Application of otolith (ear stone) geochemistry to develop natural signatures has been a common practice among fishery scientists to distinguish fish from separate nursery habitats (Gillanders and Kingsford, 2000; Zapp Sluis et al., 2012; Rooker et al., 2019), examine natal homing and movement of adult stocks (Thorrold et al., 2001; Fraile et al., 2015; Rooker et al., 2014, 2021), assess population connectivity and stock mixing (Reis-Santos et al., 2013; Sanchez et al., 2020; Rooker et al., 2023), investigate environmental stressor exposure (Limburg and Casini, 2019), and applied as an alternative aging method (Heimbrand et al., 2020; Hüssy et al., 2021; Frey et al., 2023). Otoliths are acellular, metabolically inert and accrete as the fish grows, allowing some elements from surrounding seawater to be incorporated within the otolith matrix and permanently retained throughout life (as reviewed in Campana, 1999). This property led to the development of various analytical techniques to quantify elemental concentrations in otoliths, with inductively coupled plasma mass spectrometry (ICP-MS) currently one of the most widespread applications.

Initially, solution-based ICP-MS was more commonly applied over laser ablation ICP-MS, because it was perceived to be more precise and able to obtain better quantities of trace elements (de Pontual et al., 2000). However, solution-based analysis often relies on whole otoliths being dissolved to incorporate the entire life history of an individual fish. Thus, this approach does not provide age or life-stage specific geochemical information. While micromilling devices can be used to extract otolith cores or material from specific life-history periods for solution-based assays (Arslan and Secor, 2008; Rooker et al., 2010), the size of the drilling bit limits the precision of the sample material collected. With laser ablation ICP-MS, discrete sampling locations on the otolith can be linked to a specific age or life stage of an individual through otolith microstructure analysis, providing the time of occurrence for each geochemical value (Mohan et al., 2022).

Research applications using laser ablation ICP-MS for geochemical analysis of otoliths typically include line transects or a series of spots. Line transects are a continuous ablation path that run parallel or sub-parallel to the radial growth direction of the otolith from the central core (start of life) to the exterior edge (end of life), which are commonly used to generate geochemical life profiles (Serre et al., 2018; Rooker et al., 2021; Mohan et al., 2022). In contrast, spots ablate specific locations on the otolith and are often used to investigate a certain period of life. A series of evenly spaced spots from the core to the margin of the otolith have also been used to examine changes in elemental concentrations throughout the individual’s life (Baumann et al., 2015; Catalán et al., 2018; Sanchez et al., 2020).

Three teleost fish taxa from different ecosystems with distinct physiochemical conditions were selected to evaluate different ICP-MS applications frequently used in the geochemical analysis of otoliths and other biogenic carbonate structures. The primary aim of the study was to determine whether otolith element:Ca (E:Ca) values and resulting geochemical signatures were statistically similar between two common ICP-MS applications (laser ablation and solution-based), and between two widely used laser ablation techniques (continuous line transects versus series of spots). Further, if any variations in otolith E:Ca values occurred between these applications, were similarities or differences consistent among the three species or variety of ecosystems (estuarine, coastal/reef, pelagic) assessed in this study.

Sagittal otolith pairs were obtained from three species collected in different geographic locations. The finfish taxa used in this assessment were representative of individuals inhabiting estuarine (red drum [Sciaenops ocellatus]), coastal/reef (gray parrotfish [Sparisoma axillare]), and pelagic (blackfin tuna [Thunnus atlanticus]) ecosystems. Red drum were collected along the coast of Texas, USA from either Galveston Bay or the Upper Laguna Madre. Gray parrotfish were collected from reef or seagrass habitats along the southern coast of Pernambuco, Brazil. Blackfin tuna were collected from the Caribbean Sea (Saint Lucia) and the northern Gulf of Mexico. Otoliths from 10 individuals from each region or habitat type were used in the analyses, resulting in 20 samples per species. All specimens were juveniles, with respective restricted size ranges to minimize potential age/size effects for comparisons between the two sampling locations for each species.

Otoliths were rinsed with 18 megohm doubly deionized water (DDIH₂O) and a soft bristled plastic brush was used to remove any adhering tissue residue. To further reduce any surface contamination, otoliths were then soaked in 1% ultra-pure nitric acid (HNO₃) for 1 minute, then liberally rinsed with DDIH₂O to remove the acid. After drying under a class 100 laminar flow hood, otoliths were stored in acid washed plastic vials until prepared for elemental analysis.

Accuracy of the laser ablation techniques to quantify reference standard values were high, with recovery rates ranging from 94.6-102.2% for line transects and 91.3-102.5% for spots ( Table 1 ). There was no significant difference in the accuracy of laser ablation techniques for all elements, except Mg which had significantly higher (F_1,95 = 74.35, P<0.0001) recovery rates for lines (99.2%) compared to spots (91.3%). For solution-based analysis, there was a slight overestimation of the recovery rate for Mg (125.3%) in the FEBS-1 standard, while Ca (86.6%) and Sr (86.1%) were underestimated with MACS-3 ( Table 1 ). The accuracy of the solution-based analysis with MACS-3 was significantly lower for Ca (F_1,48 = 418.81, P<0.0001) and Sr (F_1,48 = 161.16, P<0.0001) compared to line transects, with no other significant differences occurring among the other elements. Recovery rates of the FEBS-1 standard for solution-based analysis were significantly higher, although sometimes slightly overestimated, for Mg (F_1,48 = 117.19, P<0.0001), Mn (F_1,48 = 80.36, P<0.0001), and Ba (F_1,48 = 31.30, P<0.0001), while significantly lower for Ca (F_1,48 = 35.27, P<0.0001) compared to line transects. Relative standard deviations (RSDs) for all three analytical applications were below 6.5% with improved precision calculated for solution-based analysis using the FEBS-1 standard (0.9-2.1%: Table 1 ). The precision for line transects (1.2-6.3%) and spots (1.5-4.0%) was similar for all elements, except for Ba were the precision for spots (2.9%) was improved compared to line transects (6.3%). The MACS-3 standard for solution-based analysis tended to be less precise, typically producing higher RSD values compared to the other applications (4.4-5.4%; Table 1 ).

Mean otolith E:Ca profiles from line transects for all three species aligned well with mean E:Ca values of spots ablated along the same region of the otolith, with values for spots consistently within the shaded area (1 SD) of the running means for combined E:Ca values of lines ( Figure 2 ). Mean E:Ca values of spots were typically closer to the line transect means at the core versus the edge of the otolith. For red drum, mean otolith E:Ca values were only significantly different for Mg (F_2,57 = 24.50, P<0.0001) and Sr (F_2,57 = 24.68, P<0.0001) among the three applications. For both of these elements, line transects and spots had similar otolith E:Ca values, and both laser ablation techniques had significantly higher values than those from the solution-based analysis ( Figure 3 ; Table 2 ). Mean otolith E:Ca values for gray parrotfish were significantly different among the applications for Sr (F_2,48 = 10.68, P<0.001) and Ba (F_2,48 = 4.21, P<0.05) only. Otolith Sr:Ca values did not differ significantly between line transects and spots, and both were significantly higher than solution-based values. For Ba:Ca, values from the line transects did not differ significantly from spots or solution-based analysis; however, Ba:Ca values for spots were significantly lower than solution-based values ( Figure 3 ; Table 2 ). Mean otolith E:Ca values for blackfin tuna were significantly different among the three applications (Mg: F_2,57 = 6.06, P<0.01; Mn: F_2,57 = 21.94, P<0.0001; Sr: F_2,57 = 3.55, P<0.05; Ba: F_2,57 = 44.21, P<0.0001). When examining these differences further, line transects and spots had similar mean values, with both laser ablation techniques showing higher Mg:Ca and Mn:Ca values, but lower Ba:Ca values relative to the solution-based analysis. For Sr:Ca, values from the spots did not differ significantly from line transects or the solution-based analysis; however, Sr:Ca values for lines were significantly lower than solution-based values ( Figure 3 ; Table 2 ).

Linear regression models show that line transect and spot techniques produced nearly identical overall mean otolith E:Ca values for all elements assayed (R² ≥ 0.91), with slopes that deviated minimally from the 1:1 line for all species combined ( Figure 4 ). This was expected given that no differences were detected between line transects and spots for any of the three species. The strongest relationship between lines and spots based on R² values occurred for Mn:Ca and Sr:Ca (0.98 and 0.96, respectively). While the differences were not significant, line transect Sr:Ca values did appear to be slightly lower compared to spots for all species, which was also observed in the overall means ( Table 2 ). Gray parrotfish otoliths had slightly higher Mg:Ca values and lower Ba:Ca values for spots relative to line transects, which resulted in slightly lower R² values (0.94 and 0.91, respectively) for these elements ( Figure 4 ; Table 2 ).

Since a strong 1:1 relationship occurred for otolith E:Ca values between line transects and spots, only one laser ablation technique (line transects) was selected to further evaluate differences between laser and solution-based ICP-MS. Although estimates of mean otolith E:Ca values between line transect and solution-based applications were not as highly correlated as between both laser ablation techniques, robust relationships did exist (R² = 0.64 - 0.87; Figure 5 ). Again, the strongest relationship occurred for Mn:Ca (R² = 0.87), with generally higher values for line transects compared to the solution-based analysis (as also seen in Table 2 ). The weakest relationship occurred for Sr:Ca values (R² = 0.64), where otolith E:Ca values for line transects were significantly higher for red drum and gray parrotfish, and significantly lower for blackfin tuna compared to solution-based E:Ca values ( Figures 3 , 5 ).

As advances in otolith geochemistry continue to develop, it is important to examine various analytical approaches as different methodologies or techniques can impact results (Campana et al., 1997; Secor et al., 2002). While previous studies have investigated the differences of using laser ablation and solution-based ICP-MS for analyzing otoliths of fish larvae (Ludsin et al., 2006) or inter-laboratory discrepancies between milled solution-based samples versus laser ablation (Geffen et al., 2013), this study is novel in its approach by not only contrasting laser ablation and solution-based applications to quantify otolith E:Ca values for several species that occupy differing habitats or ecosystems, but also by comparing two common laser ablation techniques (transect lines vs spots). For all species investigated, otolith E:Ca values for lines and spots were comparable, with no statistical differences in overall means detected between the two laser ablation techniques. While line transects and spots produced statistically similar results, suggesting data from each may be used interchangeably, it should be noted that line transects provided more detail along the region of the otolith profiled, whereas fluctuations in otolith E:Ca values may be missed between spots depending on the spacing, as seen in Figure 2 . Yang et al. (2014) performed laser ablation line transects and spots on ostracod shells and determined that line transects provided more precise E:Ca values over spots due to a higher sample mass ablated per scan. Based on the RSD percentages of this study, the precision between line transects and spots was relatively similar, with the highest variation occurring for Ba. Further, no significant difference existed in the accuracy of the two laser ablation techniques, although line transects produced more accurate measurements of Mg compared to spots.

With unlimited volumes, longer integration times, and multiple analysis cycles, solution-based ICP-MS should have better precision and sensitivity compared to laser ablation. Laser ablation integration times are usually less than 0.05-0.1 s, whereas solution mode integration times typically range from 0.1-0.3 s allowing for better counting statistics and resolution of more representative values. Repeated analyses (typically three cycles) in solution mode of the exact same sample drives down the precision (RSD) compared to laser ablation ICP-MS. While the RSD percentages of this study were relatively good (<10%) for all application trials, solution-based analysis was more precise, producing the lowest RSDs, when using the FEBS-1 reference standard; however, the MACS-3 standard tended to be less precise than laser ablation techniques. Further, while both laser ablation techniques produced good recovery rates (90-110%) representing high accuracy, solution-based analysis using FEBS-1 slightly overestimated Mg and Mn, while MACS-3 underestimated Ca and Sr. Other studies have exposed accuracy problems when using MACS-3 for exclusive calibration compared to glass reference material, likely due to the heterogeneity of the material, and determined using NIST 612 to calibrate MACS-3 provided element concentrations within 10% of the certified reference values (Thompson et al., 2022). This method was applied to the laser ablation techniques to help improve accuracy and precision, but it was not an option with solution analysis resulting in lower precision and accuracy when using the MACS-3 standard. Geffen et al. (2013) showed similar results overall with solution-based analysis being more precise, but laser ablation ICP-MS being more accurate when examining the reference standards NIES-22 and FEBS-1. Even though solution-based analysis can resolve more elements, the required volume for the analysis can create high dilution rates that can make elements difficult to detect (Campana et al., 1997). For this reason, several elements in this study were removed from the statistical analysis due to concentrations below detection limits, with some of the smallest otoliths (gray parrotfish) being removed completely as the sample dilution resulted in a volume too small to be practically analyzed with the autosampler. Conversely, there were no issues analyzing the smallest otoliths in this study with laser ablation ICP-MS.

Among the three applications investigated, otolith Mn:Ca values consistently had the highest correlation for all paired comparisons, followed by otolith Mg:Ca values. The MACS-3 reference standard Mn concentrations are more accurate compared to the other elements (Thompson et al., 2022), potentially explaining the increased correlation between laser ablation and solution-based applications. Otolith cores also tend to be enriched in Mn due to physiological regulation during the embryonic stage as it is necessary for respiratory development, and Mn may be an important constituent in the initial development of the otolith as a substitute for Ca (Brophy et al., 2004; Ruttenberg et al., 2005). While not as pronounced as Mn, Mg values can also be elevated at the core or during the juvenile stage (Ruttenberg et al., 2005). It is important to note that allometric growth of the otolith during ontogeny may lead to variation in mean E:Ca values between laser ablation and solution-based applications on the same otolith. This is due to laser ablation techniques quantify otolith material along a single transverse plane through the otolith core while solution-based ICP-MS sums the geochemistry of all growth planes within the 3-D structure of the whole otolith. Both laser ablation techniques for red drum and blackfin tuna had significantly higher Mg:Ca values relative to solution-based analysis, whereas both laser ablation techniques produced significantly higher Mn:Ca values for blackfin tuna only. These two species consisted of older, larger juveniles (> age-1) compared to gray parrotfish (age-0) in the study samples. Therefore, a laser ablation path along a transverse section strongly influenced by core chemistry may have acquired higher concentrations in these larger individuals that could have been slightly diluted in the solution-based samples due to whole otolith dilution. The smaller parrotfish otoliths may have more uniform Mn:Ca and Mg:Ca values throughout the otolith that did not result in significant differences between laser ablation and solution-based ICP-MS.

Although Sr is highly concentrated in otoliths, associated Sr:Ca values were least correlated between laser ablation and solution-based applications, with Ba:Ca values exhibiting better correlation. In otoliths, Sr and Ba concentrations reflect ambient water conditions and are biomineralized in proportion to the surrounding environment, with higher Sr:Ca values occurring in seawater compared to freshwater and higher Ba:Ca values occurring in freshwater (Bath et al., 2000). For red drum and gray parrotfish, otolith Sr:Ca values from laser ablation techniques were higher compared to solution-based analysis, potentially due to the factors discussed above. However, line transects on otoliths of blackfin tuna had significantly lower Sr:Ca values than the solution-based analysis, with overall Sr:Ca values of blackfin tuna being generally lower than the other two species analyzed in this study. Likewise, Ba:Ca values were lower for spots of gray parrotfish and for both laser ablation techniques of blackfin tuna compared to the solution-based analysis, whereas there was no significant difference in Ba:Ca values between ICP-MS applications for red drum. Both gray parrotfish and blackfin tuna profiles start with low Ba:Ca values at the otolith core gradually increasing near the edge, while red drum profiles remain fairly constant throughout. The overall low Sr:Ca values for the blackfin tuna and the initial lower Ba:Ca values for gray parrotfish and blackfin tuna may have had the opposite effect previously discussed in the paragraph above, where now whole otolith dilution of the solution-based analysis was able to acquire greater Sr:Ca and Ba:Ca values compared to the small spatial scale of the laser ablation techniques. Otolith accretion and metabolic rate differences among the three species can result in specie specific variations in the incorporation of elements into the otolith (Hammer and Jenkins, 2007), further explaining why E:Ca values derived from laser ablation and solution-based applications were not uniform across species for certain elements.

This study demonstrates that otolith E:Ca values were statistically similar between the two laser ablation techniques for each of the three species assayed, suggesting that line transects and spots can be used interchangeably. Both laser ablation techniques can generate life history profiles that provide a tool to characterize otolith E:Ca fluctuations over time or age of the individual. Line transects provide more fine-scale and continuous resolution of otolith E:Ca variation along a life history transect that may otherwise be missed between the spacing of spots. It was also demonstrated that E:Ca values from laser ablation and solution-based applications were often statistically different, but discrepancies in E:Ca values can vary among fish species. Finally, it was shown that solution-based assays incorporating whole otolith dilution may be more comparable to laser ablation when using otoliths of smaller or younger individuals, with differences becoming more pronounced when assaying otoliths of larger, older individuals. This is attributed to otoliths of younger individuals typically beginning with a more uniform (oval) shape before differential growth along the otolith axes becomes more pronounced with age.