Otolith microchemistry of two amphidromous galaxiids across an experimental salinity gradient: A multi-element approach for tracking diadromous migrations

Abstract

An increasing number of studies are uncovering considerable flexibility in migration patterns of diadromous fishes. The development of otolith microchemical techniques has largely driven this research and led to an appreciation of the significance of facultative diadromy in the life history of numerous species. However, validation experiments need to be undertaken for each species and life stage of interest before diadromous migrations can be confidently reconstructed. These validation experiments are required to establish a salinity calibration series against which the otolith microchemistry of unknown individuals can be compared. To facilitate research on facultative amphidromy in galaxiids, we reared the larvae of two species, Galaxias maculatus and G. argenteus, in five different salinities (2, 5, 10, 20, 34). We tested whether trace element signatures of fish reflected their salinity treatment, and hence whether otolith microchemistry could reconstruct diadromous migrations. Distinguishing low salinity (2 and 5) from high salinity (20 and 34) treatments was straightforward using otolith Sr:Ca alone. The five salinity treatments resulted in five distinct multi-trace element signatures for both species (DFA classification success of 85% and 92% for G. maculatus and G. argenteus, respectively). Otolith lithium showed a similar trend to otolith Sr:Ca (ie. higher in saltwater), and otolith Rb:Ca showed a surprising negative trend with salinity despite higher ambient Rb concentrations in saltwater. Our results suggest otolith Li:Ca and Rb:Ca should be considered as part of a multi-trace element approach when investigating diadromous migrations, particularly when non-marine Sr levels may be high.

Introduction

Methodologies for characterising age-specific migration patterns are necessary to identify essential fish habitat at different life stages, as well as determining whether connectivity between different habitats is important for completion of typical life cycles (Elsdon et al., 2008). Being able to track individual migration patterns also allows an understanding of intergenerational exchange that can occur among spatially separated populations (as in a meta-population, Thorrold et al., 2007). Diadromous migrations are an extreme form of large scale movement and differential habitat utilisation, and diadromous species comprise important fisheries worldwide (McDowall, 1990). It has been difficult to track diadromous migrations due to the large distances these fishes can travel, the low recapture rates of tagged individuals, and the inability to physically tag small individuals with traditional methods, which limit the feasibility of mark/recapture studies (Secor et al., 1995). The development of otolith microchemical techniques, whereby chronologically deposited and metabolically inert geochemical information can be used to reconstruct the environmental history of a fish, has allowed diadromous migrations to be more efficiently reconstructed (Secor et al., 1995).

Otolith microchemistry has revealed a diverse range of life history strategies and habitat utilisation in a range of diadromous fish species (Secor et al., 1995). ‘Facultative diadromy’, whereby not all individuals in a presumed diadromous population actually make a migration between the river and the sea, has been identified in numerous species and populations (e.g.,: Herring, Alosa aestivalis, Limburg, 1998; Smelt, Hypomesus nipponensis, Katayama et al., 2000; Sturgeon, Acipenser guldenstadti, Arai and Miyazaki, 2001; Stickleback, Gasterosteus aculeatus, Arai et al., 2003; Striped bass, Morone saxatilis Zlokovitz et al., 2003; Shirauo, Salangichthys microdon, Yamaguchi et al., 2004; Smelt, Retropinna semoni, Crook et al., 2008). Facultative diadromy extends to all three forms of diadromy (as defined by McDowall, 1992). For example, facultative catadromy in Anguillids (Arai et al., 2004), whereby some eels complete their life entirely within the marine environment; facultative anadromy in brown trout (Arai et al., 2002), whereby some individuals are freshwater resident; and facultative amphidromy in bullies and galaxiids, whereby larvae develop in freshwater lake systems despite having open access to the ocean (Closs et al., 2003, David et al., 2004). In all cases, otolith microchemistry has been used to distinguish whether individuals have utilised marine and/or freshwater habitat during different stages of their lives.

Strontium, and to a lesser extent barium, are the two elements commonly used to distinguish residency in waters of different salinity (Elsdon et al., 2008). There is usually a much higher and lower abundance of strontium and barium, respectively, in saltwater compared with freshwater (Elsdon et al., 2008). Often, scientists are interested in distinguishing estuarine versus freshwater or marine habitat use (e.g., Arai et al., 2004), or even utilisation of different parts of an estuary (e.g., Kafemann et al., 2000). Distinguishing between freshwater and saltwater residency has usually been straightforward in previous validation experiments (e.g., Farrell & Campana, 1996, Tzeng, 1996, Kraus & Secor, 2003, Secor et al., 1995), but there has been mixed success in discriminating habitat utilisation over finer salinity scales. For example, Kraus and Secor (2003) suggested otolith Sr:Ca ratios could be used to determine residency in completely fresh, estuarine or open ocean water, but not over finer salinity scales. Importantly, the non-linear relationship between salinity and ambient element:Ca ratios would make it difficult to discriminate between medium to high salinity habitat utilisation (most variation occurs below 8 ppt — Kraus & Secor, 2003, Lowe et al., 2009), if ambient element:Ca ratio rather than ambient absolute concentrations dictate the rate of element uptake into otoliths.

The patterns of strontium and barium uptake can be both species and age dependent. Validation studies performed on one species will not necessarily extend to others (Zimmerman, 2005). Similarly, validation studies performed on adult fish will not necessarily extend to larvae. Hence, laboratory validation studies should be undertaken on the species and age group of interest before trying to infer patterns of diadromy or habitat utilisation, particularly if the goal is to reconstruct movements across finer salinity scales (Kraus & Secor, 2004, Elsdon et al., 2008).

Amphidromy is a distinct form of diadromy with adults living in freshwater where most growth and reproduction takes place, but a short larval phase is expected to take place in the ocean (McDowall, 2007). Amphidromous galaxiids are a dominant freshwater fish fauna in New Zealand, Australia and South America (McDowall, 1990). Amphidromous galaxiid species typically spend a 3–6 month larval period in the ocean, before the transparent juveniles return to rivers where they are targeted as the basis of whitebait fisheries (McDowall, 1990). Genetic homogeneity at a continental scale suggests considerable larval dispersal occurs during the marine phase (Waters et al., 2001). Landlocked populations exist in large inland lake systems that have open access to the sea (McDowall, 1990), which suggests larvae are not actively dispersing (out of the lake system). There is also evidence for larval retention in galaxiids inhabiting coastal lake systems with open access to the sea (David et al., 2004). The presence of one and three endemic migratory galaxiid species in Australia and New Zealand and nearby islands, respectively, suggests colonisation abilities may differ among galaxiids despite all species having a pelagic larval period (McDowall, 1990). Being able to track a large number of individual larvae is thus necessary to determine the scale of movement of most individuals and effective population connectivity. This is difficult when trying to track movement through a comparatively homogeneous ocean, but determining retention in lakes or estuaries should be more straightforward. Determining the salinity history of larvae, and whether patterns of larval retention in lakes and estuaries are consistent among species, will thus help explore the magnitude of self-recruitment in galaxiid species.

Before relating otolith microchemistry of wild-caught fish to their salinity history, we needed to determine the degree to which movements across salinity gradients could be confidently inferred. We thus reared larvae of two migratory galaxiid species, Galaxias maculatus and G. argenteus, in salinities of 2, 5, 10, 20 and 34 to determine whether otolith trace element signatures could be related to a range of freshwater, estuarine and saltwater conditions. We primarily wanted to ensure we could identify individuals that had not spent time in a marine environment (i.e., were non-diadromous). We also wanted to explore whether there was potential to identify movements across finer salinity scales. To advance a more general understanding of factors regulating otolith microchemistry, we also explore whether otolith element:Ca is best predicted by the salinity, ambient element:Ca or ambient element concentrations in our experimental treatments.