insufficient toguarantee successful recruitment because of other mitigating fac-tors, such as predation, lack of sheltering habitat and hydraulicdiversity, and the presence of alien species. We conclude — asmany others have before us — that there is a need for a diversityof flows, from low flows to flooding (Poff et al. 1997), and a diver-sity of hydraulic and physical complexity, to promote recruitmentfor riverine fish assemblages. Our synthesis now provides a solidrationale for why physical complexity is important for the earlieststages of fishes’ lives, whereas previously the focus of addingwoody debris, for example, to rivers has been on providing adulthabitat for fish.

We embarked on this synthesis hoping to evaluate the signifi-cance of factors affecting recruitment by drawing on multiyearand multispecies studies of riverine fishes. To our surprise, therewere no studies that we could find that simultaneously consid-ered the relationship between recruitment, life history, and othertraits of multiple fish species and reach physical complexity overa number of years under a range of flows. Such studies are desper-ately needed if we are to advance our understanding of riverinefish recruitment, particularly if we are to make informed deci-sions about environmental watering to improve fish populations.To reiterate our observation from the Introduction, fish recruit-ment research in fresh water lags behind that in marine environ-ments. Furthermore, most freshwater fish recruitment researchhas focused on single recreational species rather than fish assem-blages. Although solid foundations have been laid for understand-ing the relationships between flow and fish population dynamics(see e.g., Schlosser 1985; Schlosser and Angermeier 1995), andsome promising developments have followed (e.g., Falke et al.2010; Kerezsy et al. 2011; Balcombe et al. 2012; Hoagstrom andTurner 2015), researchers appear to be reluctant to posit recruit-ment models in rivers and then test them. Marrying the key prin-ciples of the riverscape concept (Fausch et al. 2002) and riverecosystem models (Vannote et al. 1980; Junk et al. 1989; Thorp andDelong 1994) with the drivers of fish recruitment would seem tobe a logical way forward.

AcknowledgementsThis review was funded by the Australian Department of Envi-

ronment and Energy’s Commonwealth Environmental Water Of-fice through the Murray–Darling Basin Environmental WaterKnowledge and Research project, administered by the Centre forFreshwater Ecosystems, La Trobe University. Additional supportfor the preparation of this manuscript came from Charles SturtUniversity, CSIRO, Charles Darwin University, La Trobe Univer-sity, and SARDI. The review benefited from discussions with Ste-phen Balcombe, John Koehn, and Lee Baumgartner.

ReferencesAllen-Ankins, S., and Stoffels, R.J. 2017. Contrasting fundamental and realized

niches: two fishes with similar thermal performance curves occupy differentthermal habitats. Freshwater Sci. 36(3): 635–652. doi:10.1086/693134.

Aratani, Y., Tajima, A., and Minamiyama, M. 2007. Relationship of nutrient andresidual chlorine concentration in treated wastewater with periphytic algaegrown in a stream receiving treated wastewater. In Water Science and Tech-nology. pp. 375–386.

Arthington, A.H. 2012. Environmental flows: saving rivers in the third millen-nium. University of California Press, Berkeley, California.

Arumugam, P.T., and Geddes, M.C. 1987. Feeding and growth of golden perchlarvae and fry (Macquaria ambigua Richardson). Trans. R. Soc. S. Aust. 111(1):59–65.

Bakun, A. 1996. Patterns in the ocean. California Sea Grant, in cooperation withCentro de Investigaciones Biologicas del Noroeste, La Paz, Mexico.

Bakun, A. 1998. Ocean triads and radical interdecadal variation: bane and boonto scientific fisheries management. In Reinventing fisheries management.Springer. pp. 331–358.

Bakun, A. 2010. Linking climate to population variability in marine ecosystemscharacterized by non-simple dynamics: conceptual templates and schematicconstructs. J. Mar. Syst. 79(3): 361–373. doi:10.1016/j.jmarsys.2008.12.008.

Balcombe, S.R., Lobegeiger, J.S., Marshall, S.M., Marshall, J.C., Ly, D., andJones, D.N. 2012. Fish body condition and recruitment success reflect ante-cedent flows in an Australian dryland river. Fish. Sci. 78(4): 841–847. doi:10.1007/s12562-012-0519-z.

Baldock, J.R., Armstrong, J.B., Schindler, D.E., and Carter, J.L. 2016. Juvenile cohosalmon track a seasonally shifting thermal mosaic across a river floodplain.Freshw. Biol. 61(9): 1454–1465. doi:10.1111/fwb.12784.

Baldwin, D.S., and Mitchell, A. 2000. The effects of drying and re-flooding on thesediment and soil nutrient dynamics of lowland river–floodplain systems: asynthesis. River Res. Appl. 16(5): 457–467.

Balon, E.K. 1999. Alternative ways to become a juvenile or a definitive phenotype(and on some persisting linguistic offenses). Environ. Biol. Fishes, 56: 17–38.

Beesley, L.S., Gwinn, D.C., Price, A., King, A.J., Gawne, B., Koehn, J.D., andNielsen, D.L. 2014. Juvenile fish response to wetland inundation: how ante-cedent conditions can inform environmental flow policies for native fish. J.Appl. Ecol. 51(6): 1613–1621. doi:10.1111/1365-2664.12342.

Bice, C.M., Gibbs, M.S., Kilsby, N.N., Mallen-Cooper, M., and Zampatti, B.P. 2017.Putting the “river” back into the Lower River Murray: quantifying the hy-draulic impact of river regulation to guide ecological restoration. Trans. R.Soc. South Aust. 141(2): 108–131. doi:10.1080/03721426.2017.1374909.

Binckley, C.A., Wipfli, M.S., Medhurst, R.B., Polivka, K., Hessburg, P.,Brion Salter, R., and Kill, J.Y. 2010. Ecoregion and land-use influence inverte-brate and detritus transport from headwater streams. Freshw. Biol. 55(6):1205–1218. doi:10.1111/j.1365-2427.2009.02344.x.

Briody, A.C., Cardenas, M.B., Shuai, P., Knappett, P.S.K., and Bennett, P.C. 2016.Groundwater flow, nutrient, and stable isotope dynamics in the parafluvial-hyporheic zone of the regulated Lower Colorado River (Texas, U.S.A.) over thecourse of a small flood. Hydrogeol. J.: 1-13.

Caddy, J.F., and Seijo, J.C. 2005. This is more difficult than we thought! Theresponsibility of scientists, managers and stakeholders to mitigate the un-sustainability of marine fisheries. Philos. Trans. R. Soc. B : Biol. Sci. 360(1453):59–75. doi:10.1098/rstb.2004.1567.

Caissie, D. 2006. The thermal regime of rivers: a review. Freshw. Biol. 51(8):1389–1406. doi:10.1111/j.1365-2427.2006.01597.x.

Caley, M.J., Carr, M.H., Hixon, M.A., Hughes, T.P., Jones, G.P., and Menge, B.A.1996. Recruitment and the local dynamics of open marine populations.Annu. Rev. Ecol. Syst. 27: 477–500. doi:10.1146/annurev.ecolsys.27.1.477.

Chambers, R.C., and Trippel, E. 2012. Early life history and recruitment in fishpopulations. Springer Science & Business Media.

Cohen, A.S., Gergurich, E.L., Kraemer, B.M., McGlue, M.M., McIntyre, P.B.,Russell, J.M., et al. 2016. Climate warming reduces fish production andbenthic habitat in Lake Tanganyika, one of the most biodiverse freshwa-ter ecosystems. Proc. Natl. Acad. Sci. 113(34): 9563–9568. doi:10.1073/pnas.1603237113. PMID:27503877.

Cushing, D.H. 1975. Marine ecology and fisheries. Cambridge University PressArchive, Cambridge.

Cushing, D.H. 1990. Plankton production and year-class strength in fish popula-tions: an update of the match/mismatch hypothesis. Adv. Mar. Biol. 26: 250–293. doi:10.1016/S0065-2881(08)60202-3.

De Lima, Ã.C., and Araujo-Lima, C.A.R.M. 2004. The distributions of larval andjuvenile fishes in Amazonian rivers of different nutrient status. FreshwaterBiol. 49(6): 787–800. doi:10.1111/j.1365-2427.2004.01228.x.

Dugdale, S.J., Bergeron, N.E., and St-Hilaire, A. 2015. Spatial distribution of ther-mal refuges analysed in relation to riverscape hydromorphology using air-borne thermal infrared imagery. Remote Sens. Environ. 160: 43–55. doi:10.1016/j.rse.2014.12.021.

Elliott, J.M., and Elliott, J.A. 2010. Temperature requirements of Atlantic salmonSalmo salar, brown trout Salmo trutta and Arctic charr Salvelinus alpinus: pre-dicting the effects of climate change. J. Fish Biol. 77(8): 1793–1817. doi:10.1111/j.1095-8649.2010.02762.x. PMID:21078091.

Eschbach, D., Piasny, G., Schmitt, L., Pfister, L., Grussenmeyer, P., Koehl, M., et al.2017. Thermal-infrared remote sensing of surface water–groundwater ex-changes in a restored anastomosing channel (Upper Rhine River, France).Hydrol. Process. 31(5): 1113–1124. doi:10.1002/hyp.11100.

Falke, J.A., Bestgen, K.R., and Fausch, K.D. 2010. Streamflow reductions andhabitat drying affect growth, survival, and recruitment of brassy minnowacross a Great Plains riverscape. Trans. Am. Fish. Soc. 139(5): 1566–1583.doi:10.1577/T09-143.1.

Fausch, K.D., Torgersen, C.E., Baxter, C.V., and Li, H.W. 2002. Landscapes toriverscapes: Bridging the gap between research and conservation ofstream fishes. BioScience, 52(6): 483–498. doi:10.1641/0006-3568(2002)052[0483:LTRBTG]2.0.CO;2.

Feyrer, F., Sommer, T., and Harrell, W. 2006. Managing floodplain inundationfor native fish: Production dynamics of age-0 splittail (Pogonichthys macrolepi-dotus) in California’s Yolo Bypass. Hydrobiologia 573(1): 213–226.

Francis, M.P. 1994. Duration of larval and spawning periods in Pagrus auratus

222 Can. J. Fish. Aquat. Sci. Vol. 77, 2020

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

(Sparidae) determined from otolith daily increments. Environ. Biol. Fish.39(2): 137–152. doi:10.1007/BF00004931.

Froese, R., and Pauly, D. (Editors). 2018. FishBase. World Wide Web electronicpublication [online]. Available from www.fishbase.org, version (02/2018).

Fuiman, L.A., and Werner, R.G. 2002. Fishery science: the unique contributionsof early life stages. Blackwell, Oxford.

Fullerton, A.H., Torgersen, C.E., Lawler, J.J., Faux, R.N., Steel, E.A., Beechie, T.J.,et al. 2015. Rethinking the longitudinal stream temperature paradigm:region-wide comparison of thermal infrared imagery reveals unexpectedcomplexity of river temperatures. Hydrol. Process. 29(22): 4719–4737. doi:10.1002/hyp.10506.

Gido, K.B., and Propst, D.L. 2012. Long-term dynamics of native and nonnativefishes in the San Juan River, New Mexico and Utah, under a partially managedflow regime. Trans. Am. Fish. Soc.141(3): 645–659. doi:10.1080/00028487.2012.683471.

Gordon, N.D., McMahon, T.A., and Finlayson, B.L. 2004. Stream hydrology: anintroduction for ecologists. John Wiley and Sons.

Hale, R.L., Grimm, N.B., Vörösmarty, C.J., and Fekete, B. 2015. Nitrogen andphosphorus fluxes from watersheds of the northeast U.S. from 1930 to 2000:Role of anthropogenic nutrient inputs, infrastructure, and runoff. GlobalBiogeochem. Cycles, 29(3): 341–356. doi:10.1002/2014GB004909.

Harris, J.H., and Gehrke, P.C. 1994. Development of predictive models linkingfish population recruitment with streamflow. Proceedings of the AustralianSociety for Fish Biology Workshop. In Population Dynamics for FisheriesManagement. Edited by D.A. Hancock. Bureau of Rural Resources Proceedings,Canberra. pp. 195–199.

Hein, T., Reckendorfer, W., Thorp, J.H., and Schiemer, F. 2005. The role of slack-water areas for biogeochemical processes in rehabilitated river corridors:examples from the Danube. Arch. Fur Hydrobiol. Suppl. Large Rivers, 15:425–442.

Hilborn, R., Branch, T.A., Ernst, B., Magnussson, A., Minte-Vera, C.V., Scheuerell,M.D., and Valero, J.L. 2003. State of the world’s fisheries. In Annual Review ofEnvironment and Resources. pp. 359–399.

Hjort, J. 1914. Fluctuations in the great fisheries of northern Europe. Rapport duConseil d’Exploration de Mer, 20: 1–228.

Hoagstrom, C.W., and Turner, T.F. 2015. Recruitment ecology of pelagic-broadcast spawning minnows: Paradigms from the ocean advance scienceand conservation of an imperilled freshwater fauna. Fish Fish. 16(2): 282–299.doi:10.1111/faf.12054.

Houde, E.D. 1989a. Subtleties and episodes in the early life of fishes. J. Fish Biol.35: 29–38.

Houde, E.D. 1989b. Comparative growth, mortality, and energetics of marinefish larvae: temperature and implied latitudinal effects. Fish. Bull. 87(3):471–495.

Houde, E.D. 1994. Differences between marine and freshwater fish larvae: im-plications for recruitment. ICES J. Mar. Sci. 51(1): 91–97. doi:10.1006/jmsc.1994.1008.

Houde, E.D. 2009. Recruitment variability. In Fish Reproductive Biology: Impli-cations for Assessment and Management. Edited by M. J. Fogarty, T. M. Jako-bsen, and B. A. Megrey. Wiley, New York. pp. 91–171.

Houde, E.D., and Hoyt, R. 1987. Fish early life dynamics and recruitment vari-ability. Trans. Am. Fish. Soc. Symp. 2: 17–29.

Humphries, P. 2005. Spawning time and early life history of Murray cod, Mac-cullochella peelii (Mitchell) in an Australian river. Environ. Biol. Fishes, 72(4):393–407. doi:10.1007/s10641-004-2596-z.

Humphries, P., and King, A.J. 2004. Drifting fish larvae in Murray-Darling Basinrivers: composition, spatial and temporal patterns and distance drifted. InDownstream movement of fish in the Murray-Darling Basin. Statement, rec-ommendations and supporting papers from a workshop held in Canberra3–4 June 2003. Edited by M. Lintermans and B. Phillips. Murray–Darling BasinCommission, Canberra. pp. 51–58.

Humphries, P., King, A.J., and Koehn, J.D. 1999. Fish, Flows and Flood Plains:Links between Freshwater Fishes and their Environment in the Murray–Darling River System, Australia. Environ. Biol. Fishes, 56(1): 129–151. doi:10.1023/A:1007536009916.

Humphries, P., Richardson, A., Wilson, G., and Ellison, T. 2013. River regulationand recruitment in a protracted-spawning riverine fish. Ecol. Appl. 23(1):208–225. doi:10.1890/11-2255.1. PMID:23495647.

Humphries, P., Keckeis, H., and Finlayson, B. 2014. The River Wave Concept:Integrating River Ecosystem Models. BioScience, 64(10): 870–882. doi:10.1093/biosci/biu130.

Iles, T.D., and Sinclair, M. 1982. Atlantic herring: stock discreteness and abun-dance. Science, 215(4533): 627–633. doi:10.1126/science.215.4533.627. PMID:17842372.

Jelks, H.L., Walsh, S.J., Burkhead, N.M., Contreras-Balderas, S., Diaz-Pardo, E.,Hendrickson, DA, et al. 2008. Conservation status of imperiled North Amer-ican freshwater and diadromous fishes. Fisheries, 33: 372–407. doi:10.1577/1548-8446-33.8.372.

Johnston, N., MacIsaac, E., Tschaplinski, P., and Hall, K. 2004. Effects of theabundance of spawning sockeye salmon (Oncorhynchus nerka) on nutrientsand algal biomass in forested streams. Can. J. Fish. Aquat. Sci. 61(3): 384–403.doi:10.1139/f03-172.

Jørgensen, C., Opdal, A.F., and Fiksen, Ø. 2014. Can behavioural ecology unite

hypotheses for fish recruitment? ICES J. Mar. Sci. 71(4): 909–917. doi:10.1093/icesjms/fst083.

Junk, W.J., Bayley, P.B., and Sparks, R.E. 1989. The flood pulse concept in river-floodplain systems. Can. Spec. Publ. Fish. Aquat. Sci. 106: 110–127.

Kaminskas, S., and Humphries, P. 2009. Diet of Murray cod (Maccullochella peeliipeelii) (Mitchell) larvae in an Australian lowland river in low flow and highflow years. Hydrobiologia, 636(1): 449–461. doi:10.1007/s10750-009-9973-8.

Kerezsy, A., Balcombe, S.R., Arthington, A.H., and Bunn, S.E. 2011. Continuousrecruitment underpins fish persistence in the arid rivers of far-western Queens-land, Australia. Mar. Freshw. Res. 62(10): 1178–1190. doi:10.1071/MF11021.

Kerr, S.R., and Dickie, L.M. 2001. The biomass spectrum: a predator–prey theoryof aquatic production. Columbia University Press, Cambridge.

King, A.J. 2004a. Density and distribution of potential prey for larval fish in themain channel of a floodplain river: Pelagic versus epibenthic meiofauna.River Res. Appl. 20(8): 883–897. doi:10.1002/rra.805.

King, A.J. 2004b. Ontogenetic patterns of habitat use by fishes within the mainchannel of an Australian floodplain river. J. Fish Biol. 65(6): 1582–1603. doi:10.1111/j.0022-1112.2004.00567.x.

King, A.J. 2005. Ontogenetic dietary shifts of fishes in an Australian floodplainriver. Mar. Freshw. Res. 56(2): 215–225. doi:10.1071/MF04117.

King, A.J., Humphries, P., and Lake, P.S. 2003. Fish recruitment on floodplains:The roles of patterns of flooding and life history characteristics. Can. J. Fish.Aquat. Sci. 60(7): 773–786. doi:10.1139/f03-057.

King, A.J., Tonkin, Z., and Mahoney, J. 2009. Environmental flow enhances na-tive fish spawning and recruitment in the Murray River, Australia. River Res.Appl. 25(10): 1205–1218.

King, A.J., Humphries, P., and McCasker, N.G. 2013. Reproduction and early lifehistory. In Ecology of Australian Freshwater Fishes. Edited by P. Humphriesand K.F. Walker. CSIRO Publishing, Collingwood. pp. 159–192.

King, A.J., Townsend, S.A., Douglas, M.M., and Kennard, M.J. 2015. Implicationsof water extraction on the low-flow hydrology and ecology of tropical savan-nah rivers: an appraisal for northern Australia. Freshw. Sci. 34(2): 741–758.doi:10.1086/681302.

Kobayashi, Y., Kim, C., Yoshimizu, C., Kohzu, A., Tayasu, I., and Nagata, T. 2009.Longitudinal changes in bacterial community composition in river epilithicbiofilms: Influence of nutrients and organic matter. Aquat. Microb. Ecol.54(2): 135–152. doi:10.3354/ame01258.

Kopf, S., Humphries, P., and Watts, R. 2014. Ontogeny of critical and prolongedswimming performance for the larvae of six Australian freshwater fish spe-cies. J. Fish Biol. 84(6): 1820–1841. doi:10.1111/jfb.12399. PMID:24814314.

Lake, J.S. 1967. Rearing experiments with five species of Australian freshwaterfishes. I. Inducement to spawning. Aust. J. Mar. Freshw. Res. 18(2): 137–154.doi:10.1071/MF9670137.

Lasker, R. 1978. Factors contributing to the variable recruitment of the northernanchovy (Engraulis mordax) in the Californian current: contrasting years 1975–1978. Rapp. P.-V. Réun. Cons. Int. Explor. Sci. Mer, 178: 375–388.

Lechner, A., Keckeis, H., Schludermann, E., Loisl, F., Humphries, P., Glas, M.,Tritthart, M., and Habersack, H. 2014. Shoreline configurations affect disper-sal patterns of fish larvae in a large river. ICES J. Mar. Sci. 71(4): 930–942.doi:10.1093/icesjms/fst139.

Lechner, A., Keckeis, H., and Humphries, P. 2016. Patterns and processes in thedrift of early developmental stages of fish in rivers: a review. Rev. Fish Biol.Fish. 26(3): 471–489. doi:10.1007/s11160-016-9437-y.

Lechner, A., Keckeis, H., Glas, M., Tritthart, M., Habersack, H., Andorfer, L., andHumphries, P. 2018. The influence of discharge, current speed and develop-ment on the downstream dispersal of larval nase (Chondrostoma nasus) in theRiver Danube. Can. J. Fish. Aquat. Sci. 75(2): 247–259. doi:cjfas-2016-0340.

Lewin, W.-C., Arlinghaus, R., and Mehner, T. 2006. Documented and PotentialBiological Impacts of Recreational Fishing: Insights for Management andConservation. Rev. Fish. Sci. 14(4): 305–367. doi:10.1080/10641260600886455.

Lintermans, M. 2013. Conservation and management. In Ecology of Australianfreshwater fishes. Edited by P. Humphries and K. Walker. CSIRO Publishing,Collingwood. pp. 282–316.

Lisi, P.J., Schindler, D.E., Cline, T.J., Scheuerell, M.D., and Walsh, P.B. 2015. Wa-tershed geomorphology and snowmelt control stream thermal sensitivity to airtemperature. Geophys. Res. Lett. 42(9): 3380–3388. doi:10.1002/2015GL064083.

Maceina, M., and Pereira, D.L. 2007. Recruitment. In Analysis and Interpretationof Freshwater Fisheries Data. Edited by C.S. Guy and M.L. Brown. AmericanFisheries Society, Bethesda, Maryland. pp. 121–185.

Mallen-Cooper, M., and Zampatti, B.P. 2018. History, hydrology and hydraulics:Rethinking the ecological management of large rivers. Ecohydrology, 11(5):e1965. doi:10.1002/eco.1965.

Matthews, W.J. 1998. Patterns in Freshwater Fish Ecology. Kluwer AcademicPublishers, Norewll, Massachusetts.

McCasker, N.G. 2009. Of life and death in a lowland river: investigating mortalityduring the early life stages of Murray–Darling fishes. Ph.D. thesis, School ofEnvironmental Sciences, Charles Sturt University, Albury.

McGinley, M.A., Temme, D.H., and Geber, M.A. 1987. Parental investment inoffspring in variable environments: theoretical and empirical consider-ations. Am. Nat. 130(3): 370–398. doi:10.1086/284716.

McIntyre, P.B., Liermann, C.A.R., and Revenga, C. 2016. Linking freshwater fish-ery management to global food security and biodiversity conservation. Proc.

Humphries et al. 223

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

Natl. Acad. Sci. 113(45): 12880–12885. doi:10.1073/pnas.1521540113. PMID:27791055.

Mims, M.C., and Olden, J.D. 2012. Life history theory predicts fish assemblageresponse to hydrologic regimes. Ecology, 93(1): 35–45. doi:10.1890/11-0370.1.PMID:22486085.

Mims, M.C., and Olden, J.D. 2013. Fish assemblages respond to altered flowregimes via ecological filtering of life history strategies. Freshw. Biol. 58(1):50–62. doi:10.1111/fwb.12037.

Naiman, R.J., and Décamps, H. 1997. The ecology of interfaces: riparian zones.Annu. Rev. Ecol. Syst. 28(1): 621–658. doi:10.1146/annurev.ecolsys.28.1.621.

Näslund, J., Aldvén, D., and Závorka, L. 2015. Eggs from anadromous adultsprovide marine-derived nutrients to Atlantic salmon and brown trout parr inlate autumn — observations from a Swedish coastal stream. Environ. Biol.Fish. 98(12): 2305–2313. doi:10.1007/s10641-015-0436-y.

Newbold, J.D. 1992. Cycles and spirals of nutrients. In The rivers handbook. Vol.1. Edited by P. Callow, and G. Petts. Blackwell Publising, Oxford. pp. 379–408.

Ning, N.S.P., Nielsen, D.L., Paul, W.L., Hillman, T.J., and Suter, P.J. 2010. Micro-invertebrate dynamics in riverine slackwater and mid-channel habitats inrelation to physico-chemical parameters and food availability. River Res.Appl. 26(3): 279–296. doi:10.1002/rra.1266.

Nishimura, K., and Hoshino, N. 2009. The evolutionary pattern of early lifehistory in water currents. Evol. Ecol. 23(2): 207–221. doi:10.1007/s10682-007-9213-5.

Nunn, A., Harvey, J., and Cowx, I. 2007. The food and feeding relationships oflarval and 0+ year juvenile fishes in lowland rivers and connected waterbod-ies. I. Ontogenetic shifts and interspecific diet similarity. J. Fish Biol. 70(3):726–742. doi:10.1111/j.1095-8649.2007.01334.x.

O’Neill, B.J., and Thorp, J.H. 2011. A simple channel complexity metric for ana-lyzing river ecosystem responses. River Syst. 19(4): 327–335. doi:10.1127/1868-5749/2011/0042.

Pease, A.A., Davis, J.J., Edwards, M.S., and Turner, T.F. 2006. Habitat and resourceuse by larval and juvenile fishes in an arid-land river (Rio Grande, NewMexico). Freshw. Biol. 51(3): 475–486. doi:10.1111/j.1365-2427.2005.01506.x.

Pianka, E.R. 1970. On r- and K-selection. The American Naturalist, 104: 592–597.Poff, N.L., Allan, J.D., Bain, M.B., Karr, J.R., Prestegaard, K.L., Richter, B.D.,

Sparks, R.E., and Stromberg, J.C. 1997. The natural flow regime: a paradigmfor river conservation and restoration. BioScience, 47(11): 769–784. doi:10.2307/1313099.

Pörtner, H.O., and Farrell, A.P. 2008. Physiology and climate change. Science,322: 690–692. doi:10.1126/science.1163156.

Pringle, C.M., Naiman, R.J., Bretschko, G., Karr, J.R., Oswood, M.W., Webster, J.R.,et al. 1988. Patch dynamics in lotic ecosystems: the stream as a mosaic. J.North Am. Benthol. Soc. 7(4): 503–524. doi:10.2307/1467303.

Reckendorfer, W., Keckeis, H., Winkler, G., and Schiemer, F. 1999. Zooplanktonabundance in the River Danube, Austria: the significance of inshore reten-tion. Freshwater Biol. 41(3): 583–591. doi:10.1046/j.1365-2427.1999.00412.x.

Reynolds, C.S. 2000. Hydroecology of river plankton: the role of variability inchannel flow. Hydrol. Process. 14(16–17): 3119–3132. doi:10.1002/1099-1085(200011/12)14:16/17<3119::AID-HYP137>3.0.CO;2-6.

Reynolds, C.S., and Descey, J.P. 1996. The production, biomass and structure ofphytoplankton in large rivers. Arch. Hydrobiol. Suppl. 113: 161–187.

Reynolds, C.S., Carling, P., and Beven, K. 1991. Flow in river channels: newinsights into hydraulic retention. Arch. Hydrobiol. 121(2): 171–179.

Ricker, W.E. 1954. Stock and recruitment. J. Fish. Res. Board Can. 11(5): 559–623.doi:10.1139/f54-039.

Sauvanet, J., Bourdier, G., Jouve, L., Bec, A., and Desvilettes, C. 2013. Feeding ofpike larvae (Esox lucius L.) in an alluvial river backwater: fatty acid as markersof two organic matter flows. Arch. Hydrobiol. 183(4): 337–350. doi:10.1127/1863-9135/2013/0440.

Schiemer, F., Keckeis, H., Reckendorfer, W., and Winkler, G. 2001. The “inshoreretention concept” and its significance for large rivers. Arch. Hydrobiol.Suppl. 135: 509–516.

Schiemer, F., Keckeis, H., Nemeschkal, H., Schludermann, E., Winkler, G., andZweimuler, I. 2004. Ontogenetic patterns in thermal adaptation of fish vs.long-term temperature trends in large rivers. In International Congress Se-ries. Elsevier, London. pp. 209–217.

Schlosser, I.J. 1985. Flow regime, juvenile abundance, and the assemblage struc-ture of stream fishes. Ecology, 66(5): 1484–1490. doi:10.2307/1938011.

Schlosser, I.J., and Angermeier, P.L. 1995. Spatial variability in demographicprocesses in lotic fishes: conceptual models, empirical evidence and impli-cations for conservation. Am. Fish. Soc. Symp. 17: 360–370.

Schludermann, E., Tritthart, M., Humphries, P., and Keckeis, H. 2012. Dispersaland retention of larval fish in a potential nursery habitat of a large temperateriver: an experimental study. Can. J. Fish. Aquat. Sci. 69(8): 1302–1315. doi:10.1139/f2012-061.

Sheldon, F., and Thoms, M. 2006. In-channel geomorphic complexity: The key tothe dynamics of organic matter in large dryland rivers? Geomorphology,77(3): 270–285. doi:10.1016/j.geomorph.2006.01.027.

Sinclair, M., and Iles, T.D. 1988. Population richness of marine fish species.Aquat. Living Resour. 1(1): 71–83. doi:10.1051/alr:1988009.

Snyder, C.D., Hitt, N.P., and Young, J.A. 2015. Accounting for groundwater instream fish thermal habitat responses to climate change. Ecol. Appl. 25(5):1397–1419. doi:10.1890/14-1354.1. PMID:26485964.

Stewart-Koster, B., Olden, J.D., Kennard, M.J., Pusey, B.J., Boone, E.L.,Douglas, M., and Jackson, S. 2011. Fish response to the temporal hierarchy ofthe natural flow regime in the Daly River, northern Australia. J. Fish Biol.79(6): 1525–1544. doi:10.1111/j.1095-8649.2011.03072.x. PMID:22136238.

Stoffels, R.J., Richardson, A.J., Vogel, M.T., Coates, S.P., and Müller, W.J. 2016.What do metabolic rates tell us about thermal niches? Mechanisms drivingcrayfish distributions along an altitudinal gradient. Oecologia, 180(1): 45–54.PMID:26440800.

Stoffels, R., Weatherman, K., and Allen-Ankins, S. 2017. Heat and hypoxia give aglobal invader, Gambusia holbrooki, the edge over a threatened endemic fishonAustralianfloodplains.Biol. Invasions,19(8):2477–2489.doi:10.1007/s10530-017-1457-6.

Swain, N.R., and Reynolds, J.D. 2015. Effects of salmon-derived nutrients andhabitat characteristics on population densities of stream-resident sculpins.PLoS ONE, 10(6): e0116090. doi:10.1371/journal.pone.0116090.

Thorp, J.H., and Delong, M.D. 1994. The riverine productivity model: an heuristicview of carbon sources and organic processing in large river ecosystems.Oikos, 70(2): 305–308. doi:10.2307/3545642.

Thorp, J.H., Thoms, M.C., and Delong, M.D. 2006. The riverine ecosystem synthe-sis: Biocomplexity in river networks across space and time. River Res. Appl.22(2): 123–147. doi:10.1002/rra.901.

Thorp, J.H., Thoms, M.C., and Delong, M.D. 2008. The Riverine Ecosystem Syn-thesis. Elsevier, London.

Tonkin, Z.D., King, A.J., and Mahoney, J. 2008. Effects of flooding on recruitmentand dispersal of the Southern Pygmy Perch (Nannoperca australis) at a MurrayRiver floodplain wetland. Ecol. Manage. Restor. 9(3): 196–201. doi:10.1111/j.1442-8903.2008.00418.x.

Tonolla, D., Acuña, V., Uehlinger, U., Frank, T., and Tockner, K. 2010. ThermalHeterogeneity in River Floodplains. Ecosystems, 13(5): 727–740. doi:10.1007/s10021-010-9350-5.

Townsend-Small, A., McClelland, J.W., Holmes, R.M., and Peterson, B.J. 2011.Seasonal and hydrologic drivers of dissolved organic matter and nutrients inthe upper Kuparuk River, Alaskan Arctic. Biogeochemistry, 103(1): 109–124.doi:10.1007/s10533-010-9451-4.

Turschwell, M.P., Balcombe, S.R., Steel, E.A., Sheldon, F., and Peterson, E.E. 2017.Thermal habitat restricts patterns of occurrence in multiple life-stages of aheadwater fish. Freshw. Sci. 36(2): 402–414. doi:10.1086/691553.

van Vliet, M.T.H., Ludwig, F., Zwolsman, J.J.G., Weedon, G.P., and Kabat, P. 2011.Global river temperatures and sensitivity to atmospheric warming andchanges in river flow. Water Resour. Res. 47: 19. doi:10.1029/2010WR009198.

Vannote, R.L., Minshall, G.W., Cummins, K.W., Sedell, J.R., and Cushing, C.E.1980. The river continuum concept. Can. J. Fish. Aquat. Sci. 37: 130–137.doi:10.1139/f80-017.

Vatland, S.J., Gresswell, R.E., and Poole, G.C. 2015. Quantifying stream thermalregimes at multiple scales: Combining thermal infrared imagery and station-ary stream temperature data in a novel modeling framework. Water Resour.Res. 51(1): 31–46. doi:10.1002/2014WR015588.

Verspoor, J.J., Braun, D.C., Stubbs, M.M., and Reynolds, J.D. 2011. Persistent eco-logical effects of a salmon-derived nutrient pulse on stream invertebratecommunities. Ecosphere, 2(2): 1–17. doi:10.1890/ES10-00011.1.

Vietz, G.J., Sammonds, M.J., and Stewardson, M.J. 2013. Impacts of flow regula-tion on slackwaters in river channels. Water Resour. Res. 49(4): 1797–1811.doi:10.1002/wrcr.20094.

Vranovsky, M. 1995. The effect of current velocity upon the biomass of zoo-plankton in the River Danube side arms. Biologia, 50(5): 461–464.

Walker, K.F., Sheldon, F., and Puckridge, J.T. 1995. Perspective on dryland riverecosystems. Regul. Rivers: Res. Manage. 11(1): 85–104. doi:10.1002/rrr.3450110108.

Watkins, S.C., Nielsen, D., Quinn, G.P., and Gawne, B. 2011. The influence of leaflitter on zooplankton in floodplain wetlands: changes resulting from riverregulation. Freshw. Biol. 56(12): 2432–2447. doi:10.1111/j.1365-2427.2011.02665.x.

Wawrzyniak, V., Piégay, H., Allemand, P., Vaudor, L., and Grandjean, P. 2013.Prediction of water temperature heterogeneity of braided rivers using veryhigh resolution thermal infrared (TIR) images. Int. J. Remote Sens. 34(13):4812–4831. doi:10.1080/01431161.2013.782113.

Winemiller, K.O. 1989. Patterns of variation in life history among South Ameri-can fishes in seasonal environments. Oecologia, 81(2): 225–241. doi:10.1007/BF00379810. PMID:28312542.

Winemiller, K.O. 2004. Floodplain river food webs: generalizations and implica-tions for fisheries management. RAP Publication 2004/17, FAO Regional Of-fice for Asia and the Pacific Bangkok, Thailand. pp. 285–309.

Winemiller, K.O. 2005. Life history strategies, population regulation, and impli-cations for fisheries management. Can. J. Fish. Aquat. Sci. 62(4): 872–885.doi:10.1139/f05-040.

Winemiller, K.O., and Jepsen, D.B. 1998. Effects of seasonality and fish move-ment on tropical river food webs. J. Fish Biol. 53: 267–296. doi:10.1111/j.1095-8649.1998.tb01032.x.

Winemiller, K.O., and Rose, K.A. 1992. Patterns of life-history diversification inNorth American fishes: implications for population regulation. Can. J. Fish.Aquat. Sci. 49(10): 2196–2218. doi:10.1139/f92-242.

Winemiller, K.O., and Rose, K.A. 1993. Why do most fish produce so many tinyoffspring? Am. Nat. 142(4): 585–603. doi:10.1086/285559. PMID:19425962.

224 Can. J. Fish. Aquat. Sci. Vol. 77, 2020

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.

Wolter, C., and Arlinghaus, R. 2003. Navigation impacts on freshwater fishassemblages: The ecological relevance of swimming performance. Rev. FishBiol. Fish. 13(1): 63–89. doi:10.1023/A:1026350223459.

Wolter, C., and Arlinghaus, R. 2004. Burst and critical swimming speeds of fishand their ecological relevance in waterways. Berichte des IGB 20: 77–93.

Wootton, R.J. 1998. Ecology of Teleost Fishes. Kluwer Academic Publishers, Dor-drecht, the Netherlands.

Zampatti, B.P., and Leigh, S.J. 2013. Within-channel flows promote spawning andrecruitment of golden perch, Macquaria ambigua ambigua — implications for

environmental flow management in the River Murray, Australia. Mar.Freshw. Res. 64(7): 618–630. doi:10.1071/MF12321.

Zeug, S.C., and Winemiller, K.O. 2008. Relationships between hydrology, spatialheterogeneity, and fish recruitment dynamics in a temperate floodplainriver. River Res. Appl. 24(1): 90–102. doi:10.1002/rra.1061.

Ziv, G., Baran, E., Nam, S., Rodríguez-Iturbe, I., and Levin, S.A. 2012. Trading-offfish biodiversity, food security, and hydropower in the Mekong River Basin.Proc. Natl. Acad. Sci. 109(15): 5609–14. doi:10.1073/pnas.1201423109. PMID:22393001.

Humphries et al. 225

Published by NRC Research Press

Can

. J. F

ish.

Aqu

at. S

ci. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y 13

8.80

.78.

229

on 0

3/04

/20

For

pers

onal

use

onl

y.