Evaluating food metrics of lanternfishes in waters of the southeastern Pacific Ocean

Using carbon and nitrogen stable isotope values, we analyzed the trophic position (TP) and the isotopic niche width of lanternfishes from three different areas in the Southern Pacific Ocean. Fishes from Perú had slightly higher δ13C values compared with fish from Chilean areas. In contrast, δ15N values increased with latitude (North to South). Myctophids TP differed between the three study areas (highest in Central Chile, lowest in Peru). Peruvian fishes had a smaller isotopic niche than the lanternfishes of those from Chile.


Introduction
Mesopelagic habitats are vast ecosystems, with an estimated biomass of approximately 10 billion metric tonnes (Goetsch et al. 2018). One of the biological components contributing to this high biomass is mesopelagic fish, which inhabit depths between 200 and 1000 m (Moteki et al. 2017;Christiansen et al. 2018;Belcher et al. 2019). The family Myctophidae is by far the most significant fraction of the mesopelagic fish component (Espinoza et al. 2017). For example, in the Southern Pacific Ocean (SPO), myctophids (comprising 35 species) are estimated to have a biomass of up to 200 million tonnes (Lourenco et al. 2017;Saunders et al. 2018). As a result, the SPO is considered the area with the highest abundance of myctophid fish in the world. Myctophids play a crucial role in these ecosystems since they are prey for marine mammals, birds, invertebrates, and other fish (Goetsch et al. 2018;Stewart et al. 2018 the myctophid fish can contribute to the long-term "Blue Growth" strategy set by the European Union (St. John et al. 2016). Human demands for protein from the marine environment have continued to intensify. Consequently, many fish stocks are at risk of collapse (FAO 2018), and there is a pressing from the industry to find unexploited fish stocks that can help supply this demand. Since 70's, mesopelagic fish have been the subject of developing new large-scale fisheries (Standal and Grimaldo 2021). For instance, in the Indian Ocean the myctophid Benthosema pterotum have been identified as a potential resource for future exploitation (Valinassab et al. 2007). However, empirical data showed the mesopelagic fish trends do not correspond to the assumption that these fish are a potential future resource (Standal and Grimaldo 2021) and also, their role in ecosystems is limited (St. John et al. 2016). Here, we use stable isotopes of carbon and nitrogen to characterize key features of the trophic ecology of South Pacific myctophids prior to their commercial exploitation, including the isotopic niche and trophic position (TP). The transfer of energy and nutrients from the bottom to the top links makes TP a critical concept used to describe the functional role of consumers in a food web (Quezada-Romegialli et al. 2018).
TP has historically been estimated via stomach contents through the mass and the number of prey, but those results can be over-or under-estimated due to consumer digestive biases (Hetherington et al. 2017). In recent decades, stable isotope analyses (SIA) of nitrogen (δ 15 N) and carbon (δ 13 C) have been used as a complementary approach that can reduce biases associated with stomach contents analysis. Values of predator-stable isotope composition revealed by SIA reflect prey assimilation (Goetsch et al. 2018). A consumer's carbon stable ratios typically reflect the primary production mode fueling the taxa comprising its diet. Meanwhile, δ 15 N values can be considered an indicator of TP occupied by prey and their consumers (Fry 2013). For instance, in tropical Atlantic waters, Olson and Watters (2003) examined TP in Myctophum nitidulum and Symbolophorus reversus using stomach contents, estimating a mean TP of 3.2. In contrast, stable isotopes estimated the same species to have a TP of 2.6 and 2.7, respectively (Hetherington et al. 2017).
Stable isotope values can also be used to estimate the isotopic niche-a compelling alternative to Elton's (1927) traditional definition of the trophic niche (Cherel et al. 2010). Layman et al. (2007a) developed quantitative methods using stable isotopes to measure trophic community structure. This isotopic approach is established how individuals' chemical composition is directly influenced by what they consume. Therefore, predators with high δ 15 N and a wide range of δ 13 C values will have a higher TP, hence a wide isotopic niche size (Herzka 2005). Thus, the variability in TP values displayed by a species could determine the size of the isotopic niche and provide a broader view of the species life history characteristics. We calculated TP and the isotope niche of the most abundant lanternfishes from three different areas in the southeastern Ocean Pacific to compare their ecological function.

Samples collection
We analyzed a total of 123 individuals captured during a fisheries-independent research survey in 2015. The study area was divided into three areas ( Fig. 1): Peru (n = 50), Central-Chile (n = 39), and South-Chile (n = 33). A midwater trawl captured all fish from a depth range between 200 and 300 m. After capture, they were labeled and frozen at − 20 °C until further analysis in the laboratory. We also collected water and zooplankton samples. For this purpose, 20 L of water and a random set of zooplankton (plankton net WP2, ring diameter of 57 cm; mesh with 1.5 µm and length 2.6 m) were taken. The water was stored under total darkness, while the zooplankton was filtered, separated, dehydrated, and then stored under a vacuum.

Lab work
Myctophids were thawed and identified (Table 1). Approximately one gram of wet weight from the muscle tissue was taken from each individual. Seawater samples were filtered in the laboratory and divided into two fiberglass filters (FGP Whatman™ 1.5 µm, 4.7 cm). The filter for δ 13 C was placed in a desiccator with 20 ml of 37% HCL to remove inorganic carbon; meanwhile, the filter for δ 15 N was put in aluminum foil at − 80 °C (Feuchtmayr and Grey 2003;Lorrain et al. 2003). The particulate organic matter (POM) and zooplankton left in the filters were removed and placed in individual plates. Then, all samples were dried (55 °C for 48 h) and pulverized. Subsamples of 400-500 µg were deposited in tin capsules and then analyzed in a mass spectrometer in continuous flow mode (CF) "Nu-Instruments," coupled with an elemental analyzer (EA) of Eurovector, EA-3024. Isotope radios were reported in δ notation, using standard Pee Dee Belemnite for δ 13 C and Atmospheric Nitrogen for δ 15 N. Therefore, δ 13 C or δ 15 N = [(R sample/R standard) − 1], where R is 13 C/ 14 C or 15 N/ 14 N, respectively. This analysis had an accuracy ± 0.1‰ for δ 15 N and δ 13 C.

Data analysis
For those C:N values that exceeded 3.5, we made a lipid correction to the δ 13 C values following Kiljunen et al. (2006). We calculated trophic position (TP) using two contrasting taxa as baselines based on the bulk SIA values. First, we assumed POM TP (λ) as one and copepods TP as λ = 2. Then, using the "tRophicPosition" package with Bayesian inference by the One baseline model (Quezada-Romegialli et al. 2018), we calculated myctophid TP for each of the three zones using a trophic discrimination factor (TDF) of ± SD (Δ 15 N) of 3.4 ± 1‰ (Post 2002). Therefore, TP was calculated following this equation: where δ 15 N M and δ 15 N b refer to the nitrogen values of the myctophids (M) and baseline (b), respectively. The standard ellipse area (SEA), followed by a convex-hull analysis was estimated to infer the isotopic niche's amplitude (Layman et al. 2007b). The "SIBER" package with Bayesian inference was used for this analysis (Jackson et al. 2011). A PERMANOVA: n permutations = 9.999 test was used, based on a Euclidean similarity matrix to compare the values of δ 15 N and δ 13 C. A general linear model (GLM) was used to compare the values of δ 15 N and SEA of myctophids in the different areas, and a one-way ANOVA was applied to compare the mean values of POM and copepods.
The average TP estimates using POM as the baseline (TP POM ) varied in higher ranges than those TP estimates based on copepods (TP copepods ) ( Table 2). The posterior Bayesian TP calculated for Pacific myctophids for each fishing ground was similarly dependent on the baseline (Fig. 3b). This variation could be explained by many factors, such as feeding rates and tissues turnover rates. Our finding on TP (Table 2) of lanternfishes in the SPO was similar to those obtained in other works for analogous species (Cherel et al. 2010;Bernal et al. 2015;Hetherington et al. 2017). Although the Peruvian and Chilean mesopelagic zones are considered similar oceanographically (Sutton et al. 2017), lanternfishes from the different fishing grounds showed differences in their stable isotope values. These differences likely reflect regional differences in upwelling. For example, the region of Peru sampled here has a seasonal pulse of upwelling. In contrast, the Chilean areas are subject to more permanent upwelling and, therefore, a continuous pattern of elevated primary production, explaining the isotopically enriched POM (Gutiérrez et al. 2016). Indeed, Pizarro et al. (2019 reported 15 N-enriched POM values in Northern Chile and associated this with upwelling. The Southern Pacific Ocean's lanternfishes could play an active and key role in carbon flow in the trophic web as these fishes represent prey for predators associated with different marine habitats, including surface and deep waters. In fact, Hudson et al. (2014) proposed the proportion of carbon flux carried out by the lanternfish at greater depths confirms the importance of myctophids in the biological pump of deepsea zones of the Atlantic Ocean. Olivar et al. (2019) describe secondary consumers' role in the deep-sea ecosystem, in which some species have a vertical migration in the water column. Castro et al. (2010) found the same behavior in some lanternfishes of the SPO. This behavior results in myctophid fishes transferring energy from the upper layers of water towards the depths. Conversely, when lanternfishes migrate to the surface, they are prey for pelagic predators, including penguins, sharks, and marine mammals. However,  when they migrate into deeper waters, they are prey for mesopelagic predators such as hake and squid. Therefore, our results point out that the lanternfishes from SPO likely play a primary role in the energy flow from deep waters to the surface. Myctophids are typically considered as a single homogenous guild, there are doubtless ways in which the species have differentiated. Given that our results show spatial differences in the ecological function of the guild, future studies in the SPO should compare the trophic ecology of individual species across different areas, and thus be able to adapt to the global fishery management strategy, who intend to recover valuable exploited fish, and to achieve better fisheries management in areas beyond national jurisdiction, a moratorium on deep-water bottom trawling, and no take in a vulnerable marine ecosystem (Wright et al. 2015).
Acknowledgements The authors are grateful for the valuable work done by the scientific observers of IFOP and IMARPE on the research vessels.
Author contributions All authors contributed to the study's conception and design. Material preparation, data collection and analysis were performed by SAK, IQ, FV, PE, MZ, PG, RV, AS, SC-L. The first draft of the manuscript was written by SAK, CC, ES, PE, CH, FF, AG, CC-C and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author upon reasonable request.

Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.

Ethics approval
The authors from the competent authorities have obtained all necessary permissions for sampling and observational field studies. The research cruises counted all the legal authorizations for fishing research that Perú and Chile's laws maintain in their territory.

Consent to publish
This work did not have any person as a study subject.

Consent to participate
This work did not have any person as a study subject.