Mass marking farmed Atlantic salmon with transgenerational isotopic fingerprints to trace farm fish escapees

Farmed fish sometimes escape and enter natural environments, where they mix with wild fish populations and can have negative effects. Marking farmed fish is a prerequisite for the identification of the origin of escapees and for guiding technical investigations to determine the cause of an escape event and improve farming practices. We tested transgenerational marking with enriched stable isotopes to assess its effectiveness as an accurate, feasible and cost-effective marking method for Atlantic salmon Salmo salar grown in sea-cage aquaculture. We injected a combination of 7 stable isotopes (134Ba, 135Ba, 136Ba, 137Ba, 86Sr, 87Sr and 26Mg) at 4 different concentrations (2, 0.2, 0.02 and 0.002 μg g−1 broodfish) into the abdominal cavity of female Atlantic salmon broodstock. Marking success was assessed in the otoliths of the resulting yolk sac larvae using laser ablation inductively coupled plasma mass spectrometry. Marking was 100% successful with Ba isotopes at concentrations as low as 0.002 μg and for Sr isotopes at 2 μg, when at least 3 wk had passed between the day of injection and spawning. Our results demonstrate that 63 unique fingerprint marks can be made at a low cost using enriched isotopes of Ba (US$0.0002−0.002 mark−1) and Sr (US$0.05−0.13 mark−1). Compared to other mass marking techniques, transgenerational marking of farmed salmon is an economically feasible method for tracing escapees with similarly low costs to delivery by egg bathing or vaccines, and an order of magnitude or more lower than other conventional marking methods.


INTRODUCTION
The rise of modern industrial aquaculture has introduced millions of selectively bred fish into environments where they are co-located with wild conspecifics.When they escape from aquaculture facilities, farmed fish can cause damaging ecological impacts when mixing with wild fish (Fleming et al. 2000, McGinnity et al. 2003, Hindar et al. 2006, Hutchings & Fraser 2008, Toledo-Guedes et al. 2012, Glover et al. 2013).Efforts to reduce escape events first requires detection of where the escape event occurred, so that subsequent engineering investigations can determine the cause of the escape event and make recommendations to improve the technical standards of containment systems (Jensen et al. 2010).
Atlantic salmon Salmo salar escape from sea-cage aquaculture farms in every country in which they are produced.Detecting escaped farmed salmon once they become mixed within wild populations and tracing escapees back to their farm of origin remains problematic.Although the point of escape can often be determined through DNA-based methods (Glover 2010), a fail-safe identification technique is still lacking.As an alternative to DNA-based approaches, a permanent tag or coded mark applied to all farmed fish would enable effective tracing.However, current mass marking methods, for example, fluorescent markers (Mohler 2003, Taylor et al. 2005), fin clipping and physical tags (Vander Haegen et al. 2005) or visible implant tags (FitzGerald et al. 2004), are unsuccessful with 1 or more aspects related to the ability to deliver 100% traceability to point of origin, fish welfare considerations or cost-effectiveness at industry scales.
Recently, new methods have been developed that enable 100% traceability of farmed salmon, are cost effective and have no impact on fish welfare throughout the production cycle.These methods involve the use of stable isotopes to code the otoliths of fish with unique isotope fingerprint marks during the hatchery stages of production (e.g.de-Braux et al. 2014, Warren-Myers et al. 2014, 2015a,b).To date, otolith marking with enriched stable isotopes of Ba and Sr has been highly successful in many species, and marks have been created using a range of delivery techniques, for example, via injection (Thorrold et al. 2006, Williamson et al. 2009b, Warren-Myers et al. 2014, 2015a), immersion (Munro et al. 2008, Woodcock et al. 2011b, de Braux et al. 2014, Warren-Myers et al. 2015b) or food supplementation (Woodcock et al. 2013).
The potential for identifying the origin of escaped farmed salmon with stable isotope marking is clear; between 7 and 63 mark combinations were created when Atlantic salmon parr were successfully marked with a combination of 6 isotopes mixed with a vaccine and delivered via injection (Warren-Myers et al. 2015a), and salmon embryos were marked with a combination of 3 isotopes during their egg-swelling phase immediately after fertilisation (Warren-Myers et al. 2015b).Another method with the potential to create additional multiple mark combinations with enriched stable isotopes in farmed salmon is transgenerational marking (Thorrold et al. 2006, Almany et al. 2007).This technique, which can successfully mark both freshwater (Munro et al. 2009, Starrs et al. 2014b) and marine fish species (Thorrold et al. 2006, Williamson et al. 2009b), requires an injection of enriched stable isotope into the abdominal cavity of mature females prior to spawning, which is then passed on in situ to the offspring.Marks are detectable in the core of otoliths of the resulting larvae (Thorrold et al. 2006).Many studies claim transgen-erational marking to be a successful technique for field applications to assess population connectivity (Thorrold et al. 2006, Williamson et al. 2009b, Huelga-Suarez et al. 2012), yet only 1 study (Almany et al. 2007) has demonstrated that transgenerational mar king is feasible for mass marking tens to hundreds of females.In fish farming, transgenerational marking would allow all eggs of a single broodfish to be marked with a single injection several weeks prior to stripping and fertilisation.For hatcheries, this means that no extra labour or protocol steps would be required to mark fish from the day of stripping onwards.Marking prior to stripping may also be an advantage over marking during the egg-swelling (Warren-Myers et al. 2015b), larval (de Braux et al. 2014) or parr stages (Warren-Myers et al. 2015a), as it would ensure that all fish are marked prior to any movement of eggs or fish within or between hatcheries.
Past studies on transgenerational marking have shown that timing between spawning and injection and the concentration required for 100% marking success varies greatly among species.For example, concentrations of 0.5 to 23 µg g −1 female have been successful in saltwater species (Thorrold et al. 2006, Williamson et al. 2009b) and 0.3 to 40 µg g −1 female in freshwater species (Munro et al. 2009, Huelga-Suarez et al. 2013), with spawning occurring anywhere between 1 and 170 d post injection in freshwater species (Munro et al. 2009, Starrs et al. 2014b) and 2 to 108 d in saltwater species (Cuif et al. 2014).Hence, the time between spawning and injection and the concentration required to achieve 100% marking success in farmed salmon requires optimisation to assess whether the technique will be suitable for large-scale application in aquaculture.
Here, we investigated whether transgenerational marking with enriched stable isotopes is a viable option for mass marking farmed Atlantic salmon by testing transgenerational marking on Atlantic salmon broodstock females using 7 enriched stable isotopes ( 134 Ba, 135 Ba, 136 Ba, 137 Ba, 86 Sr, 87 Sr and 26 Mg) at 4 concentrations (2, 0.2, 0.02, 0.002 µg g broodfish −1 ).We assessed marking success, mark strength and mark intensity in the otoliths of the resulting offspring.In addition, growth and mortality of offspring were monitored from hatching through to harvest size to check for any potential long-term effects of transgenerational marking with enriched stable isotopes.Finally, we provide cost estimates for the amount of isotope required to produce all successful fingerprint combinations.

Experimental design
The experiment was conducted at the Institute of Marine Research field station, at Matre, in Masfjorden, western Norway (60°N) using Atlantic salmon broodfish (AquaGen strain) that had been transferred from sea-cages to onshore freshwater tanks buffered with saltwater to a salinity of 0.7 g NaCl l −1 2 mo prior to the experiment.We tested transgenerational marking by injecting mature Atlantic salmon females (mean ± SE mass: 9.15 ± 0.26 kg) in the abdominal cavity using a hypodermic syringe with a standard volume of 60 ml prior to spawning.Each injection contained a combination of the enriched stable isotopes 134   1).Females were checked once a week post injection for ripeness, and any females ready to spawn had their eggs stripped and a subsample of eggs fertilised with 2 ml of sperm from 2 males (1 ml each).
Fertilised egg batches were kept at a constant temperature of 6°C throughout the egg incubation period (81 d) and yolk sac larval stage (52 d).Immediately prior to first feeding (Day 133), a subsample of 10 yolk sac larvae from each female's egg batch was collected and euthanized by anaesthetic overdose for otolith analysis.Sagittal otoliths from the subsampled larvae were dissected and removed, cleaned of any adhering tissue, air dried and stored individually in plastic tubes for otolith analysis.All remaining larvae from each egg batch were transferred to separate first feeding tanks, with a subsample of 50 fish from each batch randomly selected at the pre-smolt stage to be grown on to 4 kg harvest size.

Otolith preparation
Sagittal otoliths were cleaned as per Warren-Myers et al. (2014).Briefly, any remaining organic tissue was removed by immersing otoliths in a solution of ultrapure 15% H 2 O 2 buffered with 0.1 M NaOH.Following immersion, otoliths were ultrasonicated (Sonic Clean 250HT) for 5 min and then left for 6 h in the cleaning solution.The cleaning solution was then aspirated off and the otoliths were transferred through 3 Milli-Q water rinses, each of which consisted of 5 min of ultra-sonification and 30 min resting time.Otoliths were then air dried in a laminar flow bench for at least 24 h.Once dry, 1 otolith fish −1 was fixed onto gridded microscope slides using quick dry cyanoacrylate glue.No polishing of otoliths was required prior to laser ablation.

Otolith analysis
Stable isotope analyses were done on a Varian 7700x inductively coupled plasma mass spectrometer (ICP-MS) fitted with a HelEx (Laurin Technic and the Australian National University) laser ablation (LA) system constructed around a Compex 110 (Lambda Physik) excimer laser operating at 193 nm.National Institute of Standards and Technology (NIST) 612 and 610 glass standards doped with trace elements at known concentrations were used to calibrate the system.Otoliths were run in blocks of 16 samples selected randomly from all treatments and bracketed by analyses of the standards.External precision estimates (% relative standard deviation) based on 20 analyses of NIST 612 and NIST 610 standards were as follows: NIST 612, 135 Ba: 138 Ba = 0.38, 137 Ba: 138 Ba = 0.26, 86 Sr: 88 Sr = 0.51, 87 Sr: 88 Sr = 1.04 and 26 Mg: 24 Mg = 2.85; NIST 610, 134 Ba: 138 Ba = 7.17 and 136 Ba: 138 Ba = 6.49.Samples and standards were analysed in timeresolved mode, using a spot size of 157 µm, a laser energy setting of ~60 mJ and a laser repetition rate of 10 Hz.Spot ablation was performed under pure He (200 ml min −1 ) to minimise re-deposition of ablated material and the sample was then entrained into the Ar (0.95 ml min −1 ) carrier gas flow to the ICP-MS.Using this method, we were able to quantify the isotope ratios for 134 Ba: 138 Ba, 135 Ba: 138 Ba, 136 Ba: 138 Ba, 137 Ba: 138 Ba, 86 Sr: 88 Sr, 87 Sr: 88 Sr, 24 Mg: 26 Mg and 55 Mn: 43 Ca, from the edge to the core of each salmon yolk sac larval otolith using a single spot ablation ( 55 Mn: 43 Ca was used to identify when the laser had hit the core; Barbee & Swearer 2007).Data were processed off-line using a specialised MS Excel template which in - volved a low pass filter to remove any spikes (a single acquisition value > 2× the median of the adjacent acquisitions), smoothing (a running average of 3 acquisitions) and blank subtracting functions.A correction factor (K = R true /R obs , where R true is the naturally occurring isotope ratio and R obs is the average isotope ratio measured in the NIST 612 or 610 standard run before and after each set of 16 samples) was applied to all sample acquisitions to correct for mass bias.

Statistical analysis
Marking success for each treatment (Fig. 1) was evaluated using a mark detection limit (Warren-Myers et al. 2014).Briefly, detection limits for the isotope ratios 134 Ba: 138 Ba, 135 Ba: 138 Ba, 136 Ba: 138 Ba, 137 Ba: 138 Ba, 86 Sr: 88 Sr, 87 Sr: 88 Sr and 26 Mg: 24 Mg were calculated from the average isotope ratios of all control fish (i.e.0 µg l −1 treatment).To ensure a correct classification probability of 99.94%, mark detection limits were set at 3.3 standard deviations (SDs) above the mean observed ratio in control fish for each enriched isotope used.Because of the inherent instability in isotopic ratios measured on single-detector, ICP-based mass spectrometers, we conservatively set the criteria for detecting a successful mark in the otolith as at least 3 consecutive acquisitions with ratios above the detection limit.
Mark strength and mark intensity for each enriched isotope used was analysed using 2-factor ANOVAs with isotope concentration and number of weeks between injection and spawning treated as fixed factors.An interaction term was not included, as 2 combinations of concentration by weeks post injection (Week 1, 0.002 µg and Week 2, 2 µg) had no females spawn and hence no data.The response variables used were the mean maximum isotope ratio value (mark strength) and the mean proportion of acquisitions between the otolith edge and otolith core with ratio values above the detection limit (mark intensity) measured from the otoliths of the 10 subsampled fish for each egg batch.The effect of treatment on total hatchery mortality per egg batch and the number of larval deformities observed at first feeding per egg batch were analysed with 1-way ANOVAs.The effect of treatment on length, weight, Fulton's condition factor (k) (Ricker 1975) and survival of harvest size fish was analysed with 1-way ANOVAs.

Marking success
Marking success was dependent on stable isotope enrichment concentration and the number of weeks between injection and spawning (Table 2).The highest concentration (2 µg g −1 fish) achieved 100% marking success in the shortest time period for the Ba (1 wk: 134 Ba, 135 Ba 136 Ba, 137 Ba) and Sr isotopes (3 wk: 86 Sr, 87 Sr), but only 30% for 26 Mg by Week 3. When the concentration was reduced (0.2 and 0.02 µg g −1 fish), 100% marking success for the Ba isotopes was achieved when spawning occurred at least 2 wk post injection for 135 Ba and 137 Ba, or at least 3 wk post injection for 134 Ba and 136 Ba.Marking success was poor for 86 Sr, The blue line represents a control fish, and the red line represents the mark detection limit, calculated as 3.3 standard deviations above the mean ratio measured in all control fish.A successful mark was classified as 3 consecutive acquisitions above the detection limit.Photo indicates the diameter of the laser spot size (157 µm) used when ablating from the otolith edge (surface) to the core Sr and 26 Mg (0 to 10%) at a concentration of 0.2 µg g −1 fish or less, regardless of the number of weeks between injection and spawning.Marking success of 75% and 80% was achieved for 135 Ba and 137 Ba, respectively, when spawning occurred 3 wk post injection at the lowest concentration (0.002 µg g −1 fish).

Mark strength: maximum acquisition ratios
Mark strength, assessed using the maximum acquisition ratios, showed that a concentration of 2 µg g −1 fish produced the highest maximum ratios (Fig. 2, Table 3) and that maximum isotope ratios increased as the period between injection date and spawning date lengthened (Fig. 3).

Mark intensity: % of acquisition counts above detection limit
Mark intensity, assessed by the proportion of an otolith marked with acquisition counts above the detection limit, showed that the higher concentrations marked a greater proportion of the otolith (Fig. 4, Table 3).In addition, the proportion of an otolith marked increased as the period be tween injection and spawning lengthened (Fig. 5).
The number of acquisition counts above the detection limit for 26 Mg was insufficient to justify conducting mark inten sity analysis on the proportion of otolith marked.

Broodstock health, hatchery mortality, larval deformities and condition at harvest
Of the 30 females injected, 3 fish were unsuccessfully spawned.These consisted of 1 fish that Strength is the number of times the maximum isotope ratio measured in a marked otolith is greater than the threshold limit.Intensity is the percentage of the otolith marked with an isotope ratio greater than the threshold limit.Shading indicates the minimum number of weeks required between injection date and spawning date to reach 100% marking success for each isotope (dark grey: 1 wk, medium grey: 2 wk; light grey: 3 wk; no shading: 100% marking success was not achieved) died 10 d after injection for reasons that were unknown, a second fish having overripe eggs due to being stripped too late, and the third individual not reaching spawning ripeness in the time frame of egg collection (within 6 wk post injection).All other females in the experiment produced viable eggs, although there was some variation in the degree of egg ripeness when spawned.Offspring mor tality between egg fertilisation and first feeding stage (mean ± SE) averaged 15.7 ± 3% per egg batch, and there was no treatment effect of isotope enrichment (F 4, 26 = 1.2, p = 0.4).Yolk sac larval deformities that we ob served between hatching and first feeding averaged 0.25 ± 0.07% per egg batch, with no treatment effect of isotope enrichment (F 4, 26 = 0.5, p = 0.7).Fish harvested at 2.25 yr post hatch (weight: 3.79 ± 0.02 kg, fork length: 62.9 ± 2.5 cm, condition factor k: 1.39 ± 0.06) showed no significant difference in length, weight or condition among treatments (weight: F 4, 26 = 0.88, p = 0.5; fork length: F 4, 26 = 0.81, p = 0.5; k: F 4, 26 = 1.59, p = 0.2).Mortality per treatment during the sea cage stage averaged 8 ± 0.5%, with no significant difference among treatments (F 4, 26 = 1.79, p = 0.9).

DISCUSSION
We have demonstrated that producing unique isotopic fingerprint marks in the otoliths of Atlantic salmon larvae via trans generational marking is highly successful with Ba-and Sr-enriched stable isotopes.This means it is possible to mass mark farmed Atlantic salmon at the earliest possible point in the life cycle, prior to spawning.Ensuring 100% marking success is dependent on the concentration of enriched isotope used and the length of time between injection date and spawning date.

Marking success
A 6-marker fingerprint with 100% marking success was achieved using a combination of 4 Ba-and 2 Srenriched stable isotopes in the 2 µg g −1 broodfish treatment when injection date and spawning date were at least 3 wk apart.This is the first reported successful 6-mark isotope combination using the transgenerational marking technique.Only 1 other study has successfully marked fish with a 6-isotope combination (Warren-Myers et al. 2015a), but marks were delivered by an injection of stable isotopes directly into salmon parr, not via broodstock.We achieved 100% marking success for concentrations lower than 2 µg g −1 female when using the 4 Ba isotopes, but not the 2 Sr isotopes, with all Ba isotopes achieving 100% marking success at 0.02 µg g −1 female when injection date and spawning date were at least 3 wk apart.Ba concentrations as low as 0.5 µg g −1 female have been successful in saltwater species (Thorrold et al. 2006, Williamson et al. 2009b)

Mark strength and intensity
135 Ba produced the strongest (maximum ratio) and most intense (proportion of an otolith marked) tags (Table 3).On average, 80% of acquisitions in the otolith were marked with 135 Ba at the lowest successful concentration (0.02 µg g −1 female) with a maximum value 4.6 times higher than the threshold limit.137 Ba produced marks of similar strength and intensity with 71% of acquisitions marked and an average maximum value 2.8 times above the threshold limit.137 Ba is the most commonly used Ba isotope for marking fish otoliths (Thorrold et al. 2006, Munro et al. 2009, Cuif et al. 2014), and 135 Ba less so (Almany et al. 2007, Williamson et al. 2009b), yet our results suggest that 135 Ba has the potential to produce slightly stronger marks than 137 Ba, potentially due to differences in purity of the 2 enriched isotopes used ( 137 Ba: 81.7% vs. 135 Ba: 93.4%; Oak Ridge National Laboratory; www.ornl.gov).
134 Ba and 136 Ba mark strength and intensity were ~50% lower compared to 135 Ba and 137 Ba in the 0.02 µg g −1 concentration (Table 3), likely due to the higher detection limits for these isotopes resulting from isobaric interference from Xe in the carrier gases.On average, 34 and 35% of acquisitions in the otoliths were marked with 134 Ba and 136 Ba, respectively, with maximum values 2 and 2.3 times higher than the threshold limits.Al though strength and intensity were ~50% lower, marks created with 134 Ba and 136 Ba were clearly definable at a concentration of 0.02 µg g −1 female when the timing between in jection and spawning surpassed 3 wk and therefore should be highly useful for creating fingerprint combinations using 1, 2, 3 or 4 Ba isotopes.Prior to this study, neither of these isotopes had been tested or demonstrated to be 100% successful in marking otoliths using transgenerational marking.However, Warren-Myers et al. (2015a) successfully used 136 Ba and 134 Ba mixed with a vaccine and delivered via in jection in Atlantic salmon parr and produced slightly higher mark strength values (3.6 times the relative threshold limit for both).In addition, Woodcock et al. (2011a) achieved 93% marking success with 136 Ba in golden perch Macquaria ambi gua using a larval immersion technique, but reported neither mark strength nor intensity.

86
Sr and 87 Sr produced well defined marks in the otoliths of offspring that came from brood stock females injected with a concentration of 2 µg g −1 female and spawned 3 wk post injection.Mark strength maximum values were 1.1 times higher than the threshold limit for both 86 Sr and 87 Sr and 37 to 43% of acquisitions in the oto liths were marked with 86 Sr and 87 Sr, respectively.Marking success with 86 Sr and 87 Sr at a concentration of 2 µg g −1 female has not been demonstrated prior to this study using LA-ICP-MS detection methods.However, 100% success has been achieved with 87 Sr at a concentration of 20 µg g −1 female (Starrs et al. 2014b).Relative to the concentration of 0.02 µg g −1 female of all 4 Ba isotopes required to inject 10 kg Atlantic salmon broodstock to ensure successful marking of offspring, 2 µg g −1 female for Sr isotopes is high.Sr isotopes are therefore less financially feasible for mass marking programs.Sr isotopes may be more suitable if applied to smaller sized species (e.g.eastern rainbowfish Melanotaenia splendida; Starrs et al. 2014b) or by alternate delivery methods, such as immersion (Munro et al. 2008, Smith & Whitledge 2011, de-Braux et al. 2014).

Broodstock health, hatchery mortality, larval deformities and condition at harvest
Injecting broodstock with enriched stable isotopes had no effect on spawning success or broodstock sur-vival to spawning.However, when eggs were stripped, some internal bleeding in the abdominal cavity had occurred around the injection site in some females in both treatment and control fish.Smaller injection volumes may help prevent this from occurring, and should be tested in the future, particularly as the process of injection has been reported to kill broodfish in other species (e.g.Starrs et al. 2014b).Offspring of all successfully spawned females (27 of 30) showed no effect of isotope marking on egg survival or larval deformity rates, which is consistent with other studies that have marked with stable isotopes at concentrations equivalent to 2 µg g −1 female or less (Thorrold et al. 2006, Cuif et al. 2014, Warren-Myers et al. 2015b).No effect of marking on harvest size fish was found, which is consistent with results observed in fish that have been vaccinated with stable isotopes and grown to 5 kg (Warren-Myers et al. 2015a).Based on our results and previous research (Williamson et al. 2009a), transgenerational marking with stable isotopes of Ba and Sr is a safe, effective method for mass marking farmed fish.

Transgenerational marking as a mass marking tool
Mass marking millions of fish can be an expensive exercise, hence quick, accurate and cost-effective techniques that instantly batch mark numerous fish are preferred.Here, we have shown that transgenerational marking with enriched stable isotopes is another useful tool for mass marking salmon offspring prior to spawning in commercial hatcheries with 63 unique codes possible (Table 4).Marks using Ba isotopes are cheaper to apply (US$ 0.0002−0.002fish −1 ) compared to Sr isotopes (US$ 0.05−0.13fish −1 ), but Sr isotopes may still be useful if used on small numbers of broodfish.The LA-ICP-MS cost to detect marks using spot ablation was approximately US$ 15−20 otolith −1 .For the identification of escaped farmed fish, this may be considered expensive relative to the marking costs.However, the cost to analyse hundreds of fish at US$ 20 otolith −1 is small relative to the cost to apply marks to hundreds of millions of farmed fish.Whether further otolith preparation (e.g.sectioning) is required to accurately detect the marks in adult salmon otoliths must be confirmed to fully validate the method as a marker approach for application at an industry scale.
To date, transgenerational marking with enriched stable isotopes has been validated in 13 species (Table 5), including freshwater, diadromous and marine fish.Both Sr-and Ba-enriched isotopes work well 84 <$ 0.002 fish −1 <$ 0.05 fish −1 <$ 0.09 fish −1 <$ 0.13 fish Analysis of all transgenerational marking studies with enriched stable isotopes conducted to date reveals that Ba isotopes have been the most successful across all fish species tested (Table 5).For Atlantic salmon, this is also the case for different delivery methods that have tested isotope marking across a range of life history stages, for example; bathing of freshly fertilised eggs (Warren-Myers et al. 2015b), immersion of yolk sac larvae (de-Braux et al. 2014) or injection of parr (Warren-Myers et al. 2015a).In Atlantic salmon, isotopes of Ba produce strong, easily identifiable marks at concentrations 100 times lower than Sr isotopes and therefore are the most suitable and cost-effective isotopes to use for mass marking farmed fish.Transgenerational marking with Ba isotopes is another successful method to effectively mass mark fish that pinpoints the prespawning stage in the production life cycle.

Fig. 1 .
Fig. 1.Comparison of the change in 137 Ba: 138 Ba between the otolith edge (acquisition no.= 0) and core (acquisition no.150−200) in marked yolk sac larval Atlantic salmon Salmo salar.Each black line represents an otolith marked with a different concentration of 137 Ba: a = 2, b = 0.2, c = 0.02 and d = 0.002 µg l −1 .The blue line represents a control fish, and the red line represents the mark detection limit, calculated as 3.3 standard deviations above the mean ratio measured in all control fish.A successful mark was classified as 3 consecutive acquisitions above the detection limit.Photo indicates the diameter of the laser spot size (157 µm) used when ablating from the otolith edge (surface) to the core

Fig. 2 .
Fig. 2. Maximum observed isotope ratios in the otoliths of Atlantic salmon Salmo salar yolk sac larvae marked via transgenerational marking.Concentrations were 2, 0.2, 0.02, 0.002 and 0 (control) µg of each isotope per g broodfish weight.Error bars represent ± 1 SE.Different letters above bars indicate difference among concentrations based on pairwise comparisons (p < 0.05).Dashed line shows natural expected ratio for each isotope

Fig. 5 .
Fig. 5. Percentage of an otolith marked with enriched barium isotopes between the edge and the core of Atlantic salmon Salmo salar yolk sac larval otoliths marked via transgenerational marking categorized by number of weeks between injection date and spawning date.Error bars represent ± 1 SE.Different letters above bars indicate differences among concentrations based on pairwise comparisons (p < 0.05) only claimed in broodstock Table5.Fish species validated to have been marked via the transgenerational marking method with enriched Ba and Sr stable isotopes.Concentrations (conc.) and spawning times reflect the required minimums (min.) to achieve 100% marking success.LA: laser ablation, ICP: inductively coupled plasma, MS: mass spectrometry, BaCl, 135BaCl, 136 BaCl, 137 BaCl,

Table 2 .
Ba, 136Ba,137Ba, Percentage of Atlantic salmon Salmo salar yolk sac larval otoliths marked using a combination of 7 enriched stable isotopes delivered via transgenerational marking.Marking success rates of 100% are highlighted in bold

Table 3 .
Comparison of isotope mark strength and intensity.

Table 4 .
Estimated cost per stable isotope marker code (n = 63) for transgenerational marking of Atlantic salmon Salmo salar.Costs (in US $) are calculated based on a 10 kg broodfish producing at least 5000 viable offspring, injected with the minimum required amount of enriched isotope to achieve 100% marking success.Isotope pricing is based on the cost at the time isotopes were purchased from