Rev Fish Biol Fisheries (2014) 24:333–368 DOI 10.1007/s11160-013-9334-6 REVIEWS Homing and straying by anadromous salmonids: a review of mechanisms and rates Matthew L. Keefer • Christopher C. Caudill Received: 10 July 2013 / Accepted: 16 November 2013 / Published online: 22 November 2013 Ó Springer Science+Business Media Dordrecht 2013 Abstract There is a long research history addressing Reported salmonid stray rates indicate that the olfactory imprinting, natal homing, and non-natal behavior varies among species, among life-history straying by anadromous salmon and trout (Salmoni- types, and among populations within species. Most dae). In undisturbed populations, adult straying is a strays enter sites near natal areas, but long-distance fundamental component of metapopulation biology, straying also occurs, especially in hatchery popula- facilitating genetic resilience, demographic stability, tions that were outplanted or transported as juveniles. recolonization, and range expansion into unexploited A majority of past studies has estimated straying as habitats. Unfortunately, salmonid hatcheries and other demographic losses from donor populations, but some human actions worldwide have affected straying in have estimated straying into recipient populations. ways that can negatively affect wild populations Most recipient-based estimates have substantiated through competitive interactions, reduced productiv- concerns that wild populations are vulnerable to ity and resiliency, hybridization and domestication swamping by abundant hatchery and farm-raised effects, and outbreeding depression. Reduced adult strays. straying is therefore an objective for many managed populations. Currently, there is considerable uncer- Keywords Imprinting Á Olfaction Á tainty about the range of ‘natural’ stray rates and about Oncorhynchus Á Orientation Á Philopatry Á Salmo which mechanisms precipitate straying in either wild or human-influenced fish. Research in several disci- plines indicates that adult straying is affected by Introduction endocrine physiology and neurological processes in juveniles, incomplete or interrupted imprinting during Our understanding of homing and straying by anad- rearing and emigration, and by complex interactions romous salmonids (Oncorhynchus, Salmo, and Salv- among adult maturation processes, reproductive elinus species) has primarily been derived from behaviors, olfactory memory, environmental condi- studies of fish physiology and behavior under the tions during migration, and senescence physiology. purview of hatchery management. Far less is known about wild populations and there is considerable disagreement about the rates of ‘‘natural’’ versus M. L. Keefer (&) Á C. C. Caudill management-related straying. Our objectives in this Department of Fish and Wildlife Sciences, College of review are to: (1) consider adult salmonid homing and Natural Resources, University of Idaho, 975 W. 6th Street, Moscow, ID 83844-1136, USA straying within broad evolutionary, ecological, and e-mail: [email protected] management contexts; (2) summarize the mechanisms 123 334 Rev Fish Biol Fisheries (2014) 24:333–368 associated with homing versus straying; (3) provide and recolonization after local extinction (Withler 1982; recommendations on a lexicon for the variety of Anderson and Quinn 2007; Perrier et al. 2010; Pess straying behaviors that have been described in the et al. 2012). Straying also reduces inbreeding depres- literature; (4) assess potential predictors of straying in sion and density dependent effects such as competition wild and human-influenced populations; and (5) among related individuals (Quinn 1993; Hendry et al. synthesize reported straying rates and behaviors across 2004). It is less clear whether some individuals are species, life history types, and populations. genetically predisposed to straying versus homing, though it is likely that expression of these two Evolutionary context strategies is in dynamic equilibrium in wild popula- tions (Quinn 1984). Accumulating evidence suggests Philopatry (i.e., homing) to natal sites is a fundamental that a combination of predominantly philopatric indi- life-history trait of most anadromous salmon and trout. viduals plus some strays makes for robust populations Homing increases the likelihood that reproductive-age that can exploit favorable natal-site habitats, expand fish will find mates and locate habitats that are into new sites, and disperse in the face of temporary or favorable for both adult spawning and juvenile catastrophic environmental fluctuations. Indeed, sal- survival (Hendry et al. 2004; Quinn 2005). Return to monids are increasingly considered in terms of meta- natal sites is therefore highly adaptive, providing populations connected by some degree of movement fitness benefits and contributing to the evolution of (i.e., straying) among populations. thousands of locally-adapted populations (Taylor 1991; Hendry et al. 2000; McDowall 2001; Waples Ecological context et al. 2004). The spatial scale for homing varies among species, among populations, and within populations, Adult salmonids select spawning sites using a complex and ranges from very precise (i.e., within meters of combination of heritable homing behaviors plus natal sites; Stewart et al. 2003; Quinn et al. 1999, 2006, proximate behavioral responses to environmental 2012) to broader habitat units like river reaches or and social cues (Dittman and Quinn 1996). As adults river drainages (Candy and Beacham 2000; Bentzen approach potential spawning habitats, they must et al. 2001; Hamann and Kennedy 2012). simultaneously orient to natal sites and locate areas Straying is typically defined as adult migration to— with suitable substrate, water temperature, water and attempted reproduction at—non-natal sites (Quinn velocity, hyporheic flows, and other geomorphic 1993). In the context of hatcheries and other human features prior to spawning (Geist and Dauble 1998; interventions, straying is often negatively framed as a Torgersen et al. 1999). They also must avoid preda- ‘‘failure to home’’. However, straying in wild popu- tion, locate mates, defend against competitors, and lations is a critical evolutionary feature of salmonid successfully deposit gametes at the appropriate time. biology that compliments homing. Straying can be In some cases, adults hold in suitable habitat for weeks adaptive over short ecological time frames and at to months prior to the onset of spawning; this holding longer evolutionary scales. It is therefore useful to can occur at the eventual spawning location or in more distinguish proximate factors that affect straying, such distant staging areas. The degree of success in each of as sensory ecology and the physiology of orientation these ecological arenas ultimately drives reproduction and reproduction, from the ultimate factors that have and the evolution of locally-adapted traits and led to the evolution and maintenance of straying. Both populations. contribute to the observed variability in straying The relative ecological stability and quality of among populations and understanding these effects spawning and rearing habitats can be a good predictor is necessary to interpret behaviors and set manage- of homing rates. High site fidelity tends to arise when ment goals (e.g., if managing for zero straying is not high-quality habitats are stable through time because possible, what are ‘‘natural’’ straying rates?). such sites consistently attract adults and produce It is clear that straying buffers against spatial and successful offspring (Quinn and Tallman 1987; Hen- temporal variation in habitat quality, and allows colo- dry et al. 2004). Site fidelity is typically lower where nization of new habitats (Milner and Bailey 1989; unpredictable inter- or intra-annual fluctuations in Burger et al. 1997; Quinn et al. 2001; Stephenson 2006) habitat quality or quantity result in lower mean 123 Rev Fish Biol Fisheries (2014) 24:333–368 335 survival or cohort failures (Quinn 2005; Cram et al. particular importance given the use of olfaction for 2013). route finding and home site recognition. Straying is just one component of salmonid life history that varies within and among populations in Fisheries management context response to habitat stability. Like variation in juvenile residency times (Healey 1991), age at maturity (Groot Demographically, straying fish affect two populations: and Margolis 1991; Fleming 1996), or the number of their origin population (i.e., the donor population) and spawning events (i.e., iteroparity, Fleming and Rey- their selected breeding population (i.e., the recipient nolds 2004), straying can effectively ‘‘hedge’’ against population). While straying is typically considered as habitat instability. Phenotypic plasticity during juve- a per capita probability (e.g., 3 %), it is important to nile stages (e.g., in behavior or physiology) can consider the absolute number of strays, emigration maximize survival to adulthood given the availability rates from donor populations, and immigration rates and predictability of suitable habitats. In comparison, into recipient populations. Strays are always a demo- plasticity in adult life history temporally spreads the graphic loss from the donor population. This can be a risk of reproductive failure across years (i.e., multiple management concern when the donor population is maturation ages or spawning attempts) or can spread limited by the number of breeders or there are risks of the risk spatially (i.e., by straying to productive sites) genetic