Ecologists have long been interested in relevant techniques to track the field movement patterns of fish. The elemental composition of otoliths represents a permanent record of the growing habitats experienced by a fish throughout its lifetime and is increasingly used in the literature. The lack of a predictive and mechanistic understanding of the individual kinematics underlying ion incorporation/depletion limits our fine-scale temporal interpretation of the chemical signal recorded in the otolith. In particular, the rate at which elements are incorporated into otoliths is hypothesized to depend on fish physiology. However, to date, time lags have mostly been quantified on a population scale. Here, we report results from controlled experiments (translocation and artificially enriched environment) on individual trace element incorporation/depletion rates in Salmo trutta (Salmonidae). We reported significant lags (i.e. weeks to months) between changes in water chemistry and the subsequent change in otolith composition and highlighted substantial inter-individual variations in the timing and magnitude of Sr/Ca and Ba/Ca responses. These differences are partially linked to the energetic status (i.e. metabolic rate) of the individuals. It therefore appears that individuals with the highest metabolic rate are more likely to record detailed (i.e. brief) temporal changes than individuals having lower metabolic values. The time taken for environmental changes to be reflected in the growing otolith thus can no longer be assumed to remain a constant within populations. Results from the current study are a step towards the fine reconstruction of environmental histories in dynamic environments.

Knowledge of population structure is fundamental to effective management and conservation. Patterns of connectivity, or the exchange of individuals among subpopulations of a species, are of primary importance, as is a robust understanding of migration patterns. In this context, the field of movement ecology has grown rapidly in the past decade and is now providing increasingly detailed spatio-temporal data for a wide variety of taxa (Allen and Singh, 2016; Fraser et al., 2018). In particular, ecologists have paid a great deal of attention to the proper techniques to track fish movement in the wild since accurately resolving fish habitat use (i.e. movement and migration patterns) is essential to the development of effective management and conservation plans (Cooke et al., 2016; Fromentin et al., 2009).

Historically, telemetry-derived methods [especially radiotelemetry, acoustic telemetry and passive integrated transponders (PITs)] have massively improved our capacity to collect fine-scale spatiotemporal information on fish (Cooke et al., 2013; Lucas and Baras, 2000). Subsequently, significant developments were largely based on the technological and analytical advances in extracting information from the calcified structures of fish (Begg et al., 2005), which have provided some of the best examples of biochronologies in the animal kingdom (Campana and Thorrold, 2001). More specifically, chemical signatures of otoliths (bones in the fish inner ear or ‘ear stones’) have proved to be valuable tools to reconstruct environmental histories and migratory patterns of individual fish and have become widespread in both marine and freshwater fishery research (Carlson et al., 2017). These applications are possible because of how otoliths are formed. They grow continuously throughout the life of the fish, incorporate some elements and their isotopes in proportion to their ambient abundance and accrete calcium carbonate in sequential layers that preserve the timing of deposition (Campana, 1999). Given that carbonates are metabolically inert in otoliths, they act as permanent fingerprints recording environmental signatures of body waters sequentially encountered by fishes (Campana, 1999; Elsdon et al., 2008). The complete sequence of habitats experienced by a fish throughout its lifetime can therefore be assessed using the elemental composition along a transect ranging from the primordium to the otolith's edge.

Currently, otolith chemistry is used to address various issues relating to the ecology of fish, ranging from studies estimating population structure and discriminating stocks (reviewed in Tanner et al., 2015; Thresher, 1999), migration patterns and philopatric behavior (e.g. Bradbury et al., 2008; Elsdon and Gillanders, 2003a; Walther et al., 2011), site residency (e.g. Lord et al., 2011; Teichert et al., 2018), nursery habitats (e.g. Brennan et al., 2015; Marklevitz et al., 2016), environmental histories of fish (Limburg, 1995; Walther et al., 2011) and finally, investigating connectivity by retrospectively assigning adults to their areas of origin (e.g. Gillanders, 2005; Thorrold et al., 2001). In all these contexts, applications rely on the concentration of elements in otoliths changing in predictable manners in relation to changes in environmental settings. The underlying hypothesis is that the duration of elemental and isotopic singular signatures are linked to the amount of time spent by fishes in a particular water body.

Initial studies mostly focused on movement patterns through chemically distinct bodies of water over long periods. In this context, diadromous migrations have been intensively studied (Walther and Limburg, 2012). Until recently, only a few elements (strontium, magnesium, manganese and barium) were routinely measured in otoliths. For example, in the case of diadromous migrations, Sr and Ba can be used alone as environmental recorders (Sturrock et al., 2012; Walther and Limburg, 2012). The recent use of laser ablation and/or microprobe analysis on otoliths allows the simultaneous collection of various elements that are more likely to discriminate among water masses experienced by fish (Mercier et al., 2011). Multi-proxy approaches may therefore allow investigation of movement patterns in the wild at remarkably fine geographical scales (e.g. Gemperline et al., 2002; Gillanders and Kingsford, 2003; Svedäng et al., 2010). In the meantime, the rapid improvements in the analytical capabilities of sampling equipment raised expectations of unravelling environmental and migratory histories with increasingly fine temporal resolutions. Analyses have indeed shifted at time scales ranging from inter-annual to seasonal or beyond (Brennan et al., 2015; Elsdon and Gillanders, 2005a; Kennedy et al., 2000; Macdonald and Crook, 2010; Weidel et al., 2007). On a theoretical basis, it may even be possible to detect brief incursion/residence in adjacent habitats having contrasted geochemical signatures (e.g. among freshwater drainages/tributaries). However, the fine-scale analytical resolution of equipment sharply contrasts with the physiological temporal resolution of elemental incorporation in otoliths. Several investigations have reported the existence of substantial time lags (i.e. weeks to months) between a fish's exposure to changed environmental conditions and when elemental concentrations stabilize to reflect these new conditions (Lowe et al., 2009; Macdonald and Crook, 2010; Miller, 2011). Otoliths are not in direct contact with the surrounding water and elements must pass through several tissue and cell boundaries (i.e. gill epithelium, gut membrane and endolymphatic epithelium; Hüssy et al., 2020) before being incorporated into the matrix of calcified structures. Consequently, changes in environmental variables would not result in instantaneous changes in elements within calcified structures. This temporal lag is a critical aspect that limits the spatial and temporal resolution attainable for multi-elemental otolith analyses and would affect the interpretation of environmental histories (Chang and Geffen, 2013; Sturrock et al., 2012).

Retracing movements of fishes using otolith chemistry analysis is not straightforward and also requires detailed knowledge on the predictable relationships between otolith and water chemistry (otolith elemental ratios are some fraction of those present in the ambient environment). In this context, considerable research effort was focused on the exogeneous factors (i.e. salinity, temperature, ambient water concentrations) that can independently or interactively influence the uptake and incorporation of trace elements into the otolith. In contrast, endogenous factors (i.e. physiological factors such as growth rates or stress) have received relatively less attention (Walther et al., 2010) despite the long recognition of their importance (e.g. Kalish, 1989, 1991, 1992; Radtke and Shafer, 1992). The experimental interaction between exogeneous and physiological factors is even more rarely considered in the literature. Most recent advances in the field highlighted that, contrary to a long-standing assumption, physiological components can considerably influence elemental uptake and processing mechanisms (Sturrock et al., 2015, 2014). However, such rigorous investigations only focus on elemental concentrations (i.e. partition coefficients) with no specific consideration for the time taken for environmental changes to be reflected in the growing otolith (i.e. kinematics). Physiological effects such as metabolic rate are frequently suspected or hypothesized to explain inter-individual differences. On a theoretical basis, biophysical factors influencing the kinematics of both uptake and depletion of ions in the endolymph are mediated by the transport of these ions through cell membranes, which in turn may be related to some extent to metabolic rate. Nevertheless, the potential effect of individual metabolic rates on time lag has, to the best of our knowledge, never been investigated. This has been identified as an area urgently requiring further investigation (Sturrock et al., 2012; Tanner et al., 2015) that would be hugely beneficial (Walther et al., 2010).

The present study addressed this issue by examining chemical signatures in the otoliths of juvenile brown trout (Salmo trutta) following experimental changes in environmental conditions reproducing the range of concentration typically encountered by the species between different tributaries in the same watershed. More precisely, our specific aims were to: (1) precisely quantify the kinematics of both incorporation and depletion of Sr and Ba in otoliths following habitat change; (2) examine inter-individual times taken for environmental changes to be reflected in the growing otolith; and (3) examine whether inter-individual differences in time lag are related to individual metabolic rates (i.e. as revealed by oxygen consumption), with the prediction that individuals with the highest metabolic rate have the finest micro-chemical resolution (i.e. shortest time lag).

A variety of chemical tracers in otoliths have been used to reconstruct movement patterns. In particular, elemental ratios such as Sr/Ca and Ba/Ca are well documented and widely used in the literature. Apart from their use to track diadromous migrations (Walther and Limburg, 2012), these elements can also vary significantly in freshwater systems between or even within rivers depending on streambed geological composition (Ramsay et al., 2011; Walther et al., 2011; Zitek et al., 2010) and have specifically proved to be useful when studying movement patterns of salmonids species between tributaries in southwestern France (Martin et al., 2013a,b). Here, we exclusively focused on Sr/Ca and Ba/Ca because among elements used in otolith microchemistry strong positive relationships between waters and otoliths in freshwater systems were mainly found for these two elements (see review by Brown and Severin, 2009; Hüssy et al., 2020).

Site, fish collection and experimental design

For this experiment, a full-sibling group of juvenile brown trout (Salmo trutta Linnaeus 1758) spawned from captive parents (43 females and 15 males) in the Lees-Athas fish hatchery facilities, southwestern France (42°57′57″N, 0°36′50″W) were used. After fertilization, eggs were incubated and individuals were maintained in natural temperature conditions (water supplied by spring water with a constant temperature of 11.6±0.2°C during the winter of 2015–2016). On 15 April 2016, 3 months post-emergence (alevins live in the gravel, feeding off the remaining yolk that is attached to their body, then they emerge out of the gravel as fry, set up territories and start exogenous feeding), 300 individuals were transferred by car in large oxygenated coolers maintained at 12°C and reared in controlled experimental facilities (ECP, INRAE. The Ecology and Fish Population Biology Facility) with water derived from the neighbouring Nivelle River (southwestern France 43°21′14″N, 1°33′54″W). Such water has comparatively lower barium and higher strontium concentrations compared with previous conditions. During transfer, fish were marked by immersion in Alizarin Red S solution (ARS; 100 mg l−1, 3 h in a well-aerated 20 l tank maintained at 12°C) for obtaining a posteriori time-resolved elemental signals (Caudron and Champigneulle, 2006). Throughout the experiment (4.5 months, 145–153 days according to individuals), all fish were maintained under the same environmental conditions in a single tank (750 l) with water maintained at 12.0±0.3°C and ad libitum feeding conditions.

After 2 months (66 days post translocation), experimental tanks were artificially enriched in barium (fourfold increase, from ∼10 µg l−1 to ∼40 µg l−1, see Table 1) by adding the appropriate amount of BaCO3 (Sigma-Aldrich) powder, reproducing the range of concentration naturally encountered by the species between different tributaries in the same watershed in southwestern France (Martin et al., 2013a). The strontium concentration remained unchanged. On the first day of barium enrichment (20 June 2016) fish were collectively marked a second time by immersion in Alizarin Red S (same conditions as before). About 10% of the water volume was replaced every week in the experimental tank, with the addition of the appropriate amounts of barium each time. Water samples were collected at approximately 15 day intervals for a posteriori verification of the enrichment (Table 1). After additional rearing for months in these enriched conditions (i.e. 79–87 days according to individual), 51 individual fish were randomly selected for measuring oxygen consumption (as a proxy of metabolism; see ‘Metabolism’ below). They were subsequently euthanized, measured (standard length to nearest mm), weighed (to nearest 1 mg) and prepared for otolith extraction. Given that transferred individuals were initially of homogeneous size (based on a randomly selected subset of 101 fishes, mean±s.d. of 33.7±1.5 mm), fish length at the end of the experiment was considered as a proxy for somatic growth. During the rearing period, Sr/Ca and Ba/Ca ratios were the only noteworthy changed component through time. The complete experimental design is summarized in Fig. 1. Experimental design and maintaining conditions were both approved by National Ethics Committee for Fishes and Birds (CE73), with respect to the national chart.

Fig. 1.

Schematic chronology of the main steps in the experimental design. Alevins were raised at the Lees-Athas (River 1) fish hatchery facilities. They live in the gravel (left), feeding off the remaining yolk that is attached to their body, then they emerge out of the gravel as fry and start exogenous feeding, and growth for months. Fish were collectively marked by immersion in Alizarin Red S (ARS) after 3 months (red shading) and switched to water obtained from River 2 (Nivelle) after which they were immersed in ARS for a second time and exposed to enriched barium conditions (Nivelle enriched). After 79–87 days according to individual, 51 fish were randomly selected for measuring oxygen consumption. Fish were subsequently euthanized, measured, weighed and prepared for otolith extraction.

Fig. 1.

Schematic chronology of the main steps in the experimental design. Alevins were raised at the Lees-Athas (River 1) fish hatchery facilities. They live in the gravel (left), feeding off the remaining yolk that is attached to their body, then they emerge out of the gravel as fry and start exogenous feeding, and growth for months. Fish were collectively marked by immersion in Alizarin Red S (ARS) after 3 months (red shading) and switched to water obtained from River 2 (Nivelle) after which they were immersed in ARS for a second time and exposed to enriched barium conditions (Nivelle enriched). After 79–87 days according to individual, 51 fish were randomly selected for measuring oxygen consumption. Fish were subsequently euthanized, measured, weighed and prepared for otolith extraction.

Table 1.

Elemental composition of water throughout the experiment

Elemental composition of water throughout the experiment
Elemental composition of water throughout the experiment

Metabolism

We needed to quantify individual metabolic rates (i.e. the total amount of energy used by the animal at a time interval). In fishes, standard metabolic rate (SMR, the rate of energy expenditure by the body during complete physical, mental and digestive rest) and routine metabolic rate (RMR) are usually measured and provide complementary information. Since the metabolic rate varies with activity level, more active animals have a higher metabolic rate than less active animals. RMR thus integrates the intrinsic/constitutive swimming mobility of individuals that should be fully considered as a significant part of their energy budgets.

Metabolic rate was estimated by measuring oxygen consumption. A total of 51 juveniles were retrieved for oxygen consumption determination for 7 days (7–11 September and then 14–15 September 2016). Briefly each day, 8 juveniles (only 3 juveniles on the final day) were starved for 36 h before analysis. Then, they were each transferred to a respirometry chambers (diameter: 45 mm; length: 180 mm) of an intermittent flow respirometer as described by Régnier et al. (2010). Juveniles were introduced into the chambers at 15:00 h and oxygen consumption was recorded continuously every minute until 10:00 h the next day. The closed/open phase of the system was 30 min/15 min and the duration of the closed phase was determined in order that the oxygen level in the chamber was always kept above 80% O2 saturation. Every 30 min, oxygen consumption was approximated using a simple linear regression from which the regression slope was extracted. After that time, the background respiration, i.e. oxygen consumption within the respirometer due to microbial respiration, was estimated by measuring the oxygen consumption rate in the respirometer without a fish (i.e. blank run for 2 h). Temperature and photoperiod used for the acclimatization phase and oxygen measurements were similar to those used all along within experimental facilities. To estimate the SMR, the first 5 h were considered as a period of acclimation and the average oxygen consumption was calculated using the three slightest consecutive slopes (using an exhaustive search within the 08:00–20:00 h period). As some individuals may be more or less restless in the respirometric chambers, the RMR was also estimated by considering all slopes after the acclimation period. After their period in the respirometry chamber, individuals were immediately euthanized by immersion in buffered benzocaine solution (preliminary anaesthesia at 4.0 ml l−1 followed by euthanasia at 12.0 ml l−1; 11–12°C).

Water and otolith microchemical analysis

Water samples were collected at approximately 15 days intervals from experimental tanks and were processed according to the procedure given in Martin et al. (2013a,b). Briefly, using polypropylene syringes, samples were filtered through 0.45 μm PVDF Whatmann filters into acid-washed low-density polyethylene (LDPE) bottles and acidified to 2% with 70% ultrapure nitric acid (J. T. Baker, Ultrex II) to fix the sample and enable storage. Analysis of water samples was carried out on undiluted samples using solution-based inductively coupled plasma atomic emission spectrometry (ICP-AES) (ACTIVA, Jobin Yvon). Internal standard of indium (500 µg l−1) was used to correct for instrument drift. Blanks were regularly performed, using the same protocol as for the samples, with 18.2 MΩ MQ water (Millipore). Samples of known concentration, certified reference freshwater (SLRS-6, NRCC; www.nrc-cnrc.gc.ca), were used to estimate accuracy and precision; measured concentrations were within 93–103% of reported values for all elements and external precision was <4% for Ca and Sr, and ∼5% for Ba.

The same individuals (N=51) were used for both oxygen consumption and microchemistry. Sagittal otoliths were removed with acid-washed plastic forceps, gently cleaned in pure water, dried and stored in acid-washed plastic vials prior to embedding in epoxy resin (Araldite, Huntsmann). Otolith lengths (from rostrum to antirostrum) were recorded as a rough proxy of otolith growth. Embedded otoliths were ground with decreasing size abrasive paper (grit 1200, 2500 and 4000) until the otolith core was visible, polished with a polycrystalline diamond suspension (1 μm) and carefully rinsed with ultrapure water. Otolith Sr/Ca and Ba/Ca ratios were quantified using a 1030 nm femtosecond laser ablation system (Alfamet-Novalase, France) coupled with an inductively coupled plasma mass spectrometer (DRCII, Perkin Elmer). The laser was set at a pulse rate of 50 Hz with a 12 μm ablation spot and traveled at 5 μm s−1. Analysed isotopes were 86Sr, 138Ba and 43Ca used as an internal standard to account for any variations in the ablation yield. External calibrations were performed using synthetic certified reference glasses NIST614, NIST612 and NIST610 (NIST, USA). Analytical accuracy was achieved with the fish otolith FEBS-1 (NRCC, Canada; Sturgeon et al., 2005) and NIES22 (NIES, Japan; Yoshinaga et al., 2000). Limits of detection were calculated based on a 3σ criterion, where σ is the standard deviation for 20 s measurements of the blank signal. Limits of detection (ppm) were: 86Sr=1.85 ppm and 138Ba=0.06 ppm. Accepted recovery of reference materials for both elements ranged from 94% to 109%. Strontium and barium compositions were expressed as mass elemental ratios (i.e. Sr/Ca, Ba/Ca) on the basis of the stoichiometry of Ca carbonate (380,000 μg Ca g−1 otolith) (Campana, 1999).

Otolith elemental data were collected through a bipartite transect using the core as a bifurcation point (Fig. 2). Two chronologies of different lengths were therefore obtained for each fish. They were arbitrarily categorized into two groups: ‘short’ transects (located on the dorsal part of the otolith; Aymes et al., 2016) and ‘long’ transects (comparatively located on the anteroventral part) (Fig. 2). The start and end points were both located 50 μm from the outermost edge of the otolith. All transect parts were located perpendicular to daily increments to generate a time series of elemental composition during the experiment.

Fig. 2.

Position of the transects on the trout otolith. (A) Typical location of transects on the otolith. (B) Composite image (white light and TXR filter) allowing us to relate the position of the otolith core and Alizarin Red S mark to the corresponding time-resolved elemental concentrations along the transects.

Fig. 2.

Position of the transects on the trout otolith. (A) Typical location of transects on the otolith. (B) Composite image (white light and TXR filter) allowing us to relate the position of the otolith core and Alizarin Red S mark to the corresponding time-resolved elemental concentrations along the transects.

Photographs were taken under unfiltered white light and fluorescent light (filter model: RFP1; excitation, 530–550 nm/emission, 575 nm; Olympus SZX-16 stereomicroscope, Olympus DP72 camera), which was used to detect the ARS marks. For each otolith two calibrated images were superimposed, one with transmitted light and the other with fluorescent light (Olympus Cell A software) for simultaneous observation of the transect, otolith edges and ARS mark (Fig. 2B). The position of these marks along the transect was used to identify the three main experimental periods and to proceed with a signal standardization, converting distance (in μm) into time (months) using a linear intrapolation assuming constant growth (during this period, fish growth is maximized, in the absence of a sharp thermal difference; Pearson's correlation coefficient between time and otolith increment width of 0.88; Fig. S1). Time series for otolith Sr/Ca and Ba/Ca were therefore plotted against otolith distance during the experimental period.

Statistical analysis

The change in concentration over time along each transect was modelled using a non-linear mixed-model applied to raw instrumental data using the nlme package (v. 3.1; https://CRAN.R-project.org/package=nlme) in R v. 3.2.2 (https://www.r-project.org/). For convenience, we used a re-parameterized analogue of the von Bertalanffy model in which both the initial and asymptotic values are explicitly inferred. Distinct formulae were used for decreasing trends (initial Ba/Ca):
(1)
and increasing trends (Sr/Ca and enriched Ba/Ca):
(2)
where Me/Ca is metal-to-calcium ratio, t the relative time (in months) along the transect, k the kinematic parameter (increasing/decreasing rate), L the asymptotic concentration and Z is the explicit initial concentration. In the mixed-model, ‘t’ was used as the main independent variable, with ‘individual’ (51 individuals) as a random effect that allows for the three parameters (k, L and Z) to be estimated for each fish. Owing to the anisotropic growth of the otolith, the two transects obtained for each individual (replicates) are of different lengths and may exhibit distinct signals. We tested for a possible effect of these ‘replicates’ by comparing a model with and without such an effect. When considered, transects (treated as a dual factor: short versus long) were primarily included in the statistical analysis as a fixed source of intra-individual variance (i.e. fixed effect term in mixed models) with individuals nested within each transect group. Alternatively, the replicates term was discarded in a second model, pooling all data per individual. As the two models have distinct fixed effects, the maximum likelihood estimator was preferred for fitting models. Akaike weight was used for model selection (Table S1). Also, as the model taking replicates into account is far less likely than the alternative model, we further pooled for each individual data from the two transects. We also accounted for autocorrelation in the residuals due to the underfitting of the model by including an AR1 model (i.e. first-order autoregressive model).

Global kinematic trends were estimated using the fixed effects of the above-mentioned non-linear mixed-model. Moreover, at a global scale, fitted data were analysed using a quantile regression approach implemented using the ‘mqgam’ function from the qgam package (v. 1.3.2; https://CRAN.R-project.org/package=qgam) in R (Fasiolo et al., 2021) to quantify time lag in incorporation/depletion. We specifically quantified 5th and 95th time percentiles for which individuals exhibited 90% change in their concentration values. To exclusively focus on time lag by accounting for inter-individual differences in initial and asymptotic values (expressed respectively by the Z and L random terms in the mixed-model approach), fitted values were standardized (0–1) between these two limits. At the individual scale, we have identified the minimum period of exposure time required for otoliths to incorporate/deplete 90% of Sr and Ba from the initial to the asymptotic modelled value (referred hereafter as t90).

Pearson's correlation coefficient for repeated measures was used to quantify the strength of association between individuals' kinematic parameters (k, L and Z) and either standard metabolic rate (SMR), routine metabolic rate (RMR) or 90% time lag (t90). Somatic growth (fish length at the end of the experiment) and otolithic growth (otolith mass) were similarly compared with SMR and RMR.

Observed elemental trends in otoliths (Sr/Ca and Ba/Ca ratios) were consistent with environmental conditions experienced by fish during the experiment (transfer between rivers and experimental enrichment; Fig. 3). In particular, abrupt changes in geochemical signatures (both uptake and depletion) were readily noticeable on otoliths. Uptake of strontium and uptake/depletion of barium followed the expected nonlinear trend, as modelled by re-parameterized analogues of the von Bertalanffy model. The overall mean response exhibited a lag response of several weeks, before reaching equilibrium (i.e. asymptotic value). Beyond this global trend and despite the controlled chemical environment, stable temperature regime and standardized diet data exhibited substantial inter-individual variations in the timing and magnitude of Sr/Ca and Ba/Ca responses, with a large portfolio of elemental composition at equilibrium (initial Z and asymptotic L values) and kinematics (k).

Fig. 3.

Trends in microchemical concentrations over time. Raw data for Sr/Ca (A) and Ba/Ca (B) ratios (dots), predicted individual values as grey lines (N=51) and mean trend (fixed effect in the mixed model) in red. Experimental time started after the first Alizarin Red S mark (3 months post-emergence).

Fig. 3.

Trends in microchemical concentrations over time. Raw data for Sr/Ca (A) and Ba/Ca (B) ratios (dots), predicted individual values as grey lines (N=51) and mean trend (fixed effect in the mixed model) in red. Experimental time started after the first Alizarin Red S mark (3 months post-emergence).

When focusing on the kinematics by accounting for equilibrium values, our results emphasized at the individual scale a highly variable minimum period required for otoliths to reach 90% of predicted equilibrium (Fig. 4). Overall, this time lag was at least twice as long for individuals with low uptake/depletion rate, compared with those with more rapid kinematics. More specifically, for strontium uptake, 90% of the predicted equilibrium was reached after only 1.1 months for the 5th percentile, whereas it took 2.4 months for the 95th percentile. Similarly, for barium uptake, a distinct time lag was apparent among individuals before its concentration reached a stable level. Saturation occurred after 0.3 and 0.9 months, respectively, for 5th and 95th pencentiles. Ultimately, for barium depletion, a stable Ba/Ca ratio was obtained within only 0.6 months for several individuals, whereas it took 1.1 months for others.

Fig. 4.

Quantile regression on predicted individual responses. Values were standardized (0–1) between initial and asymptotic values for (A) strontium increase, (B) barium increase and (C) barium decrease. Relative time required to reach 90% of complete equilibrium (dashed line) is depicted for 5th and 95th percentiles.

Fig. 4.

Quantile regression on predicted individual responses. Values were standardized (0–1) between initial and asymptotic values for (A) strontium increase, (B) barium increase and (C) barium decrease. Relative time required to reach 90% of complete equilibrium (dashed line) is depicted for 5th and 95th percentiles.

Given that SMR and RMR were highly correlated (ρ=0.81, P=2.8×10−24) in our data and reacted similarly, we subsequently focused on only RMR. The main result of our experiment showed that the energetic status of individuals (as measured by oxygen consumption) correlated with a shift in the characteristics of chemical data (Fig. 5; Fig. S2). This was particularly true for the kinematics of barium depletion and its relative time lag (k, ρ=0.64, P=5.9×10−13; t90, ρ=0.61, P=1.9×10−11) and to a lesser extent to Ba values at equilibrium (Z, ρ=0.22, P=0.027; L, ρ=0.42, P=1.3×10−5). It was also less well marked for barium uptake (Z, ρ=0.31, P=1.7×10−3; L, ρ=0.09, P=0.36; k, ρ=0.43, P=5.8×10−8; t90, ρ=0.35, P=3.6×10−4). In contrast, for strontium uptake, RMR correlated poorly with elemental composition at equilibrium (Z, ρ=0.25, P=0.013; L, ρ=0.28, P=3.4×10−3), kinematics (k, ρ=0.02, P=0.83) and time lag (t90, ρ=0.09, P=0.39). For a more detailed review of the results, an exhaustive list of repeated-measures correlations (and associated P-values) is provided in Fig. S2. Ultimately, both somatic and otolithic growth correlated poorly with SMR (ρ=0.31, P=0.024; ρ=−0.17, P=0.21, respectively) and RMR (ρ=−0.29, P=0.034; ρ=−0.27, P=0.051, respectively).

Fig. 5.

Repeated-measures correlation between metabolic (RMR only) and chemical components. Composition at equilibrium for Z and L, kinematics (k) and time lag (t90) for (A) strontium increase, (B) barium increase and (C) barium decrease. Linear least-squares regressions are in red; shaded regions represent 95% confidence intervals. N=51 individuals.

Fig. 5.

Repeated-measures correlation between metabolic (RMR only) and chemical components. Composition at equilibrium for Z and L, kinematics (k) and time lag (t90) for (A) strontium increase, (B) barium increase and (C) barium decrease. Linear least-squares regressions are in red; shaded regions represent 95% confidence intervals. N=51 individuals.

Overall, our data exhibited substantial inter-individual variations in the time-lag response of Sr/Ca and Ba/Ca signals, ranging from a few to several weeks, up to months. Individuals with the highest metabolic rate were more likely to record detailed (i.e. brief) temporal changes than individuals having lower metabolic values.

Time lag changes in otolith microchemistry

The hard acid elements such as Sr and Ba (Walther and Thorrold, 2006) are widely used in combination to unravel anadromous movements (Sturrock et al., 2012; Walther and Limburg, 2012) but also exhibit regional variations in freshwater depending on streambed geological composition and have emerged as useful to track movement patterns at a finer spatio-temporal scale. The usefulness of these markers mostly depends on their predictable spatio-temporal distribution as well as the comprehension of the regulatory mechanisms affecting their incorporation. Although prior works have comprehensively addressed the spatial and temporal variability in these chemical tracers (e.g. Walther and Thorrold, 2009), the temporal resolution has not been thoroughly investigated for these systems.

Ambient change in chemical composition can result from a shift in habitat use (i.e. movement between water masses of distinct composition). Alternatively, fishes can remain spatially static over time in complex and temporally dynamic environments with changes in water chemistry. This should be particularly relevant in estuaries with varying physical and chemical properties as short as a tidal cycle, owing to the mixing of marine and freshwater inputs (Dorval et al., 2005; Elsdon and Gillanders, 2006). In either case, for fish that undertake rapid movements between adjacent tributaries with different geochemical signatures, the presence of time lags indicates that the residence time in a given tributary may be too short for the environmental chemical signature to be properly recorded in the otolith. Consequently, particular Sr/Ca or Ba/Ca values could either represent equilibrium with the ambient water or transitional values at some unknown stage within a time-lag period (Macdonald and Crook, 2010). Likewise, analyses that incorporate otolith material laid down during weeks following habitat change will underestimate elemental concentration in the otolith (Elsdon and Gillanders, 2005b). The presence of a time lag limits the spatial and temporal resolution achievable (Chang and Geffen, 2013; Macdonald and Crook, 2010; Sturrock et al., 2012) and poses substantial challenges to reconstruct environmental histories of fish at the relevant temporal scale. It is therefore critical to determine how rapidly otolith composition changes in response to variation in water composition.

To date, although many studies have used otolith chemistry to examine movement patterns, few manipulative experiments have specifically attempted to resolve the precise temporal resolution of their methods. In addition, publications are strongly biased toward manipulative experiments in which elemental concentrations are artificially increased. The opposite (i.e. depletion) is rarely shown but of equal interest. Here, both otolithic signals were jointly investigated and rapidly reacted to changes in environmental concentrations but otolith composition did not stabilize for weeks. The decrease/increase in concentration in our time-resolved signal was always observed after only a few sampling time points, or with no delay. Given the analytical parameters or the femtosecond laser equipment, the estimated delay between two consecutive measures encompasses only a few days (depending on the individual and transect length), indicating that the otolithic signal rapidly reacts to changes in environmental concentrations. Habitat transitions should therefore be discernible after a relatively moderate/brief residence time but the otolith may not reflect ambient water levels until a few weeks have passed. On a practical basis, even if there is an instantaneous change in otolith composition along a transect, any laser beam of a given diameter would average a mixed proportion material on either side of the centre, introducing an apparent lag (Warburton et al., 2017). The importance of this lag is defined by growth increment widths relative to the laser diameter. Similarly, given the concentric three-dimensional nature of otoliths, depth-profiling laser ablation collects data through multiple vertical layers, potentially leading to a time lag (Hoover and Jones, 2013; Warburton et al., 2017). In addition, washout/purge of the ablation cell (a sealed chamber) also plays a role in avoiding contamination from previous ablations. Overall, operating parameters of LA-ICP-MS (including spot size, ablation depth, laser speed and washout rate) can artefactually affect the time-lag response and define the absolute minimum limitation on the temporal resolution that could be identified. In our analysis, we quantified the artefactual (analytically related) time lag to be ∼20 μm (Fig. S3), corresponding to about 6 days (assuming a constant growth; Fig. S1), an order of magnitude below the one related to inter-individual variations. In addition to time lags, our data emphasize that even if fish remain in a given environment long enough to reach equilibrium, individual variations in the kinematics may hinder accurate reconstructions of their environmental histories (Macdonald and Crook, 2010).

On a theoretical basis, elements with no major structural or physiological role (e.g. Ba) may depure (depletion of ions in the endolymph) more rapidly than those that are constituent parts of functioning enzymes or structural tissues (e.g. Sr), resulting in short turnover times (Sturrock et al., 2012). While the comparison between Sr and Ba in terms of depletion kinematics was not possible, the decrease in Ba was quite rapid, emphasizing that this element (with its short turnover time) is well suitable for studying detailed temporal changes. However, experiments were conducted using fish exposed to constant environments before and after abrupt changes. For analysis aimed at reconstructing dynamic change in the environment using chemical proxies further work is required to fully characterize the response of otolith chemistry to brief/intermittent environmental variations. Such pulses of ambient Ba/Ca simulate either upwelling or flood conditions in marine systems or riverine discharge in freshwater systems.

Various studies have also reported substantial time lags of the same magnitude in elemental uptake for other diadromous species: 30–40 days (Ba/Ca and Sr/Ca, respectively) for the Australian bass (Macquaria novemaculeata) across a salinity gradient (Macdonald and Crook, 2010); 20 days (Sr/Ca) in juvenile black bream (Acanthopagrus butcheri) (Elsdon and Gillanders, 2005b); 21 days (Sr/Ca) for juvenile largemouth bass (Micropterus salmoides) (Lowe et al., 2009); over 10 days (Ba/Ca, Sr/Ca) in juvenile Chinook salmon (Oncorhynchus tshawytscha) (Miller, 2011); over weeks for juvenile striped bass Morone saxatilis (Secor et al., 1995); and more than 50 days for juvenile pike Esox lucius (Engstedt et al., 2012). A brief overview of previously cited time-lag records obtained in juveniles does not exhibit clear analytical bias (see Table S2 for operating parameters of LA-ICP-MS and recorded time lags) but rather highlights inter-specific differences, depending on local conditions. Slight differences between studies may concomitantly rely on species-specific physiology (reviewed in Chang and Geffen, 2013), as well as ambient concentration. All these studies have emphasized that otolith Sr/Ca and Ba/Ca are linked to ambient concentrations, covering a wide range of concentrations, sometimes encompassing freshwater to marine values, with no evidence of saturation (Bath et al., 2000; Elsdon and Gillanders, 2003b; Reis-Santos et al., 2013; Secor et al., 1995). Hence, the complete time lag before equilibrium is likely related to both absolute and relative change in water concentration, in a potentially non-proportional manner, with possible interactions with salinity and temperature (Bath et al., 2000; Elsdon and Gillanders, 2003b; Miller, 2009). Since diet can also influence elemental uptake (Doubleday et al., 2013; Jaecks et al., 2016; Lin et al., 2007; Marohn et al., 2009; Milton and Chenery, 2001; Walther et al., 2010; Woodcock et al., 2012), concomitant changes in both alimentation and environmental conditions remain to be investigated. In the meantime, because relative growth asymptotically decreases through time for many teleost fishes (Jones, 2000), the temporal resolution attainable should intrinsically change over time for any given analytical resolution (e.g. constant speed of ICP-MS beam). A finer temporal resolution in chemical transects is therefore expected compared with older stages. Moreover, artefactual (analytically related) lags may be apparently ‘longer’ in older fish that grow more slowly. As a consequence, generalizations about the time-lag response of elemental incorporation is a priori difficult and investigations remain to be conducted in a species-specific manner (species-specific elemental incorporation reviewed in Chang and Geffen, 2013), at the relevant life history stages and on a case-by-case basis.

Metabolic influence on kinematics

While the concentration of trace elements predictably reflects environmental conditions, the interpretation of elemental signals is not straightforward since several confounding factors can act concomitantly in an interactive manner. Otolith composition is indeed a multifactorial component that is simultaneously influenced by the composition of the surrounding water and various physiological processes that can mediate the incorporation of available elements into the otolith. While specific regulatory processes are not fully elucidated, several iono-regulatory processes may be involved since depletion and element uptake into the otolith matrix from surrounding water and food involves the transport of dissolved ions over several membranes (Campana, 1999; Kalish, 1989, 1991; Melancon et al., 2009; Payan et al., 2004). In particular, this includes modulation of uptake in the blood, binding of metal elements by circulating plasma proteins that change ion availability (at least for soft acid ions that are more strongly associated with blood proteins; Grønkjær, 2016) and mechanisms of ion transport from external plasma across the saccular macula into the endolymph (Payan et al., 2004). Finally, the incorporation of an element into the otolith depends on its compatibility to bind directly into the crystal lattice, within interstitial spaces or to the organic matrix (Campana, 1999; Sturrock et al., 2012). In this context, otolith is bathed in a supersaturated endolymphatic solution that contains all the ionic precursors and proteins necessary for otolith accretion (Hüssy et al., 2020). On a theoretical basis, the secretion of most of these precursors is an energy-demanding process and therefore linked to the bioenergetic status of the fish, which in turn can explain the (indirect) link between uptake/depletion kinematics and SMR. On a practical basis, most studies supporting metabolic regulation are based on indirect evidence (e.g. Hüssy and Mosegaard, 2004; Kalish, 1989; Wright et al., 2001). For example, Hüssy and Mosegaard (2004), observed that otolith growth continues during a period of starvation while fish lose weight, suggesting that otolith accretion is more linked to the standard metabolic rate of the fish and is independent of somatic growth. Most recent experiments in which blood samples, physiological data and water chemistry were sampled over the course of a year directly highlight physiological influences for thiophilic elements with possible metabolic implications (Sturrock et al., 2015, 2014). In a distinct but related context, isotopic composition of otoliths (a proxy for trophic level) is known to depend on prey composition, with strong physiological modulation (Lueders-Dumont et al., 2020). This study is one of the few to experimentally address the endogenous effect of metabolism on the kinematics of elemental incorporation/depletion. Our results are unambiguous and fully in line with the growing evidence of physiological effects on otolith microchemistry.

Metabolic costs are central to individual energy budgets, making estimates of metabolic rate fundamental to understanding how an organism interacts with its environment (Treberg et al., 2016). Modelling otolith growth using a bioenergetic model in the framework of the dynamic energy budget (DEB; i.e. treating otoliths as ‘metabolic products’ of organism growth and maintenance processes) indeed provided insights into otolith growth (Fablet et al., 2011; Pecquerie et al., 2012). Interestingly, both incorporation and depletion are related to SMR, suggesting that roughly the same pathways are used and regulated for these two components, although differences in kinematics exist between the two elements. More generally, the observed kinematics derives from normal metabolic processes in tissues including synthesis and degradation. Surprisingly, initial (Z parameter) and asymptotic (L parameter) elemental ratios are less well related to SMR compared with the kinematics of incorporation/depletion. The reasons for such incongruence between elemental composition and kinematics is unclear. On a theoretical basis, the two should be linked: individuals with the highest metabolic activity should have more active ionocytes (Limburg et al., 2018), leading to higher transport/availability of the element to be incorporated in the otoliths. This would in turn increase both elemental composition and kinematics of incorporation/depletion. Previous work has already emphasized that the partition coefficients of a given element can be related to the bioenergetic status of the fish (Kalish, 1989, 1991; Sturrock et al., 2015, 2014) but in the absence of additional kinematic data, this discrepancy needs to be investigated further.

Implications for individual reconstruction

While experimental investigation should be carried out at the individual level to provide an appropriate framework for future field applications and movement reconstructions (Fromentin et al., 2009; Grønkjær, 2016), such experiments have largely over-neglected inter-individual differences. In general, a fish cohort experiencing similar migration pathways of identical duration is expected to record the same detailed temporal changes in their otoliths. Our data strongly suggest that this is not the case and the temporal resolution was in fact highly variable on an individual basis. In particular, it appeared that individuals with the highest metabolic rate were more likely to record detailed (i.e. brief) temporal changes than individuals having lower metabolic values. From a biological point of view, many animals perform extensive niche shifts because of their ability to utilize resources and avoid predation changes through their ontogeny (L'Abée-Lund et al., 1993). While the proximate mechanisms controlling the niche shift in fish remain partially unresolved, several studies have highlighted that movement patterns may be conditioned by the energetic status of the individuals, possibly in relation to metabolic activity (Dodson et al., 2013; Nevoux et al., 2019; Pettersson and Hedenström, 2000). Fast-growing individuals that maintain higher metabolic rates are energetically constrained at a younger age by limited food resources compared with slow growers (e.g. Metcalfe et al., 1989, 1992; Økland et al., 1993). For example, in Salmo trutta, food consumption and energy budgets for early migrants were more than four times higher than those of the stream residents (Forseth et al., 1999). As a consequence, the finest time resolution would be available for fast-growing and more active individuals. This clearly can bias investigation conducted at the population level, although we recognize that the degree of bias remains elusive. A time series composed of juvenile transects stitched together from fast-growing and slow-growing fish in an unknown proportion may lead to an overall underestimate of movement patterns if movements are only detected for fast-growing individuals.

Whatever the iono-regulatory explanation, inter-individual differences in physiologically related kinematics of elemental incorporation/depletion in otoliths have practical implications for the reconstruction of movement patterns, especially at a fine temporal scale (as illustrated in Fig. 6). Here, we used data from two individual fish with high or low kinematics to illustrate the marked differences in otolith signature over time and reconstructed environmental history when fish follow the exact same environmental sequence between habitats with sharply contrasted barium concentration (Fig. 6). Time-resolved chemical profiles from otoliths are conventionally analysed using zoning or regime-shift algorithms that quantitatively classify otolith sequences into fish environmental histories (Fablet et al., 2007; Hedger et al., 2008; Vignon, 2015). While globally correct (i.e. at the population level), using a mean temporal scale in these algorithms will confuse noise signal with a brief incursion between contiguous habitats having contrasted geochemical signatures for some individuals. Alternatively, algorithms should no longer be completely naïve (i.e. unsupervised, using original untagged data with no information; Fablet et al., 2007) but rather should be supervised by accommodating individual covariables. Comprehensive knowledge of inter-individual differences in the kinematics of elemental incorporation/depletion and of factors that can readily be used as a proxy will enable us to determine realistic and appropriate supervised statistical methods (for statistical discussion, see Vignon, 2015).

Fig. 6.

Simplified reconstruction of environmental history depending on an individual's kinematics. (A) Here, two fish are hypothesized to follow the exact same environmental sequence (using only two habitats with sharply contrasted barium concentrations). (B) Theoretical curves from the otolith are extracted from Fig. 4, using two individuals with contrasted kinematics [i.e. one with high kinematics (solid line) and the second with low kinematics (dashed line), corresponding to 5th and 95th percentiles]. Two thresholds are reported to discriminate areas with clear assignation (below which the signal typically corresponds to habitat 1, and under which the signal typically corresponds to habitat 2). (C) Many methodologies/algorithms exist to reconstruct environmental history and for this purpose, we used the threshold-based zonation algorithm with equivocal area from Vignon (2015). Although the figure is simplistic, it illustrates that individual variability in kinematics can affect reconstruction both in terms of habitat sequence and duration. For example, signal in the otolith for the last two incursions within the habitat with high barium concentration is under the detection limit and therefore completely neglected for the individual with the lowest kinematics.

Fig. 6.

Simplified reconstruction of environmental history depending on an individual's kinematics. (A) Here, two fish are hypothesized to follow the exact same environmental sequence (using only two habitats with sharply contrasted barium concentrations). (B) Theoretical curves from the otolith are extracted from Fig. 4, using two individuals with contrasted kinematics [i.e. one with high kinematics (solid line) and the second with low kinematics (dashed line), corresponding to 5th and 95th percentiles]. Two thresholds are reported to discriminate areas with clear assignation (below which the signal typically corresponds to habitat 1, and under which the signal typically corresponds to habitat 2). (C) Many methodologies/algorithms exist to reconstruct environmental history and for this purpose, we used the threshold-based zonation algorithm with equivocal area from Vignon (2015). Although the figure is simplistic, it illustrates that individual variability in kinematics can affect reconstruction both in terms of habitat sequence and duration. For example, signal in the otolith for the last two incursions within the habitat with high barium concentration is under the detection limit and therefore completely neglected for the individual with the lowest kinematics.

Lifetime physiological proxy

While the incorporation/depletion kinematics is linked to the physiological conditions of the fish, this component is not a steady-state condition and metabolic rate can change drastically throughout the lifespan of an individual (Treberg et al., 2016). In particular, osmoregulation can change over ontogeny (Tanner et al., 2015; Varsamos et al., 2001; Walther et al., 2010), leading to changes in plasma and endolymph composition that contribute to ontogenetic shifts in elemental composition (Fuiman and Hoff, 1995; Gallahar and Kingsford, 1992; Hüssy, 2008; Morales-Nin et al., 2005; Tomas et al., 2006; Tzeng, 1996). This could be particularly visible for some species at metamorphosis or settlement when most of these variables change (e.g. Chen et al., 2008; de Pontual et al., 2003; Otake et al., 1997; Sandin et al., 2005). Here, we cannot completely assume that punctually measured metabolic status remains stable over the complete course of the experiment (neither are they intrinsic, potentially life-long characteristics of the individual). Such potential change during ontogeny may partly explain the overall relatively low correlation between physiological and chemical components. As in most experimental studies dealing with otolith microchemistry, our results were obtained from juvenile fish. Considering age-related changes in elemental incorporation dynamics, however, interpretation of otolith signals in adult fishes based on observations from young life-history stages should be treated with caution (Sturrock et al., 2012; Walther et al., 2010). In addition, several other variables can concomitantly affect deposition of otolith carbonate over an individual's lifetime, including growth rate, stress, availability of nourishment, gonadal development, reproductive investment, ambient temperature, diurnal and seasonal cycles and activity levels (Friedland et al., 1998; Kalish, 1989, 1991; Radtke and Shafer, 1992; Walther et al., 2010). All these changing variables often play a substantive role and have the potential to complicate attempts to retrospectively identify individual fish movement at the relevant temporal scale over its lifespan.

In this context, the individual interpretation of otolith signal using the relevant temporal scale would greatly benefit from a retrospective tracking of change in the bioenergetic status of individuals over time. In particular, otolith growth may be used as a metabolic proxy. While numerous studies have experimentally linked standard metabolic rate to otolith growth (e.g. Wright, 1991; Wright et al., 2001; Yamamoto et al., 1998), the relationship may not be that simple in the field owing to the potential fitness-related costs associated with fast growth (Álvarez and Nicieza, 2005; Bang et al., 2007; Bochdansky et al., 2005). In addition, a high SMR may be a prerequisite for fast growth, but does not necessarily result in fast growth, revealing a complex relationship between these two components. Several observations further suggest that processes associated with the secretory capacity and thus the metabolic rate of the endolymphatic epithelium are the driving forces behind otolith growth (Hüssy, 2008). While metabolism is known to be correlated among various tissues from the same organism (Houlihan et al., 1986), the epithelium of fast-growing otoliths would have had to increase secretory activity, thereby increasing epithelium metabolism, which would lead to an uncoupling from global metabolic rate (as measured using oxygen consumption) (Hüssy, 2008). Such decoupling may partly explain the significant but relatively moderate relationship between SMR and the observed kinematics. Unfortunately, information on the growth dynamics of the secretory area and the saccular epithelium are virtually non-existent (Hüssy, 2008), and in the absence of any other alternative several authors have preferred using otolith growth or even fish length as a proxy of metabolism. Unfortunately, otolith growth depends on metabolic activity, regardless of whether that activity goes into somatic or reproductive growth, movement, foraging or maintenance (Limburg et al., 2018). In this context, several studies have emphasized the absence of a direct relationship between otolith and somatic growth (Baumann et al., 2005; Francis et al., 1992; Mosegaard et al., 1988; Wright, 1991), precluding the direct and reliable use of fish length as a proxy of metabolism. Our data indeed indicated a poor correlation between these components (i.e. all |ρ|<0.32) and also emphasized that incorporation/depletion kinematics are much more related to metabolism itself than otolith/somatic growth. Any attempt to use otolith to unravel metabolic components may poorly mirror quantification made based on direct physiological measurement. However, recent attempts have been focused on the direct use of otolith isotopic composition (e.g. δ13C and 26/24Mg ratio in aragonite) to infer time-resolved field metabolic rates (reviewed in Grønkjær, 2016; Limburg et al., 2018; Chung et al., 2019). These studies suggest there could be promising results in the near future.

Conclusion

In contrast to the use of otoliths as group-specific natural tags for stock discrimination, the reconstruction of environmental histories of individual fish depends on effectively coupling environmental changes with otolith chemistry (Elsdon et al., 2008). Beyond assessing anadromous movement at a seasonal/annual scale, untangling movement and habitat use patterns at finer temporal scales poses a greater challenge. In particular, the lack of a predictive and mechanistic understanding of the individual kinematics underlying ion incorporation/depletion still limits our fine-scale temporal interpretation of the chemical signal recorded in the otolith. Further kinetics knowledge is fundamental since it ultimately determines the spatio-temporal scope and the ecological questions that can be adequately resolved. Such consideration is also of primary importance since accurate identification of migratory pathways and habitat use underpins the effective management of exploited populations.

The authors thank the experimental facility ECP, INRA, 2018. The Ecology and Fish Population Biology Facility (https://doi.org/10.15454/1.5572402068944548E12) for fully supporting the experiment. We are grateful to various agency personnel of the ECOBIOP and IPREM laboratories who assisted in the course of the experiment, metabolic and otolith analysis. The authors are also grateful to Stéphane Glise, Frédéric Lange, and François Guéraud for their help in maintaining fish during the experiment, and to two anonymous reviewers who made constructive suggestions on the first version of the manuscript.

Author contributions

Conceptualization: M.V., H.T., J.-C.A., G.B.; Methodology: M.V., H.T., J.-C.A., G.B.; Validation: H.T., J.-C.A., C.P., P.C.-H., G.B.; Formal analysis: M.V.; Investigation: M.V., H.T., J.-C.A., J.R.; Resources: C.P., J.R., P.C.-H., E.H.; Data curation: M.V., H.T., C.P., P.C.-H.; Writing - original draft: M.V.; Writing - review & editing: H.T., J.-C.A., G.B.; Visualization: M.V.; Supervision: G.B.; Project administration: M.V.; Funding acquisition: G.B.

Funding

This study was funded by the regional water agency (Agence de l'Eau Adour-Garonne) and the regional administration (Conseil Départemental des Pyrénées Atlantiques) as part of the CARPOMIBA project.

Data availability

Data are available from the INRAE Digital Repository (https://entrepot.recherche.data.gouv.fr/dataverse/inrae): https://doi.org/10.57745/1VSDOG.

Allen
,
A. M.
and
Singh
,
N. J.
(
2016
).
Linking movement ecology with wildlife management and conservation
.
Front. Ecol. Environ.
3
,
155
.
Álvarez
,
D.
and
Nicieza
,
A. G.
(
2005
).
Is metabolic rate a reliable predictor of growth and survival of brown trout (Salmo trutta) in the wild?
Can. J. Fish. Aquat. Sci.
62
,
643
-
649
.
Aymes
,
J. C.
,
Vignon
,
M.
,
Beall
,
E.
,
Gueraud
,
F.
and
Gaudin
,
P.
(
2016
).
Age validation of the Kerguelen Islands brown trout, Salmo trutta L., and selection of the otolith optimal zone for investigating chronological data series
.
Fish. Res.
176
,
22
-
29
.
Bang
,
A.
,
Grønkjær
,
P.
and
Folkvord
,
A.
(
2007
).
Possible fitness costs of high and low standard metabolic rates in larval herring Clupea harengus, as determined by otolith microstructure
.
Mar. Ecol. Prog. Ser.
331
,
233
-
242
.
Bath
,
G. E.
,
Thorrold
,
S. R.
,
Jones
,
C. M.
,
Campana
,
S. E.
,
McLaren
,
J. W.
and
Lam
,
J. W. H.
(
2000
).
Strontium and barium uptake in aragonitic otoliths of marine fish
.
Geochim. Cosmochem. Acta
64
,
1705
-
1714
.
Baumann
,
H.
,
Peck
,
M. A.
and
Herrmann
,
J.-P.
(
2005
).
Short-term decoupling of otolith and somatic growth induced by food level changes in postlarval Baltic sprat, Sprattus sprattus
.
Mar. Freshw. Res.
56
,
539
-
547
.
Begg
,
G. A.
,
Campana
,
S. E.
,
Fowler
,
A. J.
and
Suthers
,
I. M.
(
2005
).
Otolith research and application: current directions in innovation and implementation
.
Mar. Freshw. Res.
56
,
477
-
483
.
Bochdansky
,
A. B.
,
Grønkjær
,
P.
,
Herra
,
T. P.
and
Leggett
,
W. C.
(
2005
).
Experimental evidence for selection against larvae with high metabolic rates in a food limiting environment
.
Mar. Biol.
147
,
1413
-
1417
.
Bradbury
,
I. R.
,
Campana
,
S. E.
and
Bentzen
,
P.
(
2008
).
Otolith elemental composition and adult tagging reveal spawning site fidelity and estuarine dependency in rainbow smelt
.
Mar. Ecol. Prog. Ser.
368
,
255
-
268
.
Brennan
,
S. R.
,
Zimmerman
,
C. E.
,
Fernandez
,
D. P.
,
Cerling
,
T. E.
,
McPhee
,
M. V.
and
Wooller
,
M. J.
(
2015
).
Strontium isotopes delineate fine-scale natal origins and migration histories of Pacific salmon
.
Sci. Adv.
1
,
e1400124
.
Brown
,
R. J.
and
Severin
,
K. P.
(
2009
).
Otolith chemistry analyses indicate that water Sr:Ca is the primary factor influencing otolith Sr:Ca for freshwater and diadromous fish but not for marine fish
.
Can. J. Fish. Aquat. Sci.
66
,
1790
-
1808
.
Campana
,
S. E.
(
1999
).
Chemistry and composition of fish otoliths: pathways, mechanisms and applications
.
Mar. Ecol. Prog. Ser.
188
,
263
-
297
.
Campana
,
S. E.
and
Thorrold
,
S. R.
(
2001
).
Otoliths, increments, and elements: keys to a comprehensive understanding of fish populations?
Can. J. Fish. Aquat. Sci.
58
,
30
-
38
.
Carlson
,
A. K.
,
Phelps
,
Q. E.
and
Graeb
,
B. D. S.
(
2017
).
Chemistry to conservation: using otoliths to advance recreational and commercial fisheries management
.
J. Fish Biol.
90
,
505
-
527
.
Caudron
,
A.
and
Champigneulle
,
A.
(
2006
).
Technique de fluoromarquage en masse à grande échelle des otolithes d'alevins vésiculés de truite commune (Salmo trutta) à l'aide de l'alizarine red S
.
Cybium
30
,
65
-
72
.
Chang
,
M.-Y.
and
Geffen
,
A. J.
(
2013
).
Taxonomic and geographic influences on fish otolith microchemistry
.
Fish Fish.
14
,
458
-
492
.
Chen
,
H. L.
,
Shen
,
K. N.
,
Chang
,
C. W.
,
Iizuka
,
Y.
and
Tzeng
,
W. N.
(
2008
).
Effects of water temperature, salinity and feeding regimes on metamorphosis, growth and otolith Sr:Ca ratios of Megalops cyprinoides leptocephali
.
Aquat. Biol.
3
,
41
-
50
.
Chung
,
M.-T.
,
Trueman
,
C. N.
,
Aanestad Godiksen
,
J.
and
Grønkjær
,
P.
(
2019
).
Otolith δ13C values as a metabolic proxy: approaches and mechanical underpinnings
.
Mar. Freshw. Res.
70
,
1747
-
1756
.
Cooke
,
S. J.
,
Midwood
,
J. D.
,
Thiem
,
J. D.
,
Klimley
,
P.
,
Lucas
,
M. C.
,
Thorstad
,
E. B.
,
Eiler
,
J.
,
Holbrook
,
C.
and
Ebner
,
B. C.
(
2013
).
Tracking animals in freshwater with electronic tags: past, present and future
.
Anim. Biotelemetry
1
,
5
.
Cooke
,
S. J.
,
Martins
,
E. G.
,
Struthers
,
D. P.
,
Gutowsky
,
L. F. G.
,
Power
,
M.
,
Doka
,
S. E.
,
Dettmers
,
J. M.
,
Crook
,
D. A.
,
Lucas
,
M. C.
,
Holbrook
,
C. M.
et al. 
(
2016
).
A moving target—incorporating knowledge of the spatial ecology of fish into the assessment and management of freshwater fish populations
.
Environ. Monit. Assess.
188
,
239
.
de Pontual
,
H.
,
Lagardère
,
F.
,
Amara
,
R.
,
Bohn
,
M.
,
Ogor
,
A.
(
2003
).
Influence of ontogenetic and environmental changes in the otolith microchemistry of juvenile sole (Solea solea)
.
J. Sea Res.
50
,
199
-
211
.
Dodson
,
J. J.
,
Aubin-Horth
,
N.
,
Thériault
,
V.
and
Páez
,
D. J.
(
2013
).
The evolutionary ecology of alternative migratory tactics in salmonid fishes
.
Biol. Rev.
88
,
602
-
625
.
Dorval
,
E.
,
Jones
,
C. M.
and
Hannigan
,
R.
(
2005
).
Chemistry of surface waters: distinguishing fine-scale differences in sea grass habitats of Chesapeake Bay
.
Limnol. Oceanogr.
50
,
1073
-
1083
.
Doubleday
,
Z.
,
Izzo
,
C.
,
Woodcock
,
S.
and
Gillanders
,
B. M.
(
2013
).
Relative contribution of water and diet to otolith chemistry in freshwater fish
.
Aquat. Biol.
18
,
271
-
280
.
Elsdon
,
T. S.
and
Gillanders
,
B. M.
(
2003a
).
Reconstructing migratory patterns of fish based on environmental influences on otolith chemistry
.
Rev. Fish Biol. Fish.
13
,
217
-
235
.
Elsdon
,
T. S.
and
Gillanders
,
B. M.
(
2003b
).
Relationship between water and otolith elemental concentrations in juvenile fish, Acanthopagrus butcheri
.
Mar. Ecol. Prog. Ser.
260
,
263
-
272
.
Elsdon
,
T. S.
and
Gillanders
,
B. M.
(
2005a
).
Alternative life-history patterns of estuarine fish: barium in otoliths elucidates freshwater residency
.
Can. J. Fish. Aquat. Sci.
62
,
1143
-
1152
.
Elsdon
,
T. S.
and
Gillanders
,
B. M.
(
2005b
).
Strontium incorporation into calcified structures: separating the effects of ambient water concentration and exposure time
.
Mar. Ecol. Prog. Ser.
285
,
233
-
243
.
Elsdon
,
T. S.
and
Gillanders
,
B. M.
(
2006
).
Temporal variability in strontium, calcium, barium, and manganese in estuaries: implications for reconstructing environmental histories of fish from chemicals in calcified structures
.
Estuar. Coast. Shelf Sci.
66
,
147
-
156
.
Elsdon
,
T. S.
,
Wells
,
B. K.
,
Campana
,
S. E.
,
Gillanders
,
B. M.
,
Jones
,
C. M.
,
Limburg
,
K. E.
,
Secor
,
D. H.
,
Thorrold
,
S. R.
and
Walther
,
B. D.
(
2008
).
Otolith chemistry to describe movements and life-history parameters of fishes: hypotheses, assumptions, limitations and inferences
.
Oceanogr. Mar. Biol.
46
,
297
-
330
.
Engstedt
,
O.
,
Koch-Schmidt
,
P.
and
Larsson
,
P.
(
2012
).
Strontium (Sr) uptake from water and food in otoliths of juvenile pike (Esox lucius L.)
.
J. Exp. Mar. Biol. Ecol.
418-419
,
69
-
74
.
Fablet
,
R.
,
Daverat
,
F.
and
De Pontual
,
H.
(
2007
).
Unsupervised Bayesian reconstruction of individual life histories from otolith signatures: case study of Sr:Ca transects of eel (Anguilla anguilla) otoliths
.
Can. J. Fish. Aquat. Sci.
64
,
152
-
165
.
Fablet
,
R.
,
Pecquerie
,
L.
,
de Pontual
,
H.
,
Høie
,
H.
,
Millner
,
R.
,
Mosegaard
,
H.
and
Kooijman
,
S. A. L. M.
(
2011
).
Shedding light on fish otolith biomineralization using a bioenergetic approach
.
PLoS One
6
,
e27055
.
Fasiolo
,
M.
,
Wood
,
S. N.
,
Zaffran
,
M.
,
Nedellec
,
R.
and
Goude
,
Y.
(
2021
).
Fast calibrated additive quantile regression
.
J. Am. Stat. Assoc.
116
,
1402
-
1412
.
Forseth
,
T.
,
Naesje
,
T. F.
,
Jonsson
,
B.
and
Hårsaker
,
K.
(
1999
).
Juvenile migration in brown trout: a consequence of energetic state
.
J. Anim. Ecol.
68
,
783
-
793
.
Francis
,
M. P.
,
Williams
,
M. W.
,
Pryce
,
A. C.
,
Pollard
,
S.
and
Scott
,
S. G.
(
1992
).
Uncoupling of otolith and somatic growth in Pagrus auratus (Sparidae)
.
Fish. Bull.
91
,
159
-
164
.
Fraser
,
K. C.
,
Davies
,
K. T. A.
,
Davy
,
C. M.
,
Ford
,
A. T.
,
Flockhart
,
D. T. T.
and
Martins
,
E. G.
(
2018
).
Tracking the conservation promise of movement ecology
.
Front. Ecol. Environ.
6
,
150
.
Friedland
,
K. D.
,
Reddin
,
D. G.
,
Shimizu
,
N.
,
Haas
,
R. E.
and
Youngson
,
A. F.
(
1998
).
Strontium:calcium ratios in Atlantic salmon (Salmo salar) otoliths and observations on growth and maturation
.
Can. J. Fish. Aquat. Sci.
55
,
1158
-
1168
.
Fromentin
,
J. M.
,
Ernande
,
B.
,
Fablet
,
R.
and
de Pontual
,
H.
(
2009
).
Importance and future of individual markers for the ecosystem approach to fisheries
.
Aquat. Living Resour.
22
,
395
-
408
.
Fuiman
,
L. A.
and
Hoff
,
G. R.
(
1995
).
Natural variation in elemental composition of sagittae from red drum
.
J. Fish Biol.
47
,
940
-
955
.
Gallahar
,
N. K.
and
Kingsford
,
M. J.
(
1992
).
Patterns of increment width and strontium:calcium ratios in otoliths of juvenile rock blackfish, Girella elevara (M.)
.
J. Fish Biol.
41
,
749
-
763
.
Gemperline
,
P. J.
,
Rulifson
,
R. A.
and
Paramore
,
L.
(
2002
).
Multi-way analysis of trace elements in fish otoliths to track migratory patterns
.
Chemometr. Intell. Lab. Syst.
60
,
135
-
146
.
Gillanders
,
B. M.
(
2005
).
Using elemental chemistry of fish otoliths to determine connectivity between estuarine and coastal habitats
.
Estuar. Coast. Shelf Sci.
64
,
47
-
57
.
Gillanders
,
B. M.
and
Kingsford
,
M. J.
(
2003
).
Spatial variation in elemental composition of otoliths of three species of fish (family Sparidae)
.
Estuar. Coast. Shelf Sci.
57
,
1049
-
1064
.
Grønkjær
,
P.
(
2016
).
Otoliths as individual indicators: a reappraisal of the link between fish physiology and otolith characteristics
.
Mar. Freshw. Res.
67
,
881
-
888
.
Hedger
,
R. D.
,
Atkinson
,
P. M.
,
Thibault
,
I.
and
Dodson
,
J. J.
(
2008
).
A quantitative approach for classifying fish otolith strontium: calcium sequences into environmental histories
.
Ecol. Inform.
3
,
207
-
217
.
Houlihan
,
D. F.
,
McMillan
,
D. N.
and
Laurent
,
P.
(
1986
).
Growth rates, protein synthesis, and protein degradation rates in rainbowtrout: effects of body size
.
Physiol. Zool.
59
,
482
-
493
.
Hoover
,
R. R.
and
Jones
,
C. M.
(
2013
).
Effect of laser ablation depth in otolith life history scans
.
Mar. Ecol. Prog. Ser.
486
,
247
-
256
.
Hüssy
,
K.
(
2008
).
Otolith accretion rates: does size really matter?
J. Exp. Mar. Biol. Ecol.
362
,
131
-
136
.
Hüssy
,
K.
and
Mosegaard
,
H.
(
2004
).
Atlantic cod (Gadus morhua) growth and otolith accretion characteristics modelled in a bioenergetics context
.
Can. J. Fish. Aquat. Sci.
61
,
1021
-
1031
.
Hüssy
,
K.
,
Limburg
,
K. E.
,
de Pontual
,
H.
,
Thomas
,
O. R. B.
,
Cook
,
P. K.
,
Heimbrand
,
Y.
,
Blass
,
M.
and
Sturrock
,
A. M.
(
2020
).
Trace element patterns in otoliths: the role of biomineralization
.
Rev. Fish. Sci. Aquac.
29
,
445
-
477
.
Jaecks
,
T.
,
Bond
,
M. H.
and
Quinn
,
T. P.
(
2016
).
Can dietary reliance on Pacific salmon eggs create otolith Sr/Ca signatures that mimic anadromy in resident salmonids?
Environ. Biol. Fishes
99
,
237
-
247
.
Jones
,
C. M.
(
2000
).
Fitting growth curves to retrospective size-at-age data
.
Fish. Res.
46
,
123
-
129
.
Kalish
,
J. M.
(
1989
).
Otolith microchemistry: validation of the effects of physiology, age and environment on otolith composition
.
J. Exp. Mar. Biol. Ecol.
132
,
151
-
178
.
Kalish
,
J. M.
(
1991
).
Determinants of otolith chemistry: seasonal variation in the composition of blood plasma, endolymph and otoliths of bearded rock cod Pseudophycis barbatus
.
Mar. Ecol. Prog. Ser.
74
,
137
-
159
.
Kalish
,
J. M.
(
1992
).
Formation of a stress-induced chemical check in fish otoliths
.
J. Exp. Mar. Biol. Ecol.
162
,
265
-
277
.
Kennedy
,
B. P.
,
Blum
,
J. D.
,
Folt
,
C. L.
and
Nislow
,
K. H.
(
2000
).
Using natural strontium isotopic signatures as fish markers: methodology and application
.
Can. J. Fish. Aquat. Sci.
57
,
2280
-
2292
.
L'Abée-Lund
,
J. H.
,
Langeland
,
A. L.
and
Jonsson
,
B.
and
Ugedal
,
O.
(
1993
).
Spatial segregation by age and size in brown trout and Arctic charr: a trade-off between feeding possibilities and risk of predation
.
J. Anim. Ecol.
62
,
160
-
168
.
Limburg
,
K. E.
(
1995
).
Otolith strontium traces environmental history of subyearling American shad Alosa sapidissima
.
Mar. Ecol. Prog. Ser.
119
,
25
-
35
.
Limburg
,
K. E.
,
Wuenschel
,
M. J.
,
Hüssy
,
K.
,
Heimbrand
,
Y.
and
Samson
,
M.
(
2018
).
Making the otolith magnesium chemical calendar-clock tick: plausible mechanism and empirical evidence
.
Rev. Fish. Sci. Aquac.
26
,
479
-
493
.
Lin
,
S.-H.
,
Chang
,
C.-W.
,
Iizuka
,
Y.
and
Tzeng
,
W. N.
(
2007
).
Salinities, not diets, affect strontium/calcium ratios in otoliths of Anguilla japonica
.
J. Exp. Mar. Biol. Ecol.
341
,
254
-
263
.
Lord
,
C.
,
Tabouret
,
H.
,
Claverie
,
F.
,
Pécheyran
,
C.
and
Keith
,
P.
(
2011
).
Femtosecond laser ablation ICP-MS measurement of otolith Sr:Ca and Ba:Ca composition reveal differential use of freshwater habitats for three amphidromous Sicyopterus (Teleostei: Gobioidei: Sicydiinae) species
.
J. Fish Biol.
79
,
1304
-
1321
.
Lowe
,
M. R.
,
DeVries
,
D. R.
,
Wright
,
R. A.
,
Ludsin
,
S. A.
and
Fryer
,
B. J.
(
2009
).
Coastal largemouth bass (Micropterus salmoides) movement in response to changing salinity
.
Can. J. Fish. Aquat. Sci.
66
,
2174
-
2188
.
Lucas
,
M. C.
and
Baras
,
E.
(
2000
).
Methods for studying the spatial behaviour of freshwater fishes in the natural environment
.
Fish Fish.
1
,
238
-
316
.
Lueders-Dumont
,
J. A.
,
Sigman
,
D. M.
,
Johnson
,
B. J.
,
Jensen
,
O. P.
,
Oleynik
,
S.
and
Ward
,
B. B.
(
2020
).
Comparison of the isotopic composition of fish otolith-bound organic N with host tissue
.
Can. J. Fish. Aquat. Sci.
77
,
264
-
275
.
Macdonald
,
J. I.
and
Crook
,
D. A.
(
2010
).
Variability in Sr:Ca and Ba:Ca ratios in water and fish otoliths across an estuarine salinity gradient
.
Mar. Ecol. Prog. Ser.
413
,
147
-
161
.
Marklevitz
,
S. A. C.
,
Fryer
,
B. J.
,
Johnson
,
J.
,
Gonder
,
D.
and
Morbey
,
Y. E.
(
2016
).
Otolith microchemistry reveals spatio-temporal heterogeneity of natal sources and inter-basin migrations of Chinook salmon in Lake Huron
.
J. Gt. Lakes Res.
42
,
668
-
677
.
Marohn
,
L.
,
Prigge
,
E.
,
Zumholz
,
K.
,
Klügel
,
A.
,
Anders
,
H.
and
Hanel
,
R.
(
2009
).
Dietary effects on multi-element composition of European eel (Anguilla anguilla) otoliths
.
Mar. Biol.
156
,
927
-
933
.
Martin
,
J.
,
Bareille
,
G.
,
Berail
,
S.
,
Pecheyran
,
C.
,
Daverat
,
F.
,
Bru
,
N.
,
Tabouret
,
H.
and
Donard
,
O.
(
2013a
).
Spatial and temporal variations in otolith chemistry and relationships with water chemistry: a useful tool to distinguish Atlantic salmon Salmo salar parr from different natal streams
.
J. Fish Biol.
82
,
1556
-
1581
.
Martin
,
J.
,
Bareille
,
G.
,
Berail
,
S.
,
Pécheyran
,
C.
,
Gueraud
,
F.
,
Lange
,
F.
,
Daverat
,
F.
,
Bru
,
N.
,
Beall
,
E.
,
Barracou
,
D.
et al. 
(
2013b
).
Persistence of a southern Atlantic salmon population: diversity of natal origins from otolith elemental and Sr isotopic signatures
.
Can. J. Fish. Aquat. Sci.
70
,
182
-
197
.
Melancon
,
S.
,
Fryer
,
B. J.
and
Markham
,
J. L.
(
2009
).
Chemical analysis of endolymph and the growing otolith: fractionation of metals in freshwater fish species
.
Environ. Toxicol. Chem.
28
,
1279
-
1287
.
Mercier
,
L.
,
Darnaude
,
A. M.
,
Bruguier
,
O.
,
Vasconcelos
,
R. P.
,
Cabral
,
H. N.
,
Costa
,
M. J.
,
Lara
,
M.
,
Jones
,
D. L.
and
Mouillot
,
D.
(
2011
).
Selecting statistical models and variable combinations for optimal classification using otolith microchemistry
.
Ecol. Appl.
21
,
1352
-
1364
.
Metcalfe
,
N. B.
,
Huntingford
,
F. A.
,
Graham
,
W. D.
and
Thorpe
,
J. E.
(
1989
).
Early social status and the development of life-history strategies in Atlantic salmon
.
Proc. R. Soc. B
236
,
7
-
19
.
Metcalfe
,
N. B.
,
Wright
,
P. J.
and
Thorpe
,
J. E.
(
1992
).
Relationships between social status, otolith size at first feeding and subsequent growth in Atlantic salmon (Salmo salar)
.
J. Anim. Ecol.
61
,
585
-
589
.
Miller
,
J. A.
(
2009
).
The effects of temperature and water concentration on the otolith incorporation of barium and manganese in black rockfish Sebastes melanops
.
J. Fish Biol.
75
,
39
-
60
.
Miller
,
J. A.
(
2011
).
Effects of water temperature and barium concentration on otolith composition along a salinity gradient: implications for migratory reconstructions
.
J. Exp. Mar. Biol. Ecol.
405
,
42
-
52
.
Milton
,
D. A.
and
Chenery
,
S. R.
(
2001
).
Sources and uptake of trace metals in otoliths of juvenile barramundi (Lates calcarifer)
.
J. Exp. Mar. Biol. Ecol.
264
,
47
-
65
.
Morales-Nin
,
B.
,
Swan
,
S. C.
,
Gordon
,
J. D. M.
,
Palmer
,
M.
,
Geffen
,
A. J.
,
Shimmield
,
T.
and
Sawyer
,
T.
(
2005
).
Age-related trends in otolith chemistry of Merluccius merluccius from the northeastern Atlantic Ocean and the western Mediterranean Sea
.
Mar. Freshw. Res.
56
,
599
-
607
.
Mosegaard
,
H.
,
Svedäng
,
H.
and
Taberman
,
K.
(
1988
).
Uncoupling of somatic and otolith growth rates in Arctic Char (Salvelinus alpinus) as an effect of differences in temperature response
.
Can. J. Fish. Aquat. Sci.
45
,
1514
-
1524
.
Nevoux
,
M.
,
Finstad
,
B.
,
Davidsen
,
J. G.
,
Finlay
,
R.
,
Josset
,
Q.
,
Poole
,
R.
,
Höjesjö
,
J.
,
Aarestrup
,
K.
,
Persson
,
L.
,
Tolvanen
,
O.
et al. 
(
2019
).
Environmental influences of life history strategies in partial anadromous brown trout (Salmo trutta, Salmonidae)
.
Fish Fish.
20
,
1051
-
1081
.
Økland
,
F.
,
JONSSON
,
B.
,
Jensen
,
A.
and
Hansen
,
L. P.
(
1993
).
Is there a threshold size regulating smolt size in brown trout and Atlantic salmon?
J. Fish Biol.
42
,
541
-
555
.
Otake
,
T.
,
Ishii
,
T.
,
Ishii
,
T.
,
Nakahara
,
M.
and
Nakamura
,
R.
(
1997
).
Changes in otolith strontium:calcium ratios in metamorphosing Conger myriaster leptocephali
.
Mar. Biol.
128
,
565
-
572
.
Payan
,
P.
,
De Pontual
,
H.
,
Boeuf
,
G.
and
Mayer-Gostan
,
N.
(
2004
).
Endolymph chemistry and otolith growth in fish
.
Comptes Rendus Palevol.
3
,
535
-
547
.
Pecquerie
,
L.
,
Fablet
,
R.
,
de Pontual
,
H.
,
Bonhommeau
,
S.
,
Alunno-Bruscia
,
M.
,
Petitgas
,
P.
and
Kooijman
,
S. A. L. M.
(
2012
).
Reconstructing individual food and growth histories from biogenic carbonates
.
Mar. Ecol. Prog. Ser.
447
,
151
-
164
.
Pettersson
,
L. B.
and
Hedenström
,
A.
(
2000
).
Energetics, cost reduction and functional consequences of fish morphology
.
Proc. R. Soc. Lond. Ser. B
267
,
759
-
764
.
Radtke
,
R. L.
and
Shafer
,
D. J.
(
1992
).
Environmental sensitivity of fish otolith microchemistry
.
Aust. J. Mar. Freshw. Res.
43
,
935
-
951
.
Ramsay
,
A. L.
,
Milner
,
N. J.
,
Hughes
,
R. N.
and
McCarthy
,
I. D.
(
2011
).
Comparison of the performance of scale and otolith microchemistry as fisheries research tools in a small upland catchment
.
Can. J. Fish. Aquat. Sci.
68
,
823
-
833
.
Régnier
,
T.
,
Bolliet
,
V.
,
Labonne
,
J.
and
Gaudin
,
P.
(
2010
).
Assessing maternal effects on metabolic rate dynamics along early development in brown trout (Salmo trutta): an individual-based approach
.
J. Comp. Physiol. B
180
,
25
-
31
.
Reis-Santos
,
P.
,
Tanner
,
S. E.
,
Elsdon
,
T. S.
,
Cabral
,
H. N.
and
Gillanders
,
B. M.
(
2013
).
Effects of temperature, salinity and water composition on otolith elemental incorporation of Dicentrarchus labrax
.
J. Exp. Mar. Biol. Ecol.
446
,
245
-
252
.
Sandin
,
S. A.
,
Regetz
,
J.
and
Hamilton
,
S. L.
(
2005
).
Testing larval fish dispersal hypotheses using maximum likelihood analysis of otolith chemistry data
.
Mar. Freshw. Res.
56
,
725
-
734
.
Secor
,
D. H.
,
Henderson-Arzapalo
,
A.
and
Piccoli
,
P. M.
(
1995
).
Can otolith microchemistry chart patterns of migration and habitat utilization in anadromous fishes?
J. Exp. Mar. Biol. Ecol.
192
,
15
-
33
.
Sturgeon
,
R. E.
,
Willie
,
S. N.
,
Yang
,
L.
,
Greenberg
,
R.
,
Spatz
,
R. O.
,
Chen
,
Z.
,
Scriver
,
C.
,
Clancy
,
V.
,
Lam
,
J. W.
and
Thorrold
,
S.
(
2005
).
Certification of a fish otolith reference material in support of quality assurance for trace element analysis
J. Anal. At. Spectrom.
20
,
1067
-
1071
.
Sturrock
,
A. M.
,
Trueman
,
C. N.
,
Darnaude
,
A. M.
and
Hunter
,
E.
(
2012
).
Can otolith elemental chemistry retrospectively track migrations in fully marine fishes?
J. Fish Biol.
81
,
766
-
795
.
Sturrock
,
A. M.
,
Trueman
,
C. N.
,
Milton
,
J. A.
,
Waring
,
C. P.
,
Cooper
,
M. J.
and
Hunter
,
E.
(
2014
).
Physiological influences can outweigh environmental signals in otolith microchemistry research
.
Mar. Ecol. Prog. Ser.
500
,
245
-
264
.
Sturrock
,
A. M.
,
Hunter
,
E.
,
Milton
,
J. A.
,
Johnson
,
R. C.
,
Waring
,
C. P.
,
Trueman
,
C. N.
and
Leder
,
E.
(
2015
).
Quantifying physiological influences on otolith microchemistry
.
Methods Ecol. Evol.
6
,
806
-
816
.
Svedäng
,
H.
,
André
,
C.
,
Jonsson
,
P.
,
Elfman
,
M.
and
Limburg
,
K.
(
2010
).
Migratory behaviour and otolith chemistry suggest fine-scale sub-population structure within a genetically homogenous Atlantic Cod population
.
Environ. Biol. Fishes
89
,
383
-
397
.
Tanner
,
S.
,
Reis-Santos
,
P.
and
Cabral
,
H.
(
2015
).
Otolith chemistry in stock delineation: a brief overview, current challenges and future prospects
.
Fish. Res.
173
,
206
-
213
.
Teichert
,
N.
,
Tabouret
,
H.
,
Lagarde
,
R.
,
Grondin
,
H.
,
Ponton
,
D.
,
Pécheyran
,
C.
and
Bareille
,
G.
(
2018
).
Site fidelity and movements of an amphidromous goby revealed by otolith multi-elemental signatures along a tropical watershed
.
Ecol. Freshw. Fish
27
,
834
-
846
.
Thorrold
,
S. R.
,
Latkoczy
,
C.
,
Swart
,
P. K.
and
Jones
,
C. M.
(
2001
).
Natal homing in a marine fish metapopulation
.
Science
291
,
297
-
299
.
Thresher
,
R. E.
(
1999
).
Elemental composition of otoliths as a stock delineator in fishes
.
Fish. Res.
43
,
165
-
204
.
Tomas
,
J.
,
Geffen
,
A. J.
,
Millner
,
R. S.
,
Pineiro
,
C. G.
and
Tserpes
,
G.
(
2006
).
Elemental composition of otolith growth marks in three geographically separated populations of European hake (Merluccius merluccius)
.
Mar. Biol.
148
,
1399
-
1413
.
Treberg
,
J. R.
,
Killen
,
S. S.
,
MacCormack
,
T. J.
,
Lamarre
,
S. G.
and
Enderse
,
E. C.
(
2016
).
Estimates of metabolic rate and major constituents of metabolic demand in fishes under field conditions: methods, proxies, and new perspectives?
Comp. Biochem. Physiol. A
202
,
10
-
22
.
Tzeng
,
W.-N.
(
1996
).
Effects of salinity gradient and ontogenetic shift on strontium: calcium ratio in otolith of the Japanese eel, Anguilla japonica Temminck & Schlegel
.
J. Exp. Mar. Biol. Ecol.
199
,
111
-
122
.
Varsamos
,
S.
,
Connes
,
R.
,
Diaz
,
J. P.
,
Barnabé
,
G.
and
Charmantier
,
G.
(
2001
).
Ontogeny of osmoregulation in the European sea bass Dicentrarchus labrax L
.
Mar. Biol.
138
,
909
-
915
.
Vignon
,
M.
(
2015
).
Extracting environmental histories from sclerochronological structures - Recursive partitioning as a mean to explore multielemental composition of fish otolith
.
Ecol. Inform.
30
,
159
-
169
.
Walther
,
B. D.
and
Limburg
,
K. E.
(
2012
).
The use of otolith chemistry to characterize diadromous migrations
.
J. Fish Biol.
81
,
796
-
825
.
Walther
,
B. D.
and
Thorrold
,
S. R.
(
2006
).
Water, not food, contributes the majority of strontium and barium deposited in the otoliths of a marine fish
.
Mar. Ecol. Prog. Ser.
311
,
125
-
130
.
Walther
,
B. D.
and
Thorrold
,
S. R.
(
2009
).
Inter-annual variability in isotope and elemental ratios recorded in otoliths of an anadromous fish
.
J. Geochem. Explor.
102
,
181
-
186
.
Walther
,
B. D.
,
Kingsford
,
M. J.
,
O'Callaghan
,
M. D.
and
McCulloch
,
M. T.
(
2010
).
Interactive affects of ontogeny, food ration and temperature on elemental incorporation in otoliths of a coral reef fish
.
Environ. Biol. Fishes
89
,
441
-
451
.
Walther
,
B. D.
,
Dempster
,
T.
,
Letnic
,
M.
and
McCulloch
,
M. T.
(
2011
).
Movements of diadromous fish in large unregulated tropical rivers inferred from geochemical tracers
.
PLoS One
6
,
e18351
.
Warburton
,
M. L.
,
Reid
,
M. R.
,
Stirling
,
C. H.
and
Closs
,
G. P.
(
2017
).
Validation of depth-profiling LA-ICP-MS in otolith applications
.
Can. J. Fish. Aquat. Sci.
74
,
572
-
581
.
Weidel
,
B. C.
,
Ushikub
,
T.
,
Carpenter
,
S. R.
,
Kita
,
N. T.
,
Cole
,
J. J.
,
Kitchell
,
J. F.
,
Pace
,
M. L.
and
Valley
,
J. W.
(
2007
).
Diary of a bluegill (Lepomis marcrochirus): daily d13C and d18O records in otoliths by ion microprobe
.
Can. J. Fish. Aquat. Sci.
64
,
1641
-
1645
.
Woodcock
,
S. H.
,
Munro
,
A. R.
,
Crook
,
D. A.
and
Gillanders
,
B. M.
(
2012
).
Incorporation of magnesium into fish otoliths: determining contribution from water and diet
.
Geochim. Cosmochim. Acta
94
,
12
-
21
.
Wright
,
P. J.
(
1991
).
The influence of metabolic rate on otolith increment width in Atlantic salmon parr, Salmo salar L
.
J. Fish Biol.
38
,
929
-
933
.
Wright
,
P. J.
,
Fallon-Cousins
,
P.
and
Armstrong
,
J. D.
(
2001
).
The relationship between otolith accretion and resting metabolic rate in juvenile Atlantic salmon during a change in temperature
.
J. Fish Biol.
59
,
657
-
666
.
Yamamoto
,
T.
,
Ueda
,
H.
and
Higashi
,
S.
(
1998
).
Correlation among dominance status, metabolic rate and otolith size in masu salmon
.
J. Fish Biol.
52
,
281
-
290
.
Yoshinaga
,
J.
,
Nakama
,
A.
,
Morita
,
M.
and
Edmonds
,
J. S.
(
2000
).
Fish otolith reference material for quality assurance of chemical analyses
.
Marine Chemistr
69
,
91
-
97
.
Zitek
,
A.
,
Sturm
,
M.
,
Waidbacher
,
H.
and
Prohaska
,
T.
(
2010
).
Discrimination of wild and hatchery trout by natural chronological patterns of elements and isotopes in otoliths using LA-ICP-MS
.
Fish. Manag. Ecol.
17
,
435
-
445
.

Competing interests

The authors declare no competing or financial interests.

Supplementary information