Age and Growth of Predatory Mesopelagic Fishes in a Low- Latitude Oceanic Ecosystem

Home , Hammerjaw, Lampanyctus, Lampanyctus festivus, Lampanyctus pusillus, Myctophum affine, Myctophum nitidulum

Nova Southeastern University NSUWorks

All HCAS Student Capstones, Theses, and Dissertations HCAS Student Theses and Dissertations

7-30-2020

Natalie Slayden Nova Southeastern University

Follow this and additional works at: https://nsuworks.nova.edu/hcas_etd_all

Share Feedback About This Item

NSUWorks Citation Natalie Slayden. 2020. Age and Growth of Predatory Mesopelagic Fishes in a Low-Latitude Oceanic Ecosystem. Master's thesis. Nova Southeastern University. Retrieved from NSUWorks, . (10) https://nsuworks.nova.edu/hcas_etd_all/10.

This Thesis is brought to you by the HCAS Student Theses and Dissertations at NSUWorks. It has been accepted for inclusion in All HCAS Student Capstones, Theses, and Dissertations by an authorized administrator of NSUWorks. For more information, please contact [email protected]. Thesis of Natalie Slayden

Submitted in Partial Fulfillment of the Requirements for the Degree of

Master of Science Marine Science

Nova Southeastern University Halmos College of Arts and Sciences

July 2020

Approved: Thesis Committee

Major Professor: Tracey Sutton, Ph.D

Committee Member: David Kerstetter, Ph.D

Committee Member: Paul Arena, Ph.D

This thesis is available at NSUWorks: https://nsuworks.nova.edu/hcas_etd_all/10 NOVA SOUTHEATERN UNIVERSITY

HALMOS COLLEGE OF ARTS AND SCIENCES

Age and Growth of Predatory Mesopelagic Fishes in a Low-Latitude Oceanic Ecosystem

Natalie Slayden

Submitted to the Faculty of Halmos College of Arts and Sciences in partial fulfillment of the requirements for the degree of Master of Science with a specialty in:

Marine Biology

Nova Southeastern University

August 8, 2020

ABSTRACT Mesopelagic and bathypelagic fishes provide important global ecosystem services, such as carbon sequestration via the biological pump and provision of food for economically important (billfishes and tuna) and federally protected (cetaceans and seabirds) species. These attributes are becoming increasingly recognized, while simultaneously mesopelagic fisheries are becoming of interest for direct harvest as coastal fisheries have become overexploited. Additionally, climate change, ocean acidification, and seabed mining threaten deep-sea fishes. With increasing interest in deep-sea fisheries and anthropogenic threats, age and growth information on these fishes is necessary for management and conservation. Currently ecosystem models lack data such as sexual maturity and lifespan on deep-pelagic fishes, hindering our ability to quantify production rates and resilience to disturbance. Here we examine four numerically dominant predatory fishes from the Gulf of Mexico exhibiting a range of trophic ecologies and vertical distributions: Lampanyctus lineatus (Myctophidae), Omosudis lowii (Omosudidae), Stomias affinis (Stomiidae), and Chauliodus sloani (Stomiidae). In this thesis, the otoliths (‘ear stones’) of each species were examined in order to estimate age and duration of specific life history stages (e.g. juvenile, intermediate, and adult) and landmarks (e.g., sexual maturity and determine longevity). Given that otolith ring validation was not possible (these fishes cannot be kept alive, marked and recaptured, and hourly sampling is not possible for species living at great depths), we present putative minimum and maximum estimates at landmarks based on two scenarios: 1) total rings = days of life; and 2) major (darkest) otolith increments = years of life). Comparing both estimates to available validated ages of their prey, we conclude that the maximum age scenario is most appropriate. Results of this study suggest that the deep-living myctophid species and higher-level predators investigated have relatively long generation times (L. lineatus one year, C. sloani 9 - 17 years, and S. affinis 10 – 19 years until sexual maturity), and thus likely have low resilience to population level-perturbation.

Keywords: Otoliths, Deep-pelagic, Age, Growth, Mesopelagic, Bathypelagic

ACKNOWLEDGEMENTS I would first like to thank my advisor, Dr. Tracey Sutton, for his guidance throughout my time at NSU. The knowledge and experience I gained while working with Dr. Tracey Sutton are indispensable. I would also like to thank my committee members, Dr. David Kerstetter and Dr. Paul Arena, for their time and the knowledge they shared with me during my thesis. Another special thanks to Dr. John Gartner for the email conversations full of good advice and guidance when I first started to process otoliths.

I would also like to thank my lab mates in the Oceanic Ecology Lab for all the laughs and support. I have had endless good memories of long nights in lab together, attending research conferences, going on research cruises, and participating in lab canoe race competitions. You all kept me going.

Lastly, I would like to thank my family and friends specifically my Mom, Brother, Memaw, Brendan, Kirby, and Dakota. I could not have done all this without you. I am so happy to have you all as a part of my team.

iii

TABLE OF CONTENTS

ABSTRACT ...... ii

ACKNOWLEDGEMENTS ...... iii

TABLE OF CONTENTS ...... iv

LIST OF TABLES ...... vi

LIST OF FIGURES ...... vii

1. INTRODUCTION...... 1

1.1. The open ocean zones and the deep-pelagic environment ...... 1

1.2. Global importance of mesopelagic and bathypelagic fishes ...... 1

1.2.1. Global threats of mesopelagic and bathypelagic fishes ...... 2

1.3. The Gulf of Mexico and the Deepwater Horizon oil spill ...... 3

1.4. The necessity of fish ageing in the open ocean environment ...... 4

1.5. Introduction to age determination ...... 4

1.6. A summary of information on deep-pelagic otoliths ...... 6

1.7. Objectives ...... 10

2. Methods ...... 11

2.1. Sampling collection and processing...... 11

2.2. Otolith Processing ...... 12

2.2.1. Otolith extraction ...... 12

2.2.3. Otolith morphology ...... 12

2.2.4. Otolith preparation ...... 15

2.2.5. Otolith increment interpretation ...... 16

2.2.6. Otolith microincrement analysis ...... 16

2.3. Case Studies ...... 17

3. Results ...... 19

3.1. Chauliodus sloani ...... 19

3.1.1. Length-weight regressions ...... 19

3.1.2. Size-frequency distribution and vertical distribution ...... 19

3.1.3. Length-increment regressions and lifespan ...... 21

3.1.4. External otolith morphology ...... 22

3.1.5. Otolith length and otolith width vs. SL regressions ...... 23

3.1.6. Chauliodus sloani internal otolith morphology ...... 23

3.2. Stomias affinis ...... 26

3.2.1. Length-weight regression ...... 26

3.2.2. Size frequency of occurrence and vertical distribution ...... 27

3.2.3. Length-increment regressions and lifespan ...... 28

3.2.4. External Otolith morphology ...... 30

3.2.5. Otolith length and otolith width-fish length regressions...... 30

3.2.6. Stomias affinis internal otolith morphology ...... 31

3.3. Omosudis lowii...... 34

3.3.1. Length-weight regression ...... 34

3.3.2. Size frequency of occurrence and vertical distribution ...... 35

3.3.3. Length-increment regressions and lifespan ...... 37

3.3.4. Omosudis lowii external otolith morphology ...... 38

3.3.5. Otolith length and otolith width-fish length regressions...... 38

3.3.6. Omosudis lowii internal otolith morphology ...... 39

3.4. Lampanyctus lineatus...... 42

3.4.1. Length-weight regression ...... 42

3.4.2. Size frequency of occurrence and vertical distribution ...... 43

3.4.3. Length-increment regressions and lifespan ...... 44

3.4.4. Lampanyctus lineatus external otolith morphology ...... 46

3.4.5. Otolith length and otolith width-fish length regressions...... 46

3.4.6. Lampanyctus lineatus internal otolith morphology ...... 47

3.5. Summary of all fishes ...... 49

4. Discussion...... 52

4.1. The importance of aging fishes and the rationale of this study ...... 52

4.2. Vertical migration patterns ...... 53

4.3. Growth patterns ...... 54

4.4. Age at maturity ...... 55

4.5. Longevity ...... 55

5. Conclusions ...... 56

6. References ...... 57

LIST OF TABLES

Table 1. Age records for deep-pelagic fishes collected in various ocean basins.* indicates demersal species occurring in the pelagic domain as juveniles...... 7

Table 2. Metrics of gear types utilized during Pisces and DEEPEND research cruises to collect fishes ...... 12

Table 3. The age estimations of juvenile, intermediate, and adult Chauliodus sloani if the total number of increments represent days and dark major increments represent years...... 22

Table 4. Kendall’s tau correlation test on the left and right otoliths of Chauliodus sloani...... 23

Table 5. The age estimations of juvenile, intermediate, and adult Stomias affinis if the number of increments represent days and dark major increments represent years...... 30

Table 6. Pearson’s correlation test on right and left otoliths of Stomias affinis...... 30

Table 7. The age estimations of juvenile, intermediate, and adult Omosudis lowii if the number of increments represents days and dark major increments represent years...... 38

Table 8. Kendall’s tau correlation test on left and right otoliths of Omosudis lowii...... 39

Table 9. The age estimations of juvenile, intermediate, and adult Lampanyctus lineatus if the number of increments represents days and dark major increments represent years...... 45

Table 10. Kendall’s tau correlation of left and right otoliths of Lampanyctus lineatus...... 46

Table 11. Length-weight regressions of Chauliodus sloani, Stomias affinis, Omosudis lowii, and Lampanyctus lineatus...... 49

Table 12. Length-increment regressions and Otolith-SL regressions of Chauliodus sloani, Stomias affinis, Omosudis lowii, and Lampanyctus lineatus...... 51

LIST OF FIGURES

Figure 3. a.) Otolith components and terminology. b.) Terminology used for otolith shape. c.) Otolith anterior and posterior shapes. d.) Otolith margin terminology (Tuset et al., 2008)...... 14

Figure 4. Ostial opening and sulcus shape terminology utilized during this study (Schwarzhans, 1978) ...... 15

Figure 5. An image of Chauliodus sloani caught in the Gulf of Mexico. Image take by Danté Fenolio...... 17

Figure 6. An image of Stomias affinis caught in the Gulf of Mexico. Image taken by Danté Fenolio...... 18

Figure 7. An image of Omosudis lowii from the Harvard Museum of Comparative Zoology. .... 18

Figure 8. Length-weight regression of Chauliodus sloani...... 19

Figure 9. Length frequency of all Chauliodus sloani collected during DEEPEND and Pisces cruises. The red boxes represent: a.) juveniles, b.) intermediates, and c.) adult individuals...... 20

Figure 10. The vertical distribution pattern of: a.) all Chauliodus sloani collected during DEEPEND cruises; b.) juveniles, c.) intermediate, and d.) adult size classes...... 21

Figure 11. Otolith increment regressions for Chauliodus sloani...... 21

vii

Figure 12. Otoliths of Chauliods sloani at 100 mm SL, 80 mm SL, and 34 mm SL. Scale bars are set to 100 microns...... 22

Figure 13. Otolith length and width vs. SL regressions for Chauliodus sloani...... 23

Figure 14. The number of zones in each otolith vs. SL regression for Chauliodus sloani...... 24

Figure 16. Plots of the zone length, increment width, and the number of increments within each zone of individual otoliths of Chauliodus sloani...... 26

Figure 17. Length-weight regression for Stomias affinis...... 27

Figure 18. Length frequency of all Stomias affinis collected during DEEPEND and Pisces cruises. The red boxes represent: a.) juveniles, b.) intermediates, and c.) adult individuals...... 28

Figure 19. The vertical distribution pattern of: a.) all Stomias affinis collected during DEEPEND cruises; b.) juveniles, c.) intermediate, and d.) adult size classes...... 28

Figure 20. Otolith increment regressions of Stomias affinis...... 29

Figure 21. Otoliths of Stomias affinis at 181 mm SL, 98 mm SL, and 47 mm SL. Scale bars are set to 100 microns...... 30

Figure 22. Otolith length and width vs. standard length (mm) regressions for Stomias affinis. ... 31

Figure 23. The number of zones in each otolith vs. SL regression for Stomias affinis...... 32

Figure 25. Plots of zone length, increment width, and the number of increments within each zone of individual otoliths of Stomias affinis...... 34

Figure 26. Length – weight regression of Omosudis lowii...... 35

Figure 27. Length frequency of all Omosudis lowii collected during DEEPEND and Pisces cruises. The red boxes represent: a.) juveniles, b.) intermediates, and c.) adult individuals...... 36

Figure 28. Vertical distribution of Omosudis lowii...... 36

Figure 29. Otolith increment regressions of Omosudis lowii...... 37

Figure 30. Otoliths of Omosudis lowii at 164 mm SL, 84 mm SL, and 24 mm SL. Scale bar is at 100 microns...... 38

viii

Figure 31. Otolith length and width vs. standard length regressions of Omosudis lowii...... 39

Figure 32. The number of zones in each otolith vs. SL regression for Omosudis lowii...... 40

Figure 34. Plots of zone length, increment width, and the number of increments within each zone of individual otoliths of Omosudis lowii...... 42

Figure 35. Length-weight regression of Lampanyctus lineatus...... 43

Figure 36. Length frequency of all Lampanyctus lineatus collected during DEEPEND and Pisces cruises. The red boxes represent: a.) juveniles, b.) intermediates, and c.) adult individuals...... 44

Figure 37. The vertical distribution pattern of: a.) all Lampanyctus lineatus collected during DEEPEND cruises; b.) juveniles, c.) intermediate, and d.) adult size classes...... 44

Figure 38. Otolith increment regressions of Lampanyctus lineatus...... 45

Figure 39. Otoliths of Lampanyctus lineatus at 71 mm SL, 57 mm SL, and 32 mm SL. Scale bars are set at 100 microns...... 46

Figure 40. Otolith length and width vs. standard length (mm) regressions for Lampanyctus lineatus ...... 47

Figure 41. Number of zones – standard length (mm) regression of Lampanyctus lineatus...... 47

Figure 43. Plots of zone length, increment width, and the number of increments within each zone of individual otoliths of Lampanyctus lineatus...... 49

Figure 44. ShapeR analysis of the otoliths of Chauliodus sloani, Stomias affinis, Omosudis lowii, and Lampanyctus lineatus...... 52

1. INTRODUCTION

1.1. The open ocean zones and the deep-pelagic environment The open ocean can be classified into zones by depth. The epipelagic zone is the upper- most, sunlit zone from the surface to 200 m. The epipelagic zone is the most explored and known zone. This bathome contains enough light for photosynthesis to occur (Sutton, 2013). Below the epipelagic zone lie the mesopelagic and bathypelagic zones, which collectively make up the deep- pelagic bathome. The mesopelagic zone extends from 200 m to 1000 m and is often referred to as the twilight zone since this bathome has enough light for organisms to tell day from night, but not enough light for photosynthesis (Robinson et al., 2010; Sutton et al., 2017). In the mesopelagial, the environment changes with respect to light and temperature as depth increases. The majority of the organisms that live in the mesopelagial use bioluminescence to communicate with potential mates, lure in prey, or avoid predation. Most bioluminescent species undertake diel vertical migrations to the surface to feed at night and swim back to depth to hide from predators during the day (Sutton, 2013). Mesopelagic fishes also possess visual adaptations that allow them to utilize dim downwelling light (Robinson et al., 2010). The bathypelagic zone extends from 1000 m to ~100 m above the ocean floor and is characterized by the complete absence of sunlight and little variation in temperature. Organisms that live in this bathome usually have watery tissue, are often dark black, brown, or red in color, have small eyes, and a reduction in photophores (Sutton, 2013). This study will focus on deep-pelagic fishes that that live primarily within the mesopelagic zone, but may span into the bathypelagic zones.

1.2. Global importance of mesopelagic and bathypelagic fishes Globally, mesopelagic and bathypelagic fishes are economically and ecologically important. Mesopelagic fishes play a key role in connecting the epipelagic and mesopelagic food webs due to their daily vertical migration from deep depths to the epipelagic zone at night and then back to depth before daybreak (Davison et al., 2015). In the epipelagic zone at night, these vertical migratory fishes are a food source for economically important fishes such as tuna and billfishes, along with marine mammals (Gjosæter and Kawaguchi, 1980; St. John et al., 2016). Additionally, mesopelagic fishes play an important role in the biological pump through their consumption of

zooplankton (Longhurst et al., 1989). The vertically migrating species transport prey carbon to deeper depths, where it is sequestered via respiration and waste production, thus drawing down atmospheric carbon dioxide (Proud et al., 2017).

Bathypelagic fishes and non-vertical migratory mesopelagic fishes also play an important role in the biological pump through predatory consumption. As bathypelagic and non-vertical migrating mesopelagic fishes consume vertically migrating mesopelagic fishes. This carbon remains remain at depth due to its release through waste production below the thermocline (Wang et al., 2019).

1.2.1. Global threats of mesopelagic and bathypelagic fishes Mesopelagic fishes are a potential fishery target because of their daily vertical migrations, which concentrate individuals in shallower waters at night and allows these fishes to be caught more easily (Moore, 1999). The biomass of mesopelagic fishes is estimated at 10 billion tons (Irigoien et al., 2014; St. John et al., 2016). As human population steadily increases, the concern with food security rises. The continuous overexploitation of coastal fisheries is also an impetus for mesopelagic fisheries, whose product would be fishmeal for aquaculture. Likewise, the demand for omega-3 oil supplements for human dietary needs has triggered fisheries for mesopelagic fishes due to their high fatty acid content (St. John et al., 2016). Norway and Pakistan have already begun licensing several fishing vessels to catch mesopelagic fishes (The Economist, 2017). The increasing interest in mesopelagic fisheries may be symbolic of the “fishing down the food web” paradigm (Pauly et al., 1998). As epipelagic fisheries resources increasingly become overexploited (Worm et al., 2009), it is likely that new fisheries will develop for deep-pelagic fishes.

Age and size structure are key components of fish population dynamics (Brunel and Piet, 2013). Fish population and stock assessment models require age-related parameters such as growth rate, mortality, length at age, age at maturity, and mean age (Pope et al. 2010). Fishes with a longer lifespan will generally take longer to replace losses than those with shorter lifespans. For example, some bathypelagic fishes, such as members of the family Melamphaidae, use most of their energy for growth rather than reproduction, becoming reproductively mature later in life (~ seven to nine years; Childress et al., 1980). If a disturbance affects melamphaid assemblages, these fishes will have difficulty recovering since they reproduce much later.

Mesopelagic and bathypelagic fishes are threatened by climate change, ocean acidification, deoxygenation, oil extraction (section 1.3), and seabed mining, in addition to the potential challenges inherent in fisheries management. Climate warming could threaten the carbon-sink ecosystem service of mesopelagic fishes as increasing temperatures promote ocean stratification. Stratification prevents ocean mixing from delivering oxygen to deeper depths, which would expand the Oxygen Minimum Zone (OMZ). The increase in the oxygen-depleted layer could prevent fishes from passing through this zone, causing them to remain either below or above the OMZ. The removal of the vertical migration of fishes due to the expansion of the OMZ may not allow the fishes to transport carbon to the deep ocean during their decent. Climate change may also change the geographic distribution of fishes due to increasing temperatures. Additionally, the input of anthropogenic carbon dioxide threatens to lower the pH of the ocean making it more acidic. Acidification can suppress fish metabolism and prevent animals from growing or reproducing (Robison, 2009).

In order to understand the effects of numerous threats on mesopelagic and bathypelagic fishes, baseline information is needed (Robison, 2009). Age and growth studies, such that presented here, provide valuable information regarding life history events and production rates of fishes, which could assist in determining how environmental and anthropogenic changes will affect mesopelagic and bathypelagic fish assemblages in the future (Campana and Jones, 1992). The importance of age and growth studies on deep-pelagic fishes is highlighted in Section 1.4.

1.3. The Gulf of Mexico and the Deepwater Horizon oil spill The Gulf of Mexico (GOM) is a semi-enclosed basin surrounded by land. It is fueled with an inflow of warm water from the Caribbean Sea that transports fishes into the GOM. This influx of water forms the GOM Loop Current, which then exits through the Straits of Florida, forming the Florida Current, and then entering the western North Atlantic, where it becomes the Gulf Stream. The Gulf Stream pushes warm water up the western boundary of the North Atlantic Ocean. Even with these connections to the Caribbean Sea and the Atlantic Ocean, the GOM is distinct with warm tropical waters, winter water cooling, and the input of freshwater from the Mississippi River. Due to its ecotonal geographic position, the GOM contains one of the most speciose mesopelagic fish assemblages known in the World Ocean (Sutton et al., 2017).

The Deepwater Horizon oil spill was a disaster that occurred in the northern GOM in 2010. This spill lasted for 87 days, spilling around 4.9 million barrels of oil into the GOM (McNutt et al., 2012). This oil spill occurred from the deep Macondo well at a depth of 1500 m. Deep plumes were reported, with one hydrocarbon plume that was documented at around 1000 – 1200 m depth (Camilli et al., 2010). This oil plume presumably impacted deep-pelagic fishes due to their behavioral habits. Deep-pelagic fishes live and swim through these depths as they undergo diel vertical migration; thus, deep-pelagic fishes were exposed and vulnerable to the oil. The vulnerability of deep-pelagic fishes has further been proven through their significant decline in standardized abundances after the spill (Sutton, in review). With the decline in deep-pelagic fish abundance, we now ask the question “How resilient are these fishes to disturbances?” Age and growth information can be used to understand the resilience of deep-pelagic fishes.

1.4. The necessity of fish ageing in the open ocean environment Growth rates and longevity are crucial parameters needed to quantify and understand population dynamics (Greely et al. 1999). For deep-pelagic fishes, information is lacking on lifespan, age at first reproduction, and growth rate for all but a few species. Only four records of age exist for mesopelagic fishes collected in the GOM, of which three have been validated (Gatner, 1991; Lancraft et al., 1998; Section 1.6).

Ecosystem models have been developed to determine the impact that disturbances have on the environment. The Atlantis Ecosystem Model is a framework model to be used as a management tool that was created for the GOM. This model, in addition to many ecosystem models, characterize organisms into functional groups by utilizing information on growth rate, mortality, and age at maturity (Ainsworth et al., 2015). With respect to deep-pelagic fishes, these basic life-history traits are mostly unknown, making it difficult to understand how disturbances affect deep-pelagic fishes and how management decisions will impact these fishes.

1.5. Introduction to age determination The age of a fish can be determined through the examination of various body structures, including scales, otoliths, opercular bones, fin rays, and vertebrae. Each structure possesses a number of rings or increments that increase in number with age. Changes in deposition rate on several timescales (hourly, daily, seasonally, yearly) create patterns of ring-like bands. Otoliths are the standard structures used for aging studies of teleosts (bony fishes; chondrichthyans lack

otoliths) and will be examined in this study. Otoliths are present at the earliest developmental stage providing initial growth increments, whereas scales develop at the end of the larval stage (Campana and Neilson, 1985; Quist and Isermann, 2017). Additionally, scales and fin rays can be absent, damaged, grow irregularly, or even be resorbed during periods of biological stress, making them less accurate for aging studies (Campana and Neilson, 1985).

An otolith, sometimes referred to as an ear stone, is made up of calcium carbonate and inorganic salts that form within a protein matrix. There are three pairs of otoliths in an individual fish, the sagitta, lapillus, and asteriscus, which are located in the inner ear canals within the cranial cavity of fishes (Figure 1a) (Campana, 2004; Quist and Isermann, 2017). The three pairs of otoliths are positioned within compartments in semi-circular canals (Figure 1b). The sagitta is located in the saccule compartment, the lapillus positioned in the utricle, and the asteriscus in the lagena (Campana, 2004; Tuset et al., 2008). The canals and compartments are fluid-filled. Sensory macula covered in cilia are present on the compartment walls. Since the otoliths are the only solid object within the canals and compartments, the slight movement of the otoliths within the fluid and the movement of the macular cilia indicate orientation, providing balance for fishes (Campana, 2004; Quist and Isermann, 2017). Otoliths also provide sound detection for most fishes due to their higher densities (sound energy is absorbed at a different frequency than the rest of the fish’s body). The lapillus is usually associated with balance while the sagitta is associated with sound detection (Campana, 2004).

A B

Figure 1 (a.) Location of each otolith within the cranial cavity (Campana, 2004). (b.) The fluid-filled semicircular canals of the deep-pelagic fish Poromitra crassiceps, with each otolith within its compartment (Deng et al., 2013). Otoliths provide information used in the fields of ichthyology, paleontology, archaeology, and zoology (Tuset et al., 2008). Otoliths can be used to identify fishes since otolith shape is often

species-specific, which is valuable in understanding taxonomy and trophic interactions (Campana, 2004; Gimenez et al., 2016). The shape of preserved otoliths can also be used to identify fossils. Since otoliths are somewhat resistant to degradation, diet analysis can be performed by examining otoliths in the digestive tracts of predators. An otolith’s chemical composition can reveal life history information, such as year of hatch and migration behavior along with identifying the populations and the location where they live or have lived. Of most importance to this thesis, an otolith can reveal age and ontogeny (Campana, 2004; Tuset et al., 2008).

Since the sagitta is usually the largest otolith, it is commonly used for age and growth analyses. Within the sagitta is a sequence of rings or increments. The rings are deposited as a white calcium carbonate band (slow growth) followed by a dark protein-rich band (fast growth). This sequence begins at the primordia (center of growth) within the otolith and continues with age as the fish grows (Campana and Neilson, 1985). Otoliths are valuable for age analysis because they record the age of the fish from the hatch date until death (Campana, 2004). Life stages, stressful events, and habitat shifts can be etched into the otolith microstructure due to their effects on growth rate (Campana and Neilson, 1985; Quist and Isermann, 2017). A dark cluster of bands can indicate metamorphosis or the migration of larvae from the surface waters to the deep-sea zones (Gartner, 1991a). Checks, which are defined as deeply etched increments, can also be present in otoliths and usually indicate sexual maturity or other stressful events that cause absence or reduction in somatic growth (Campana and Neilson, 1985).

1.6. A summary of information on deep-pelagic otoliths Most age studies have been performed on fishes that inhabit the coastal and/or epipelagic zones. Age and growth studies on mesopelagic and bathypelagic fishes are rare, and the vast majority of these studies focused on the mesopelagic fishes within the family Myctophidae (Table 1). The age of only four fishes occurring in the GOM, three myctophids, and one gonostomatid, exist, all of which are lower-trophic-level, planktivorous fishes (Gartner, 1991a; Gartner, 1991b, Lancraft et al., 1998). This study will add age records for dominant mesopelagic top predators, plus the numerically dominant, large-bodied, mid-trophic-level lanternfish in the GOM. Annual increments are commonly utilized for epipelagic fish ageing. However, for mesopelagic and bathypelagic fishes, it is often uncertain what their increments represent (days, years, feeding events, metamorphosis) due to the major data gaps present for these fishes (Webb et al., 2010).

Table 1. Age records for deep-pelagic fishes collected in various ocean basins.* indicates demersal species occurring in the pelagic domain as juveniles.

Size at Age at Max Size Lifespan Reference Maturity Maturity Chloropthalmidae Chlorophthalmus agassizii* >10 years D’Onghia et al., 2006 Gonostomatidae Cyclothone alba 1 year Maynard, 1982 Male 16.7 mm SL 1-2 years 21 mm SL 2 years Miya & Nemoto, 1986a Female 25 mm SL 2 years 26 mm SL 2 years Miya & Nemoto, 1991 Cyclothone pallida 2.4 years Maynard, 1982 Male 28.9 mm SL 3-4 years Miya & Nemoto, 1987 Female 40-45 mm SL 4-5 years Miya & Nemoto, 1987 Cyclothone pseudopallida 2-3 years 1.2 years Maynard, 1982 Male 20 mm SL 2-3 years 25.5 mm SL 3 years Miya & Nemoto, 1986b Female 30.7 mm SL 2-3 years 38.0 mm SL 4-5 years Miya & Nemoto, 1986b Sigmops elongatus 3 years Krueger and Bond, 1972 Male 110 mm SL 9 months 225 mm SL <2 years Lancraft et al., 1988 Female 135 mm SL 12 months 225 mm SL <2 years Lancraft et al., 1988 Melamphaidae Melamphaes pumilus 24 mm SL 1 year Keene et al., 1987 Melamphaes ebelingi 137 mm SL 3-4 years Keene et al., 1987 Melamphaes typhlops 73 mm SL 2 years Keene et al., 1987 Poromitra crassiceps 9 years Childress et al., 1980 Scopeloberyx opisthopterus 39 mm SL 2 years Keene et al., 1987 Myctophidae

7 Benthosema suborbitale 23 mm SL 140 days 32 mm SL <1 year Gartner, 1991

Bolinichthys supralateralis 97 mm SL 2 years Karnella, 1987

Ceratoscopelus warmingii 51-54 mm SL 3-8 months 78 mm SL >2 years Giragosov et al., 2000 Male 46 mm SL 68 mm SL 9-10 months Linkowski et al., 1993 Female 46 mm SL 68 mm SL >1 year Linkowski et al., 1993 75 mm SL 2 years Karnella, 1987 Diaphus dumerilii 52 mm SL 240 days 63 mm SL 2 years Gartner, 1991 Diaphus effulgens 150 mm SL >2 years Karnella, 1987 Diaphus mollis 50 mm SL 1 year Karnella, 1987 Diogenichthys atlanticus 22 mm SL 1 year Karnella, 1987 Gonichthys coco 45 mm SL 1 year Karnella, 1987 Hygophum hygomii 64 mm SL 1 year Karnella, 1987 Hygophum reinhardtii 45 mm SL 1 year Karnella, 1987 Lampanyctus cuprarius 79 mm SL >2 years Karnella, 1987 Lampanyctus festivus 110 mm SL >1.5 years Karnella, 1987 Lampanyctus lineatus 165 mm SL >3 years Karnella, 1987 Lampanyctus pusillus 39 mm SL 1 year Karnella, 1987 Lepidophanes guentheri 63 mm SL >1 year Karnella, 1987 43 mm SL 180 days 65 mm SL 1 year Gartner, 1991 Lobianchia dofleini 38 mm SL 1 year Karnella, 1987 Myctophum asperum 81.2 mm SL 1 year Hayashi et al., 2001 Myctophum nitidulum 1 year Giragosov & Ovcharov, 1992 80 mm SL >2 years Karnella, 1987 Myctophum selenops 65 mm SL 2 years Karnella, 1987 Notolychnus valdiviae 22 mm SL 1 year Karnella, 1987 Notoscopelus caudispinosus 130 mm SL 2 years Karnella, 1987 Notoscopelus replendens 73 mm SL 2 years Karnella, 1987

Male 1.7 years 4 years Sarmiento-Lezcano et al., 2018 8 Female 2.05 years 4 years Sarmiento-Lezcano et al., 2018

Phosichthyidae Vinciguerria nimbaria 27 mm SL <1 year Clarke, 1974 56 mm SL <1 year Tomas & Panfili, 2000 6-7 months Stequert, Menard, & Marchal 30 mm SL 1 year Menon et al., 1996 Sternoptychidae Agyropelecus aculeatus 2 years 2 years Howell & Krueger, 1987 Agyropelecus hemigymnus 1 year 1 year Howell & Krueger, 1987 30 mm SL 1 year 1 year Kawaguchi & Mauchline, 1987 Sternoptyx diaphana 36 mm SL 1 year Howell & Krueger, 1987 Valenciennellus tripunctulatus 6-9 months 1 year Howell & Krueger, 1987 Stomiidae Astronesthes indicus 117 mm SL 3-4 years Clarke, 1974 Chauliodus sloani Several years Clarke, 1974

Previous studies using microincrement analysis show that mesopelagic fishes that perform diel vertical migrations have otoliths with increments that represent days (Gartner, 1991a). However, it is unknown whether increments represent days or years for mesopelagic fishes that do not vertically migrate daily or bathypelagic fishes that remain at depth (Childress et al., 1980; Sarmiento-Lezcano et al., 2018). Variations in temperature, photoperiod, and food intake in the deep-sea environment make interpreting increments difficult (Campana and Neilson, 1985). This study will estimate lifespan based on two separate assumptions: 1) each discernable increment represents one day, or 2) each major (i.e. dark) increment represents one year. In this manner, a first-order approximation is made for the growth rate and longevity of ecologically important fishes whose otolith increments cannot be validated. Due to the limitations imposed by the nature of deep midwater sampling (e.g., trawl times approach six hours) the marginal increment analysis daily ring validation method of Gartner (1991a, 1991b) was not possible, as this method requires hourly sampling across a diel cycle.

1.7. Objectives The first aim of this study is to provide otolith descriptions with corresponding images for the four target fish species. The majority of deep- pelagic fish species do not have otolith descriptions, so this is an important first step. As an added value, these descriptions can be utilized for identification of these taxa in trophic ecology studies (i.e. otoliths in stomach contents).

The second aim of this study is to determine what an otolith increment represents for a deep-pelagic fish, specifically one that does not vertically migrate, or migrates on a non-daily cycle (termed ‘asynchronous vertical migration’ hereafter). Otolith increment counts and increment pattern analysis were documented and related to fish life histories. Understanding what an increment represents in time for deep-pelagic fishes is essential for age determination.

The third aim of this study is to provide age estimations for the four target fishes from the GOM to better understand production rates and longevity of fishes in the mesopelagic and bathypelagic zones. The fishes examined in this study, Chauliodus sloani, and Stomias affinis, Omosudis lowii, and Lampanyctus lineatus were selected due to their importance as high-level predators (first three species), and numerical dominance (L. lineatus) in the GOM.

2. Methods

2.1. Sampling collection and processing The specimens for this study were collected during two sets of research cruises in the GOM. The Offshore Nekton Sampling and Analysis Program (ONSAP) was developed as a means to assess the damage to deep-sea assemblages due to the Deepwater Horizon oil spill that occurred in 2010. Sampling was performed from 2010 - 2011 on the research vessel NOAA FRV Pisces using a commercial-sized midwater trawl (High-Speed Rope Trawl). The High-Speed Rope Trawl is a large-area net that sampled from the surface to 700 m depth (shallow tow) and from the surface to 1500 m depth (deep tow), both day and night. The large-area net sample scheme allowed collection of larger, more mobile fauna. Pisces 8 (PC8) sampling occurred from December 2 to December 19, 2010. Pisces 9 (PC9) sampling occurred from March 23 to April 6, 2011. Pisces 10 (PC10) sampling was from June 23 to July 13, 2011 and Pisces 12 (PC12) sampling from September 8 to September 27, 2011.

The DEEPEND (Deep Pelagic Nekton Dynamics of the Gulf of Mexico) Consortium conducted six research cruises between 2015 and 2018. Sampling occurred from May 1 to May 8, 2015 (DP01), August 8 to August 21, 2015 (DP02), April 30 to May 14, 2016 (DP03), August 5 to August 19, 2016 (DP04), May 1 to May 11, 2017 (DP05), and July 19, 2018 to August 2, 2018 (DPO6). A Multiple Opening/Closing Net and Environmental Sensing System (MOCNESS) was used for sampling during the DEEPEND research cruises. The MOCNESS is a discrete-depth sampling system that utilizes six nets, sampling with one oblique tow and five nets that were opened and closed remotely based on the depth targeted (Wiebe et al., 1985). The MOCNESS had a 10-m² mouth area and a mesh size of 3 mm. During the DEEPEND cruises, the MOCNESS sampled day and night in depth intervals from the surface to 1500 m (Net 0; N0), 1500 – 1200 m (N1), 1200 – 1000 m (N2), 1000 – 600 m (N3), 600 – 200 m (N4), and 200 m to the surface (N5). This sampling scheme was repeated at 12:00 am and 12:00 pm, which allowed for the understanding of depth distributions and the quantification of vertical migration.

Table 2. Metrics of gear types utilized during Pisces and DEEPEND research cruises to collect fishes

Trawl Type Mouth Area Mesh Profile Tow Speed Target Rectangular 10 m² 3 mm, Discrete 1 – 2 knots Plankton – Midwater uniform depth Micronekton Trawls with Multiple nets (MOCNESS) High-Speed 162 m² 6 – 320 cm, Oblique 5 knots Micronekton Rope Trawl graded mesh – Nekton

Initial sample processing occurred on board the research vessels. Specimens were further identified to species and curated by members of the Oceanic Ecology Laboratory at the Halmos College of Natural Sciences and Oceanography at Nova Southeastern University (NSU). For this study, specimens were stored frozen to prevent otolith degradation. Before otolith processing, fishes were weighed to the nearest 0.001 g and measured for standard length (SL; mm).

2.2. Otolith Processing

2.2.1. Otolith extraction An incision was made near the cranium and sagittal otoliths were removed from the inner ear. Otoliths were used to age deep-pelagic fishes since scales were often damaged or absent due to the trawling net, and because some species have an absence of scales. Sagittal otoliths were the target otolith since they are the largest otoliths. Otoliths were extracted, cleaned with water, set out to dry, and stored in glass vials for later analysis.

2.2.3. Otolith morphology Image analysis was conducted using a dissecting microscope with an attached camera (Axiocam) and Zen software, and imageJ was utilized for otolith morphology measurements. Otolith length, maximum width, and area were measured in microns. Otolith length was measured from the farthest posterior region to the farthest anterior region (Figure 2)Error! Reference source not found.. The maximum width was the longest length from the dorsal margin to the

ventral margin (Campana, 2016). Otolith length and maximum width were examined by plotting non-linear growth curves as a function of fish length.

Figure 2. Length and max width measurements of a Lampanyctus lineatus otolith. Length was measured from the furthest anterior region to the furthest posterior region, and max width was measured from the furthest dorsal region to the furthest ventral region. Otolith morphology (e.g. otolith shape, anterior and posterior shape, sulcus shape, caudal shape, distal shape, and otolith margins; Figure 3; Figure 4.) was examined under a compound microscope. The morphology was documented, and images were taken to assist with otolith morphology descriptions. The morphological terminology used in this study follows the Werner Schwarzhans (1978) and Tuset et al. guides (2008). Otolith morphology was further analyzed to visualize the mean shape of each species by utilizing the ShapeR package which outlines the otoliths by performing a Fourier or Wavelet transformation (Libungan and Palsson, 2015)

Figure 3. a.) Otolith components and terminology. b.) Terminology used for otolith shape. c.) Otolith anterior and posterior shapes. d.) Otolith margin terminology (Tuset et al., 2008).

Figure 4. Ostial opening and sulcus shape terminology utilized during this study (Schwarzhans, 1978) Furthermore, within each species, fish were divided into size classes according to peaks in frequency of occurrence, with these size classes putatively representing life stages (juvenile, intermediate, and adult). Otoliths for each size class were analyzed with ShapeR to determine if there were any variations in otolith shape between each group.

2.2.4. Otolith preparation Sagittal otoliths were mounted on microscope slides such that the plane of growth was parallel to the microscope slide. Cyanoacrylate glue (Krazy Glue, Toagose America Inc.) was used to attach the otoliths to the slide with the proximal side (surface with sulcus) facing upward according to the recommendation of Campana (2016). Once the otoliths were positioned flat on the middle of the microscope slide, a drop of Krazy Glue was placed next to the otolith. Glue was spread around and over the otoliths with a probe until the otoliths were encompassed by glue. Once mounted, the otoliths were left to dry overnight.

Diamond lapping film was used to grind and polish the otoliths. The microscope slides were oriented so that the mounted otolith faced down onto the lapping film. Grinding began using a 30-micron lapping film with light pressure in circular motions followed by polishing with 3- and

1-micron lapping film for 3- to 10-second intervals. Throughout the grinding and polishing process, otoliths were checked using a compound light microscope. Once the first round was finished, the otoliths were assessed to determine whether the other side needed to be polished to display all increments. If so, the mounted otolith was flipped (Campana, 2016).

To flip the mounted otoliths, the slide was submerged in water for 10 to 30 minutes. Then a scalpel was used to cut a small square in the Krazy Glue around the otoliths to dislodge the mounted otoliths. Once the otoliths were detached and dried, they were re-glued by placing the Krazy Glue around the previously cut square-shaped glue sections containing the otoliths’ distal side facing upward. The otoliths were ground and polished for a second time with the same method as was performed on the proximal side of the otolith. Polishing both sides of the otoliths created thin sections that better displayed growth increments (Campana, 2016).

2.2.5. Otolith increment interpretation A compound light microscope with an attached camera (Axiocam) was used to analyze otolith increments. An increment was defined as a unit of a dark and light band (Quist et al., 2017). The counting axis was determined by increment clarity and the shortest radius with the complete sequence (Campana, 1992). Image analysis allowed for more accurate increment interpretation via computer-assisted analyses. Increment counts were determined while also making note of zones and checks. Checks were defined as deeply etched increments, and zones were determined as areas surrounded by dark etched increments. Growth models were used to examine increment counts vs. the standard length (mm) of the fishes. Two methods were utilized to determine age: 1) considering all discernable increment counts as days, and 2) considering major growth intervals (dark bands) as years. Non-linear regressions were created to relate age with fish growth utilizing the two aging methods (days vs. years). The best-fit growth models were determined by the lowest Akaike Information Criterion (AIC) values. Light fine increments with dark major increments on the outer edge (preceding and following the light fine increments) were considered zones. The number of increments per zone was also analyzed by a non-parametric one-way ANOVA Kruskal-Wallis rank to determine if there was a difference in the number of increments between each zone.

2.2.6. Otolith microincrement analysis Imagej software was utilized to measure increment width and zone width within the otoliths of each species. Variation in increment width and zone width was analyzed using a non-parametric

one-way ANOVA Kruskal-Wallis rank. Furthermore, increment width vs. zone width was analyzed with a non-parametric Kendall’s rank correlation to determine if there was a correlation between the two measurements. Otolith microincrement patterns were observed and noted for species descriptions.

Vertical migration was investigated by life stages (i.e. juvenile, intermediate, adult) determined through frequency of occurrence analysis for each species to identify and treat vertical migration patterns as landmarks. This rationale was utilized throughout the study with age reported per each life stage.

2.3. Case Studies Chauliodus sloani (Figure 5), the viperfish, is a stomiiform fish within the family Stomiidae (dragonfishes). Chauliodus sloani are numerically dominant predators that represent the majority (38.7%) of stomiid fishes caught during the DEEPEND research cruises in the GOM from 2015 - 2018. While most stomiids have a mental (chin) barbel thought to serve as a mechanism for luring prey, C. sloani instead has a unique, forward-positioned dorsal fin that may be used similarly to lure prey (Richards, 2005). Chauliodus sloani feed on crustaceans and fishes mainly within the family Myctophidae (Sutton and Hopkins, 1996). Clarke (1974) indicated that C. sloani take numerous years to reach maturity, but lifespan is unknown.

Figure 5. An image of Chauliodus sloani caught in the Gulf of Mexico. Image take by Danté Fenolio. Stomias affinis (Figure 6) the scaley dragonfish, is a stomiiform fish within the family Stomiidae. Stomias affinis are numerically dominant predators and was one of the top three species of Stomiidae (8.3%) caught during the DEEPEND research cruises in the GOM in 2015 - 2018. Morphologically, Stomias affinis is a prototypical dragonfish, with a chin barbel for luring prey,

an elongated body form, and posteriorly placed median fins. These fishes are high-level predators that feed mainly on fishes of the family Myctophidae (lanternfishes) (Butler et al., 2001).

Figure 6. An image of Stomias affinis caught in the Gulf of Mexico. Image taken by Danté Fenolio. Omosudis lowii (Figure 7), the hammerjaw, is an aulopiform fish in the family Omosudidae. Omosudis lowii is a dominant predator in the Gulf of Mexico. These fish feed solely on squid and were the main squid predator caught during the DEEPEND and Pisces research cruises. Omosudis lowii have adapted to the deep-sea environment as functional simultaneous hermaphrodites, which allows them to function as both males and females (Smith and Atz, 1973; Sadovy de Mitcheson and Liu, 2008).

Figure 7. An image of Omosudis lowii from the Harvard Museum of Comparative Zoology. Lampanyctus lineatus (Formerly Nannobrachium lineatum, sensu Zahuranec, 2000) is a large-bodied myctophiform fish in the family Myctophidae. Myctophids made up over 8% of the fishes caught during DEEPEND cruises. Of these, L. lineatus is the numerically dominant deep- living (>600 m) species in the GOM. It feeds on copepods and euphausiids.

3. Results

3.1. Chauliodus sloani

3.1.1. Length-weight regressions A power model was the best fit for C. sloani, represented by the equation: W=aSLb. In this equation, W represents weight (g), SL represents standard length (mm), a is the scaling factor moving SLb as a increases are decreases, and b is the exponent or power that determines the functions rate of growth. For C. sloani, the value of b is close to three, which represents isometric growth and is normal in most fishes (Figure 8).

W=8.45·10-7 SL3.19

Figure 8. Length-weight regression of Chauliodus sloani.

3.1.2. Size-frequency distribution and vertical distribution In general, the MOCNESS caught more small-sized individuals of C. sloani, while the High-Speed Rope Trawl caught larger-sized and higher total quantities of C. sloani (Figure 9). Peaks in the size-frequency analyses (Figure 9) were treated separately in further analysis, representing juvenile (Figure 9a), intermediate (Figure 9b), and adult (Figure 9c) life stages.

Overall, C. sloani is an asynchronous vertical migrator in the GOM, with individuals at depth during the day and some of the population vertically migrating to the surface waters during the night while others remain at depth (Figure 10a). Juveniles ranging from 0 – 43 mm SL occurred mostly between the depths of 200 – 600 m during the day, with a small portion occurring between

0-200 m. At night, a portion of the juvenile population migrated into the upper 200 m while others remain at depth (Figure 10b). Intermediate C. sloani exhibited a similar vertical migration pattern as the juveniles except for the complete avoidance of the upper 200 m during the day. A large portion of the intermediate population migrated into the upper 200 at night (Figure 10c) Adult C. sloani occurred from 0 – 1500 m depth and undergo an asynchronous vertical migration pattern with the majority within 600 m – 1200 m (Figure 10d).

Figure 9. Length frequency of all Chauliodus sloani collected during DEEPEND and Pisces cruises. The red boxes represent: a.) juveniles, b.) intermediates, and c.) adult individuals.

Figure 10. The vertical distribution pattern of: a.) all Chauliodus sloani collected during DEEPEND cruises; b.) juveniles, c.) intermediate, and d.) adult size classes.

3.1.3. Length-increment regressions and lifespan For SL vs. increment count regression the best model was the logistic model which is 푎 represented by the equation: SL= . In this equation, SL is standard length (mm), 1+푒(푏−푖푛푐푟푒푚푒푛푡 푛표.)/푥 increment no. is the number of increments present, a is the upper asymptote of the standard length, b is the time (number of increments = days) at the inflection point, which can be positive or negative, thus increasing or decreasing increment no. as x increases. The rate when growth decreases with size is represented by x (Figure 11). Additionally, the logistic model also was the best model for the SL vs. dark major increment regression. This logistic model is represented by 푎 the equation: SL= , where SL is standard length (mm), dark major 1+푒(푏−푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠)/푥 increments represents the number of dark increments, a is the upper asymptote of the standard length, b is the time (number of dark major increments = years) at the inflection point which can be positive or negative thus increasing or decreasing dark major as x, the rate when growth decreases with size, increases (Figure 11).

187.93 183.08 푆퐿 = (6.51−푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠)ൗ 푆퐿 = (179.33−푖푛푐푟푒푚푒푛푡 푛표.) 1 + 푒 3.14 1 + 푒 ൗ56.64

Figure 11. Otolith increment regressions for Chauliodus sloani. 21

The ages of C. sloani by life stage (e.g. juvenile, intermediate, adult) were estimated using the two growth curve scenarios (number of increments = days, and number of dark major increments = years; Figure 11). If the number of increments is representative of days, then juvenile C. sloani are approximately 2 – 3 months old, intermediate stages are approximately 4 – 7 months, and C. sloani are adults in just less than a year. If dark major increments equal years, then juvenile C. sloani have a duration 0 – 1 year, intermediate are between 4 – 9 years old, and adult C. sloani have a lifespan of 17 years (Table 3).

Table 3. The age estimations of juvenile, intermediate, and adult Chauliodus sloani if the total number of increments represent days and dark major increments represent years.

N Standard length (mm) Number of Dark major increments increments Juvenile 72 23 - 43 78 - 92 0 - 1 Intermediate 42 47 - 123 121 - 238 4 - 9 Adult 35 128 - 199 235 - 337 9 - 17

3.1.4. External otolith morphology Chauliodus sloani have an oval shaped otolith with entire margins. The rostrum extends further than the antirostrum. The anterior end of the otoliths is double-peaked, whereas the posterior end is round. The sulcus is o-heterosulcoid with an ostial classified ostial opening. The ostium is funnel-like, and the caudal is round-oval and straight. Smaller individuals of C. sloani have otoliths with less pronounced rostrums and antirostrums (Figure 12).

Figure 12. Otoliths of Chauliods sloani at 100 mm SL, 80 mm SL, and 34 mm SL. Scale bars are set to 100 microns.

3.1.5. Otolith length and otolith width vs. SL regressions A Kendall’s tau correlation test determined that the left and right otoliths of C. sloani were not significantly different, thus the remaining analysis was performed on right otoliths (Table 4). The power model was the best model for otolith length and otolith width in relation to SL.

Table 4. Kendall’s tau correlation test on the left and right otoliths of Chauliodus sloani.

Chauliodus sloani Kendall’s Tau Correlation T p-value tau Otolith length 669 2.2·10-16 0.805668 Otolith Width 708 2.2·10-16 0.9109312

For otolith length - SL, the power model is represented by the equation: OL=aSLb. In this equation, OL represents otolith length (µm), SL represents standard length (mm), a is the scaling factor moving SLb as a increases are decreases, and b is the exponent or power that determines the functions rate of growth. For otolith width - SL, the power model is represented by the equation: OW=aSLb. In this equation, OW represents otolith width (µm), SL represents standard length (mm), a is the scaling factor moving SLb as a increases are decreases, and b is the exponent or power that determines the functions rate of growth (Figure 13).

OL = 3.60SL0.96 OW = 3.54SL0.93

Figure 13. Otolith length and width vs. SL regressions for Chauliodus sloani.

3.1.6. Chauliodus sloani internal otolith morphology The best model was the logistic model, which is represented by the equation: 푆퐿 = 푎 . In this equation, SL is standard length (mm), zones represents the number of zones, 1+푒(푏−푧표푛푒푠)/푥.

a is the upper asymptote of the number of zones, b is the standard length at the inflection point which can be positive or negative, thus increasing or decreasing zone as x increases (Figure 14).

14.64 푆퐿 =

1 + 푒(71.82−푧표푛푒푠)/32.76

No. of No. Zones

Figure 14. The number of zones in each otolith vs. SL regression for Chauliodus sloani. Non-parametric one-way ANOVA’s were used to examine zone width, increment width, and the number of increments between each zone to determine if there were any significant patterns. Chauliodus sloani otoliths have darkly etched increments that outlined the inner lightly etched increments and signified zones. A Kruskal-Wallis rank test indicated that zone width was significantly different between zones (p-value = 1.671e-08) (Figure 15a). Overall, the zone length of zones 1 and 2 were significantly different from zones 3 - 7 and zone 9. A Kruskal-Wallis rank test indicated that increment width was not significantly different between zones (p-value = 0.3754) (Figure 15b). Increment width was steady throughout the otoliths and averaged 1.2 microns in width. For the number of increments in each zone, a Kruskal-Wallis rank test indicated that increment number was significantly different between zones (p-value = 1.814e-06) (Figure 15c). Zone 1 and zone 2 were significantly different than zone 4. Additionally, zone 2 was significantly different than zone 3 in increment number.

Figure 15. Non-parametric one-way ANOVA Kruskal-Wallis rank displaying the significant differences in a.) zone length, b.) increment width, and c.) number of increments of each zone in the otoliths (d.) of Chauliodus sloani. Within each zone per C. sloani otolith, there was variation in the pattern in which zones were deposited, therefore the width of zones 4 – 17 were variable among individuals (Figure 16). Furthermore, there was little variation in increment width within each zone since the majority are within the scale of one to two microns. Within each zone, there was variation in the pattern at which the number of increments were deposited, therefore the number of increments were variable from zones 5 – 17.

Figure 16. Plots of the zone length, increment width, and the number of increments within each zone of individual otoliths of Chauliodus sloani.

3.2. Stomias affinis

3.2.1. Length-weight regression A power model was the best fit for S. affinis, represented by the equation: W=aSLb. In this equation, W represents weight (g), SL represents standard length (mm), a is the scaling factor moving SLb as a increases are decreases, and b is the exponent or power that determines the functions rate of growth. For S. affinis, the value of b is greater than three, which means this fish grows faster in weight than in length (Figure 17).

W = 7.21·10-9 SL4.13

Figure 17. Length-weight regression for Stomias affinis.

3.2.2. Size frequency of occurrence and vertical distribution Overall, the MOCNESS caught more small-sized individuals of S. affinis, with the majority being juveniles (Figure 18). Additionally, the MOCNESS caught some intermediate-sized S. affinis and a small number of adult individuals. Furthermore, the High-Speed Rope Trawl caught more large-sized individuals than the MOCNESS. Intermediate and adult S. affinis were the main targeted size ranges caught with the High-Speed Rope Trawl. In total, the High-Speed Rope Trawl caught more S. affinis individuals. As with C. sloani, the peaks in the size-frequency analysis were treated separately for further analysis, representing juvenile (Figure 18a), intermediate (Figure 18b), and adult (Figure 18c).

Stomias affinis is an asynchronous vertical migrator in the GOM, with individuals at depth during the day and some of the population vertically migrating to the surface waters at night while others remain at depth (Figure 19a). During the day, the juveniles ranging from 0 – 55 mm SL mostly occurred between the depths of 200 – 1000m, with a small portion occurring between 0 – 200 m. At night, a portion of the juvenile population migrated into the upper 200 m while others remain at depth (Figure 19b). Intermediate S. affinis ranging from 56 – 149 mm SL completely avoided the upper 200 m during the day and occurred between the depths 600 – 1000 m. Furthermore, at night a large portion of the intermediate population migrated into the upper 200

while others remain at depth (Figure 19c). For adult (150 – 220 mm SL) S. affinis, there were not enough individuals collected to plot their vertical migration pattern (Figure 19d).

Figure 18. Length frequency of all Stomias affinis collected during DEEPEND and Pisces cruises. The red boxes represent: a.) juveniles, b.) intermediates, and c.) adult individuals.

Figure 19. The vertical distribution pattern of: a.) all Stomias affinis collected during DEEPEND cruises; b.) juveniles, c.) intermediate, and d.) adult size classes.

3.2.3. Length-increment regressions and lifespan Gompertz model was the best fit for the SL vs. increment counts regression. The Gompertz

−푥(푖푛푐푟푒푚푒푛푡 푛표.) model is represented by a double exponential equation:푆퐿 = 푎푒−푏푒 , where SL is

standard length, increment no. represents the number of increments counted, a is the asymptote of standard length, b is an exponent setting the displacement of increment no (days). and x is an exponent determining the function's growth rate. Furthermore, For SL vs. dark major increments, the Gompertz model was also the best fit. The Gompertz model is represented by a double

−푥(푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠) exponential equation:푆퐿 = 푎푒−푏푒 , where SL is standard length, a is the asymptote of standard length, b is an exponent setting the displacement of the dark major increments (years), and x is an exponent determining the function growth rate (Figure 20).

−0.85(푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠.) −9.91∙10−1(푖푛푐푟푒푚푒푛푡 푛표.) −3.09푒 푆퐿 = 2.14 ∙ 102푒−4.18푒 푆퐿 = 213.72푒

Figure 20. Otolith increment regressions of Stomias affinis. The ages of S. affinis by life stage (e.g. juvenile, intermediate, adult) were estimated using the two growth curve scenarios: total number of increments = days, and number of dark major increments = years. If the number of increments is representative of days, then juvenile S. affinis are approximately 3 – 5 months old, intermediate stages are approximately 4 – 8 months, and S. affinis are adults at just about a year. If dark major increments equal years, then juvenile S. affinis have a duration 3 – 6 years, intermediate are between 6 – 11 years old, and adult S. affinis have a lifespan of 10 – 19 years (Table 5).

Table 5. The age estimations of juvenile, intermediate, and adult Stomias affinis if the number of increments represent days and dark major increments represent years.

N Standard length (mm) Number of Increment Dark Major Increments Juvenile 5 47 - 52 104 – 157 3 - 6 Intermediate 25 56 - 146 122 – 243 6 - 11 Adult 3 161 - 181 269 – 397 10 - 19

3.2.4. External Otolith morphology Stomias affinis have an oval-shaped otolith with entire margins. The anterior end of the otolith is peaked, and the posterior is round. The sulcus is homosulcoid with a medial ostium opening. The ostium and caudal are round-oval. The caudal is also straight (Figure 21).

Figure 21. Otoliths of Stomias affinis at 181 mm SL, 98 mm SL, and 47 mm SL. Scale bars are set to 100 microns.

3.2.5. Otolith length and otolith width-fish length regressions Pearson’s correlation test determined that the left and right otoliths of S. affinis were not significantly different, thus the remaining analysis was performed on right otoliths (Table 6). The power model was the best model for S. affinis otolith length and otolith width in relation to SL.

Table 6. Pearson’s correlation test on right and left otoliths of Stomias affinis.

Stomias affinis Pearson’s Correlation p-value df cor Otolith length 2.2·10-16 36 0.9469886 Otolith Width 2.2·1016 36 0.9337911

0.62 OL = 13.89SL0.68 OW = 15.70SL

Figure 22. Otolith length and width vs. standard length (mm) regressions for Stomias affinis.

3.2.6. Stomias affinis internal otolith morphology A power model was the best fit for S. affinis length vs. the number of zones. This model is represented by the equation: SL = aZonesb. where SL is standard length (mm), a is the scaling factor which moves the Zonesb as it increases or decreases, and b is an exponent determining the function’s rate of growth (Figure 23)

SL = 0.63 Zones0.62

Zones

No. of No.

Figure 23. The number of zones in each otolith vs. SL regression for Stomias affinis. Non-parametric one-way ANOVA’s were used to examine zone width, increment width, and the number of increments between each zone to determine if there were any significant patterns. Stomias affinis otoliths have darkly etched increments that surround lightly etched increments and signified zones. A Kruskal-Wallis rank test indicated that zone width was significantly different between zones (p-value = 8.392e-10 (Figure 24a). Overall, the zone length of zone 1 was significantly different from zones 2 - 12. Furthermore, A Kruskal-Wallis rank test indicated that increment width between zones was significantly different (p-value = 1.247e-05), where zone 1 was significantly different from zones 6 - 11 (Figure 24b). For the number of increments in each zone, a Kruskal-Wallis rank test indicated that increment number was significantly different between zones (p-value = 0.001325) (Figure 24c). Increment number in zone 1 was significantly different than zone 2 and 3.

Figure 24. Non-parametric one-way ANOVA Kruskal-Wallis rank displaying the significant differences in a.) zone length, b.) increment width, and c.) number of increments of each zone in the otoliths (d.) of Stomias affinis. Within the zones of S. affinis otoliths, there was variation in the pattern in which zones were deposited, therefore the width of the zones from zones 3 to 13 was variable among S. affinis -individuals (Figure 25). However, there was little variation in increment width within each zone since the majority are within the scale of one to three microns. Within each zone, there was variation in the pattern at which the number of increments were deposited, resulting in the number of increments being variable from zones 3 – 13.

Figure 25. Plots of zone length, increment width, and the number of increments within each zone of individual otoliths of Stomias affinis.

3.3. Omosudis lowii

3.3.1. Length-weight regression A power model was the best fit for O. lowii, represented by the equation: W=aSLb. In this equation, W represents weight (g), SL represents standard length (mm), a is the scaling factor moving SLb as a increases are decreases, and b is the exponent or power that determines the functions rate of growth. For O. lowii, the value of b is close to three, which represents isometric growth and is normal in most fishes (Figure 26).

W = 3.54·10-6 SL3.03

Figure 26. Length – weight regression of Omosudis lowii.

3.3.2. Size frequency of occurrence and vertical distribution The MOCNESS caught large quantities of small O. lowii individuals, while the High-Speed Rope Trawl caught mainly large individuals. Additionally, the High-Speed Rope Trawl caught a larger quantity of O. lowii. Both gear types were needed in order to analyze the full-size range of O. lowii (Figure 27).

Peaks in the size-frequency analysis were treated separately in further analysis, representing juvenile (Figure 27a), intermediate (Figure 27b), and adult (Figure 27c). For O. lowii, juveniles were from 0 – 35 mm SL, intermediates were from 36 – 149 mm SL, and adults were from 150 – 290 mm SL. Omosudis lowii is a non-vertical migrator in the GOM, thus all of the population remains at depth during day and night (Figure 28).

Figure 27. Length frequency of all Omosudis lowii collected during DEEPEND and Pisces cruises. The red boxes represent: a.) juveniles, b.) intermediates, and c.) adult individuals.

Figure 28. Vertical distribution of Omosudis lowii.

3.3.3. Length-increment regressions and lifespan A Gompertz model was the best fit for O. lowii SL vs. increment regression. The Gompertz

−푥(푖푛푐푟푒푚푒푛푡 푛표.) model is represented by a double exponential equation:푆퐿 = 푎푒−푏푒 , where SL is standard length, a is the asymptote of standard length, b is an exponent setting the displacement of the number of increments (days), increment no. is the number of increments counted, and x is an exponent determining the functions growth rate. For SL vs. dark major increments, the Gompertz model was also the best fit. The Gompertz model is represented by a double exponential

−푥(푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠) equation:푆퐿 = 푎푒−푏푒 , where SL is standard length, a is the asymptote of standard length, b is an exponent setting the displacement of the dark major increments (years), and x is an exponent determining the function’s growth rate (Figure 29).

−3.54푒−0.92(푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠) −9.52∙10−1(푖푛푐푟푒푚푒푛푡 푛표.) 푆퐿 = 262.93푒 푆퐿 = 2.59 · 102푒−4.65푒

Figure 29. Otolith increment regressions of Omosudis lowii. The ages of O. lowii by life stage (e.g. juvenile, intermediate, Adult) were estimated using the two growth curve scenarios: number of increments = days, and number of dark major increments = years (Figure 29). If the number of increments is representative of days, then juvenile O. lowii are approximately 3 – 5 months old, intermediate stages are approximately 7 – 14 months old, and O. lowii are adults at just over a year. If dark major increments equal years, then juvenile O. lowii have a duration 3 – 8 years, intermediates are between 10 – 20 years old, and adult O. lowii have a lifespan of 22 years (Table 7).

Table 7. The age estimations of juvenile, intermediate, and adult Omosudis lowii if the number of increments represents days and dark major increments represent years.

N Standard length (mm) Number of Increment Dark Major Increments Juvenile 18 12 - 32 116 - 173 3 - 8 Intermediate 25 57 - 139 211 - 424 10 - 20 Adult 1 164 409 22

3.3.4. Omosudis lowii external otolith morphology Omosudis lowii have an otolith that is square-shaped dorsally and discoidal ventrally with entire margins. The anterior and posterior of the otoliths are rounded. The rostrum and antirostrum of O. lowii otoliths are not distinct or pronounced. The sulcus is homosulcoid with a pseudo-ostial ostial opening. Within the sulcus margins, the ostium is tubular, and the caudal is round-oval and straight (Figure 30).

Figure 30. Otoliths of Omosudis lowii at 164 mm SL, 84 mm SL, and 24 mm SL. Scale bar is at 100 microns.

3.3.5. Otolith length and otolith width-fish length regressions A Kruskal-Wallis correlation test determined that the left and right otoliths of O. lowii were not significantly different, thus the remaining analysis was performed on right otoliths (Table 8). The power model was the best fit for O. lowii otolith length and otolith width in relation to SL.

Table 8. Kendall’s tau correlation test on left and right otoliths of Omosudis lowii.

Omosudis lowii Kendall’s Tau Correlation p-value tau Otolith length 2.2·10-16 0.9550279 Otolith Width 4.4·10-16 0.9414141

OL = 55.10SL0.49 OW = 45.10SL0.54

Figure 31. Otolith length and width vs. standard length regressions of Omosudis lowii.

3.3.6. Omosudis lowii internal otolith morphology A power model was the best fit for O. lowii length vs. the number of zones. This model is represented by the equation: SL = aZonesb. where SL is standard length (mm), Zones represent the number of zones counted, a is the scaling factor which moves the Zonesb as it increases or decreases, and b is an exponent determining the function’s rate of growth (Figure 32).

SL = 1.25Zones0.58

No. of No. Zones

Figure 32. The number of zones in each otolith vs. SL regression for Omosudis lowii. Non-parametric one-way ANOVA’s were used to examine zone width, increment width, and the number of increments between each zone to determine if there were any significant differences between zones. Omosudis lowii otoliths have clusters of darkly etched increments that signified the outer boundaries of zones. A Kruskal-Wallis rank test indicated that zone width was significantly different between zones (p-value = 2.2e-16) (Figure 33a). Overall, the zone length of zone 1 was significantly different from zones 2 – 21. Zone length in zone 2 was significantly different than zone 1, zones 5 - 14, and zone 16, and zone length in zone 3 was significantly different than zone 1, and zones 7 - 12. Furthermore, A Kruskal-Wallis rank test indicated that increment width between zones was significantly different (p-value = 1.759e-07), where zone 1 was significantly different from zones 1 – 7, zone 13, and zone 18 (Figure 33b). For the number of increments in each zone, a Kruskal-Wallis rank test indicated that the increment number was significantly different between zones (p-value = 2.2e-16) (Figure 33c). Increment number in zone 1 was significantly different from zones 2 – 4, zone 15, and zone 18 – 19. Increment number in zone 2 was significantly different than zone 1, zone 4 – 14, and increment number in zone 3 was significantly different than zones 5, 8, 10, and 13.

Figure 33. Non-parametric one-way ANOVA Kruskal-Wallis rank displaying the significant differences in a.) zone length, b.) increment width, and c.) number of increments of each zone in the otoliths (d.) of Omosudis lowii. Within each zone of O. lowii otoliths, there were variations in the pattern in which zones and the number of increments were deposited, therefore the width of the zones and the number of increments from zones 3 - 22 were variable among O. lowii individuals (Figure 34). Likewise, variation was present in increment width within each zone although, the scale qas between 0.25 to one micron. Within each zone per otolith, there was variation in the pattern at which the number of increments were deposited, resulting in the number of increments being variable from zones 3 – 22.

Figure 34. Plots of zone length, increment width, and the number of increments within each zone of individual otoliths of Omosudis lowii.

3.4. Lampanyctus lineatus

3.4.1. Length-weight regression A power model was the best fit for L. lineatus, represented by the equation: W=aSLb. In this equation, W represents weight (g), SL represents standard length (mm), a is the scaling factor moving SLb as a increases are decreases, and b is the exponent or power that determines the functions rate of growth. For L. lineatus, the value of b is close to three, which represents isometric growth and is normal in most fishes (Figure 35).

W = 1.41·10-6 SL3.28

Figure 35. Length-weight regression of Lampanyctus lineatus.

3.4.2. Size frequency of occurrence and vertical distribution Overall, the MOCNESS caught larger quantities of small L. lineatus, with the majority being juveniles and intermediate-sized individuals (Figure 36). Additionally, the MOCNESS caught a small number of adult individuals. The High-Speed Rope Trawl caught larger quantities of intermediate and adult individuals than the MOCNESS. The High-Speed Rope Trawl caught larger quantities of L. lineatus. Peaks in the size-frequency analysis were treated separately for further analysis, representing juvenile (Figure 36a), intermediate (Figure 36b), and adult (Figure 36c).

Lampanyctus lineatus is an asynchronous vertical migrator in the GOM, with individuals at depth during the day and some of the population vertically migrating into the surface waters at night while others remain at depth (Figure 37a). Juveniles ranging from 0 – 56 mm SL occurred mostly between the depths of 0 – 1000 m, avoiding the upper 200 m during the day. However, at night a portion of the juvenile population migrated into the upper 200 m while others remained at depth (Figure 37b). Intermediate (57 – 92 mm SL) L. lineatus exhibited a similar vertical migration pattern as the juveniles with a deeper depth range, spanning from 0 – 1200 m (Figure 37c). Adult L. lineatus ranging from 93 – 180 mm SL occurred between the depths of 200 – 1500 m deep, and tended to remain at depth (Figure 37d).

Figure 36. Length frequency of all Lampanyctus lineatus collected during DEEPEND and Pisces cruises. The red boxes represent: a.) juveniles, b.) intermediates, and c.) adult individuals.

Figure 37. The vertical distribution pattern of: a.) all Lampanyctus lineatus collected during DEEPEND cruises; b.) juveniles, c.) intermediate, and d.) adult size classes.

3.4.3. Length-increment regressions and lifespan For the number of increments vs. SL, a Weibull model was the best fit for L. lineatus. The

푖푛푐푟푒푚푒푛푡 푛표. 훿 Weibull model is represented by the equation: 푆퐿 = 푎 − 푏푒−푒 푥 , where SL is standard length, increment no. is the number of increments present, a is the upper asymptote of standard length, b is the rate inflection point which increases or decreases increment no. (increment no. =

days) and x is raised to the power of 훿 (Figure 38). An exponential model was the best fit for dark major increments vs. SL of L. lineatus. This model is represented by the equation: 푆퐿 = 푎푒푏(푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠), where SL is standard length, a is the scale of the standard length, and b is an exponent that increases or decreases dark major increments (dark major increments = years) (Figure 38).

푖푛푐푟푒푚푒푛푡 푛표. 6.92 푆퐿 = 75.24 − 44.11푒−푒 −39.31 푆퐿 = 27.23푒0.31(푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠)

Figure 38. Otolith increment regressions of Lampanyctus lineatus. The ages of L. lineatus by life stage (e.g. juvenile, intermediate, adult) were estimated using the two growth curve scenarios: number of increments = days, and number of dark major increments = years (Figure 38). If the number of increments is representative of days, then juvenile L. lineatus are approximately 3 – 9 months old and intermediate are approximately 9 – 12 months. If dark major increments equal years, then juvenile L. lineatus have a duration of 0 - 1 year and intermediate have an agespan of 2 – 3 years (Table 9).

Table 9. The age estimations of juvenile, intermediate, and adult Lampanyctus lineatus if the number of increments represents days and dark major increments represent years.

N Standard length (mm) Number of Increment Dark Major Increments Juvenile 13 30 - 50 119 - 266 1 Intermediate 9 58 - 84 291 - 360 2 - 3 Adult 0 NA NA NA

3.4.4. Lampanyctus lineatus external otolith morphology Lampanyctus lineatus have square and tall otoliths with entire margins. The rostrum and antirostrum of L. lineatus otoliths are small, rounded, and both protrude about the same distance on the anterior end of the otolith. The anterior and posterior are both flattened. The sulcus is archaesulcoid with an ostial opening that is ostial. Within the sulcus margins, the ostium and caudal are tubular and straight (Figure 39).

Figure 39. Otoliths of Lampanyctus lineatus at 71 mm SL, 57 mm SL, and 32 mm SL. Scale bars are set at 100 microns.

3.4.5. Otolith length and otolith width-fish length regressions A Kruskal-Wallis correlation test determined that the left and right otoliths of L. lineatus were not significantly different, thus the remaining analysis was performed on right otoliths (Table 10). A Gompertz model was the best fit for L. lineatus otolith length and otolith width in relation to SL.

Table 10. Kendall’s tau correlation of left and right otoliths of Lampanyctus lineatus.

Lampanyctus lineatus Kendall’s Tau Correlation p-value Tau Otolith length 2.2·10-16 0.8412698 Otolith Width 2.2·10-16 0.847619

For otolith length-SL, the Gompertz model is represented by a double exponential

−푥(푆퐿) equation:푂퐿 = 푎푒−푏푒 , where OL is otolith length, SL is standard length, a is the asymptote of otolith length, b is an exponent setting the displacement of SL and x is an exponent determining the function’s growth rate. For otolith width-SL, the Gompertz model is represented by a double

−푥(푆퐿) exponential equation:푂푊 = 푎푒−푏푒 , where OW is otolith width, SL is standard length, a is

the asymptote of otolith width, b is an exponent setting the displacement of SL, and x is an exponent determining the functions growth rate (Figure 40).

−0.96(푆퐿) −2.39푒 −0.96(푆퐿) 푂퐿 = 685.24푒 푂푊 = 849.43푒−2.39푒

Figure 40. Otolith length and width vs. standard length (mm) regressions for Lampanyctus lineatus

3.4.6. Lampanyctus lineatus internal otolith morphology For the number of zones vs. SL, a Weibull was the best fit for L. lineatus. The Weibull

푁표.표푓 푍표푛푒푠. 훿 model is represented by the equation: 푆퐿 = 푎 − 푏푒−푒 푥 , where SL is standard length, No. of Zones is the number of zones present, a is the upper asymptote of SL, b is the rate inflection point which increases or decreases the No. of Zones and x is raised to the power of 훿 (Figure 41).

푁표.표푓 푍표푛푒푠. 27.26

푆퐿 = 4.83 − 1.74푒−푒 −110.66

No. of No. Zones

Figure 41. Number of zones – standard length (mm) regression of Lampanyctus lineatus.

Non-parametric one-way ANOVA’s were used to examine zone width, increment width, and the number of increments between each zone to determine if there were any significant differences between zones. Lampanyctus lineatus otoliths have clusters of dark etched increments that signified the outer boundaries of zones. A Kruskal-Wallis rank test indicated that zone width was significantly different between zones (p-value = 0.0003691), where zone 1 was significantly different from zone 2 and zone 2 was significantly different from zone 3 (Figure 42a). Likewise, Kruskal-Wallis rank test indicated that increment width between zones was significantly different (p-value = 3.689e-05), where zone 1 was significantly different from zone 2, and zone 2 was significantly different from zones 4 and 5 (Figure 42b). For the number of increments in each zone, a Kruskal-Wallis rank test indicated that increment number was not significantly different between zones (p-value = 0.194) (Figure 42c).

Figure 42. Non-parametric one-way ANOVA Kruskal-Wallis rank displaying the significant differences in a.) zone length, b.) increment width, and c.) number of increments of each zone in the otoliths (d.) of Lampanyctus lineatus. Within each zone of L. lineatus otoliths per otolith, there was little variation in the pattern in which zones were deposited (Figure 43). Likewise, variation was present in increment width 48

among all zones. Within each zone per otolith, there was variation in the pattern at which the number of increments was deposited, resulting in the number of increments being variable from zones 4 – 5.

Figure 43. Plots of zone length, increment width, and the number of increments within each zone of individual otoliths of Lampanyctus lineatus.

3.5. Summary of all fishes For length-weight regressions, a power model was the best fit for C. sloani, S. affinis, L. lineatus, and O. lowii (Table 11).

Table 11. Length-weight regressions of Chauliodus sloani, Stomias affinis, Omosudis lowii, and Lampanyctus lineatus.

Length-Weight Regressions Species Growth equation Model

Chauliodus sloani W=8.45·10-7 SL3.19 Power

Stomias affinis W = 7.21·10-9 SL4.13 Power Omosudis lowii W = 3.54·10-6 SL3.03 Power

Lampanyctus lineatus W = 1.41·10-6 SL3.28 Power

For length-age regression, a logistic model was the best fit for C. sloani. A Gompertz model was the best fit for S. affinis and O. lowii (Table 12). As for L. lineatus, a Weibull model was the best fit for length vs. increment number, while an exponential model was the best fit when considering length vs. dark major increments. For otolith length-SL and otolith width-SL, a power model was the best fit for C. sloani, S. affinis, and O. lowii, while a Gompertz model was the best fit for L. lineatus.

Table 12. Length-increment regressions and Otolith-SL regressions of Chauliodus sloani, Stomias affinis, Omosudis lowii, and Lampanyctus lineatus.

Length-Age Regressions Species Growth equation (dark major Growth equation (increment no.) Model increments) 푆퐿 183.08 Chauliodus sloani 푆퐿 = 187.93 Logistic (179.33−푖푛푐푟푒푚푒푛푡 푛표.)ൗ 1 + 푒 57.64 = (6.51−푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠) 1 + 푒 ൗ3.14 푆퐿 푆퐿 Stomias affinis Gompertz −9.91∙10−1(푖푛푐푟푒푚푒푛푡 푛표.) −0.85(푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠.) = 2.14 ∙ 102푒−4.18푒 = 213.72푒−3.09푒 푆퐿 푆퐿 Omosudis lowii Gompertz −9.52∙10−1(푖푛푐푟푒푚푒푛푡 푛표.) −0.92(푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠) = 2.59 · 102푒−4.65푒 = 262.93푒−3.54푒 Lampanyctus 푆퐿 Weibull / 0.31(푑푎푟푘 푚푎푗표푟 푖푛푐푟푒푚푒푛푡푠) 푖푛푐푟푒푚푒푛푡 푛표. 6.92 푆퐿 = 27.23푒 lineatus = 75.24 − 44.11푒−푒 −39.31 Exponential Otolith-length Regressions Otolith Length-SL Otolith Width-SL Model

Chauliodus sloani OL = 3.60·SL0.96 OW = 3.54·SL0.93 Power

Stomias affinis OL = 13.89·SL0.68 OW = 15.70·SL0.62 Power

Omosudis lowii OL = 55.10SL0.49 OW = 45.10SL0.54 Power

Lampanyctus −0.96(푆퐿) −0.96(푆퐿) 푂퐿 = 685.24푒−2.39푒 푂푊 = 849.43푒−2.39푒 Gompertz

lineatus

ShapeR analysis indicted that C. sloani and S. affinis have nearly identical otoliths, while L. lineatus cluster separately. Omosudis lowii otoliths are variable in shape (Figure 44).

Figure 44. ShapeR analysis of the otoliths of Chauliodus sloani, Stomias affinis, Omosudis lowii, and Lampanyctus lineatus.

4. Discussion

4.1. The importance of aging fishes and the rationale of this study Aging fishes is a vital element for the assessment of fish populations and their resource management (Cailliet et al., 2001). Age and growth information is essential for estimating fish production (Mayank et al., 2015). Deep-pelagic fishes are difficult to age due to the depth at which they are caught (i.e., sample sizes are necessarily limited in time and space due to sampling logistics), the small size of their otoliths, and the lack of literature associated with these fishes. In general, the deep-pelagic is the most data-deficient zone in the ocean (St. John et al., 2016).

The majority of deep-pelagic fish species conduct diel vertical migrations on a variety of scales (Brierley, 2014). Diel vertical migration is unique to pelagic, oceanic systems, with only a small number of freshwater fishes (e.g., salmon) and those that live in deep, cold lakes undergoing a similar migration pattern (Mehner, 2012). Three of the four fish species in this study undertake an asynchronous form of diel vertical migration, with some of the population vertically migrating

upwards at night while others remain at depth (Brierley, 2014). Similar to fishes migrating from estuaries to the open ocean, changes in vertical migration pattern with ontogeny are landmarks that may be reflected as changes in growth rate (Thorpe, 1994). Thus, in his study, different size classes within each species identified through length frequency analysis were treated as separate “ecological units” with respect to vertical migration. This method of treating size classes as separate units was utilized throughout the study and proved useful since these size classes represented differing growth trajectories.

4.2. Vertical migration patterns Chauliodus sloani, S. affinis, and L. lineatus displayed an asynchronous vertical migration pattern, while O. lowii apparently does not vertically migrate at any life stage. Chauliodus sloani exhibited asynchronous vertical migration throughout its life (i.e. juvenile, intermediate, and adult). However, as C. sloani grow larger, they occupy deeper depths, with juveniles and intermediate individuals found mainly between 0 – 600 m and adults found throughout a 0 – 1500 m range. Similarly, S. affinis exhibited this same asynchronous migration pattern for juvenile and intermediate fish. Adult vertical migration could not be examined in detail due to the lack of data for that size range. These findings agree with past studies reporting that stomiids such as C. sloani and S. affinis undertake an asynchronous vertical migration (Sutton and Hopkins, 1996; Kenaley, 2008). Lastly, L. lineatus exhibited asynchronous vertical migration at all life stages. As L. lineatus grow larger, less of the population vertically migrates into shallow waters and instead prefers deeper depths. Other myctophid species, such as Benthosema glaciale Ceratoscopelus maderensis, Hygophum hygomii, Lobianchia dofleini, Notoscopelus elongatus, Symbolophorus veranyi, Lampanyctus pusillus, Hygophum benoiti, and Myctophum punctatum have been shown to exhibit similar asynchronous vertical migration patterns (Dypvik et al., 2012; Olivar et al., 2012).

It is generally accepted that diel vertical migration is mostly driven by food and light, with fishes swimming into the surface at night within the protection of darkness to feed and avoid predators followed by their return to depth during the day. Asynchronous vertical migration then may represent portions of populations feeding on different days, perhaps due to it taking several days to digest larger meals (Pearre, 2003) or perhaps reduced metabolism of asynchronous migrators reducing the need to feed daily. Integrating data presented in this study with what is known about these species’ feeding and metabolism, it would be logical that the former hypothesis

(large meals) applies to C. sloani, S. affinis, while the latter (reduced metabolism) holds for L. lineatus given that this species is elongated, poorly muscled, and has small eyes compared to synchronously migrating myctophids. Omosudis lowii does not vertically migrate and has adapted to life at depth, this apparently striking a successful balance between feeding needs and predation risk. It is possible that this species successfully feeds at depth during daytime, thus releasing it from the need to vertically migrate to acquire food.

4.3. Growth patterns Chauliodus sloani exhibits logistic growth when standard length is compared to the number of total otolith increments (equaling days) and the number of dark major increments (equaling years), while S. affinis and O. lowii growth demonstrate a Gompertz growth trajectory. These growth curves indicate a slower initial growth, followed by a rapid increase in growth, then a decrease in growth rate in adulthood. Mesopelagic fishes such as Myctophum affine and Notoscopelus replendens (both Myctophidae) also display growth patterns that best fit a Gompertz model (Hayashi at al., 2001; Sarmiento-Lezcano et al., 2019). The majority of previous studies on myctophids have strictly applied a Von Bertalanffy function to growth; however, it has been noted that additional models may better represent growth of these fishes (Sarmiento-Lezcano et al., 2019). Lampanyctus lineatus standard length in relation to the number of increments was best represented by a Weibull model pattern, while exponential growth was revealed when examining standard length in relation to the number of dark major increments. The Weibull distribution indicated a longer period of slow growth earlier on, followed by a rapid increase in growth, which then decreased as the fish gets larger. Exponential growth suggests an increase in growth rate throughout life.

Two strategies have been proposed to explain slow vs. fast growth rates in larval fishes. One larval strategy is to put more effort and energy into growth early on in order to reach a large enough size to outgrow predation and increase survival (Stearns and Koella, 1986). Another strategy is to undergo slower growth during the larval stage in order to avoid increased predation at intermediate size classes. In the latter strategy, larvae delay the increased activity it takes to acquire prey in order to reduce activity-induced predation (Biro et al., 2006).

4.4. Age at maturity Marks et al. (2020) reported that female C. sloani reaches sexual maturity at ~155 mm SL. This size is within the adult size range (128 – 199 mm SL) classification of this study. Thus, if total increments represent days then C. sloani reach maturity between approximately 7 - 11 months old. If dark major increments equal years, then C. sloani reach maturity around 9 – 17 years old. Although the exact size at maturity is unknown for S. affinis, it is possible that S. affinis reaches maturity (161 – 181 mm SL) at a size similar to C. sloani. If so, then S. affinis would reach sexual maturity at the age of approximately 8 months to just over a year assuming that total increments equal days. If dark major increments equal years, then S. affinis reaches sexual maturity at 10 – 19 years. Myctophid fishes are known to mature at smaller sizes and ages than stomiids, befitting the fact that myctophids are the primary prey of most stomiids. Gartner (1993) indicates that generally myctophids reach maturity early on (Gartner, 1993). If so, then sexual maturity would be reached by L. lineatus during the juvenile (30 – 50 mm SL) stage. If increments represent days, then L. lineatus reaches sexual maturity at approximately 3 – 8 months. Likewise, if dark major increments equal years then L. lineatus reaches sexual maturity at 1 year.

4.5. Longevity Two scenarios were presented for longevity. If increments represent days, then C. sloani and L. lineatus have a life span of just under a year. Likewise, O. lowii and S. affinis would have a life span of approximately just over a year. This scenario would indicate that these fishes are short-lived, and so would be able to replace themselves quickly. This scenario seems unlikely given that these fishes are higher–level predators that consume myctophids, squid, and large euphausiids. GOM myctophids are thought to have a lifespan between 1 – 3 years (Gartner, 1993), and as a rule, predators have much longer generation times than their prey (Bax, 1998; Yamamichi and Miner, 2015). If dark major increments represent years, then these fishes are much older. Lampanyctus lineatus would have a lifespan of over 3 years, since only intermediate-sized individuals were examined in this study. Adult individuals, not available for this study, would likely be much older. Chauliodus sloani would have a lifespan of 17 years, which seems plausible since it has been reported that these fishes take several years to reach sexual maturity off Hawaii (Clarke, 1974), another low-latitude, oligotrophic ecosystem. Stomias affinis would have a lifespan of 19 years and Omosudis lowii a lifespan of 22 years. Currently, 55

there is a lack of additional age estimations for these fishes. If these fishes do have older lifespans and become sexually mature (section 4.4.) later in life, then these conditions indicate that these fishes would have a slower recovery rate when impacted by anthropogenic disasters or threats (Hughes et al., 2005). This would result in long generation times for populations to recover from population-level perturbation.

5. Conclusions Currently, the majority of ecosystem models cannot include or represent deep-pelagic fishes properly due to the basic lack of life-history information on these fishes. This study provides life history information, such as vertical migration patterns of different size classes, and age parameter estimations, such as age at sexual maturity and lifespan, that are necessities in building whole-ecosystem models. The most likely scenario revealed by this study is that the higher-level predators and deep-living myctophid species investigated here have relatively long generation times, and thus have low resilience to population level-perturbation.

6. References Ainsworth, C. H., M. J. Schirripa, and H. N. Morzaria-Luna. (2015). “An Atlantis ecosystem model for the Gulf of Mexico supporting integrated ecosystem assessment.” Bax, N. J. (1998). “The significance and prediction of predation in marine fisheries.” ICES Journal of marine Science 55(6), pp. 997-1030.

Biro, P. A., M. V. Abrahams, J. R. Post, and E. A. Parkinson. (2006). “Behavioural trade‐offs between growth and mortality explain evolution of submaximal growth rates.” Journal of animal ecology, 75(5), pp. 1165-1171. Brierley, A. S., (2014). “Diel vertical migration.” Current Biology, 24(22), pp. R1074-R1076. Butler, M., Bollens, S. M., Burkhalter, B., Madin, L. P., and Horgan, E. (2001). “Mesopelagic fishes of the Arabian Sea: distribution, abundance and diet of Chauliodus pammelas, Chauliodus sloani, Stomias affinis, and Stomias nebulosus.” Deep Sea Research Part II: Topical Studies in Oceanography, 48(6), 369-1383. doi:https://doi.org/10.1016/S0967- 0645(00)00143-0

Cailliet, G. M., A. H. Andrews, E. J. Burton, D. L. Watters, D. E. Kline, and L. A. Ferry- Graham. (2001)."Age determination and validation studies of marine fishes: do deep- dwellers live longer?." Experimental gerontology, 36(4-6), pp. 739-764.

Camilli, R., Reddy, C. M., Yoerger, D. R., Van Mooy, B. A., Jakuba, M. V., Kinsey, J. C., Mclntyre, A. P., Sylva, S. P., and Maloney, J. V. (2010). Tracking hydrocarbon plume transport and biodegradation at Deepwater Horizon. Science, 330(6001), pp. 201-204.

Campana, S. E. (1992). “Measurements and interpretation of the microstructure of fish otoliths.” In D. K. Stevenson and S. E. Campana [ed.] Otolith microstructure examination and analysis. Canadian Special Publication of Fisheries and Aquatic Sciences, 117, pp. 59-71 Campana, S. E. (2004). “Photographic atlas of fish otoliths of the northwest Atlantic Ocean.” Canadian Special Publication of Fisheries and Aquatic Sciences, 133: NRC Research Press. Campana, S. E. (2016, January 15). Otolith microstructure preparation. Retrieved from http://uni.hi.is/scampana/otoliths/methods/otolith-microstructure-preparation/ Campana, S. E. and C. M. Jones. (1992). “Analysis of otolith microstructure data.” In D. K. Steavenson and S. E. Campana [ed.] Otolith microstructure examination and analysis. Canadian Special Publication of Fisheries and Aquatic Sciences, 117. Campana, S. E. and J. D. Neilson. (1985). “Microstructure of fish otoliths.” Canadian Journal of Fisheries and Aquatic Sciences, 42(5), pp. 1014-1032. doi:10.1139/f85-127 Childress, J. J., S. M. Taylor, G. M. Cailliet, and M. H. Price. (1980). “Patterns of growth, energy utilization and reproduction in some meso- and bathypelagic fishes off southern California.” Marine Biology, 61(1), pp. 27-40. doi:10.1007/bf00410339

Clarke, T. A. (1974). “Some aspects of the ecology of stomiatoid fishes in the Pacific Ocean near Hawaii.” fishery bulletin, 72(2). Clarke, T. A. (1978). “Diel feeding patterns of 16 species of mesopelagic fishes from Hawaiian waters.” Fishery bulletin, 76(3), pp. 495-513. Davison, P. C., J. A. Koslow, and R. J. Kloser. (2015). “Acoustic biomass estimation of mesopelagic fish: backscattering from individuals, populations, and communities.” ICES Journal of Marine Science, 72(5), pp. 1413-1424. doi:10.1093/icesjms/fsv023 Deng, X., H. J. Wagner, and A. N. Popper. (2013). “Interspecific variations of inner ear structure in the deep-sea fish family Melamphaidae.” Anatomical Record (Hoboken), 296(7), pp. 1064-1082. doi:10.1002/ar.22703 Dypvik, E., A. Røstad, and S. Kaartvedt. (2012). “Seasonal variations in vertical migration of glacier lanternfish, Benthosema glaciale.” Marine biology, 159(8), pp. 1673-168 Gartner, J. V. (1991a). “Life histories of three species of lanternfishes (Pisces: Myctophidae) from the eastern Gulf of Mexico I. Morphological and microstructural analysis of sagittal otoliths.” Marine Biology (111), pp. 11-20. Gartner, J. V. (1991b). “Life histories of three species of lanternfishes (Pisces: Myctophidae) from the eastern Gulf of Mexico II. Age and growth patterns.” Marine Biology (111), pp. 11-20. Gartner, J. V. (1993). "Patterns of reproduction in the dominant lanternfish species (Pisces: Myctophidae) of the eastern Gulf of Mexico, with a review of reproduction among tropical-subtropical Myctophidae." Bulletin of Marine Science, 52(2), pp. 721-750. Gimenez, J., A. Manjabacas, V. M. Tuset, and A. Lombarte. (2016). “Relationships between otolith and fish size from Mediterranean and north-eastern Atlantic species to be used in predator-prey studies.” Journal of Fish Biology, 89(4), 2195-2202. doi:10.1111/jfb.13115 Giragosov, V. E. (2000). “Age and growth of the lanternfish Ceratoscopelus warmingii (Myctophydae Myctophidae).” Ekologiya Morya, 53, pp. 64-67.

Gjosaeter, J. and K. Kawaguchi. (1980). “A review of the world resources of mesopelagic fish.” FAO Fisheries Technical Papers, 193(151). Greely, T. M., J. V. Gartner, and J. J. Torres. (1999). “Age and growth of Electrona antarctica (Pisces: Myctophidae), the dominant mesopelagic fish of the Southern Ocean.” Marine Biology, 133(1), pp. 145-158. Hayashi, A., K. Kawaguchi, H. Watanabe, and M. Ishida. (2001). “Daily growth increment formation and its lunar periodicity in otoliths of the myctophid fish Myctophum asperum (Pisces: Myctophidae).” Fisheries Science, 67(5), pp. 811-817. doi:10.1046/j.1444- 2906.2001.00327.x Hayashi, A., H. Watanabe, M. Ishida, and K. Kawaguchi. (2001). “Growth of Myctophum asperum (Pisces: Myctophidae) in the Kuroshio and transitional waters.” Fisheries Science, 67(5), pp. 983-984. doi:10.1046/j.1444-2906.2001.00350.x

Howell, W. H. and W. H. Krueger. (1987). “Family Sternoptychidae, marine hatchetfishes and related species.” Smithsonian contributions to zoology, 452, pp. 32-50 Hughes, C. T., C. E. Lowie, and J. M. Haynes. (2005). "Age, growth, relative abundance, and scuba capture of a new or recovering spawning population of lake sturgeon in the lower Niagara River, New York." North American Journal of Fisheries Management, 25(4), pp. 1263-1272. Irigoien, X., T. A. Klevjer, A. Rostad, U. Martinez, G. Boyra, J. L. Acuna, A. Bodde, F. Echevarria, J. I. Gonzalez-Gordillo, S. Hernandez-Leon, S. Agusti, D. L. Aksnes, C. M. Duarte, and S. Kaartvedt. (2014). “Large mesopelagic fishes biomass and trophic efficiency in the open ocean.” National Communications, 5, pp. 3271. doi:10.1038/ncomms4271 Karnella, C. (1987). “Family Myctophidae, Lanternfishes.” Smithsonian contributions to zoology, 452, pp. 51-168 Kawaguchi, K. and J. Mauchline. (1987). “Biology of Sternoptychid fishes rockall trough, northeastern Atlantic Ocean.” Biological Oceanography, 4(2), pp. 99-120. doi:10.1080/01965581.1987.10749486 Keene, M. J., R. H. Gibbs, and W. H. Krueger. (1987). “Family Melamphaidae, bigscales.” Smithsonian contributions to zoology, 452, pp. 169-185. Kenaley, C. P. (2008). “Diel vertical migration of the loosejaw dragonfishes (Stomiiformes: Stomiidae: Malacosteinae): a new analysis for rare pelagic taxa.” Journal of Fish Biology, 73(4), pp. 888-901. Krueger, W. H. (1972). “Biological studies of the Bermuda Ocean Acre III. Vertical distribution and ecology of the bristlemouth fishes (family Gonostomatidae).” Represented to U.S. Navy Underwater System center, Washington

Lancraft, T., T. Hopkins, and J. Torres. (1988). “Aspects of the ecology of the mesopelagic fish Gonostoma elongatum (Gonostomatidae, Stomiiformes) in the eastern Gulf of Mexico.” Marine Ecology Progress Series, 49(12), pp. 27-40. Libungan L. A. and S. Palsson. (2015). ”ShapeR: An R Package to Study Otolith Shape Variation among Fish Populations.” PLoS ONE, 10(3), pp. 0121102. Linkowski, T. B., R. L. Radtke, and P. H. Lenz. (1993). “Otolith microstructure, age and growth of two species of Ceratoscopelus (Oosteichthyes: Myctophidae) from the eastern north Atlantic.” Journal of Experimental Marine Biology and Ecology, 167(2), pp. 237-260. doi:10.1016/0022-0981(93)90033-k Longhurst, A. R. and W. G. Harrison. (1989). “The biological pump: profiles of plankton production and consumption in the upper ocean.” Progress in Oceanography, 22(1), pp. 47-123. Marks, A. D., D. W. Kerstetter, D. M. Wyanski, and T. T. Sutton. (2020). "Reproductive ecology of dragonfishes (Stomiiformes: Stomiidae) in the Gulf of Mexico." Frontiers in Marine Science, 7(101).

Mayank, P., R. K. Tyagi, and A. C. Dwivedi. (2005). "Studies on age, growth and age composition of commercially important fish species, Cirrhinus mrigala (Hamilton, 1822) from the tributary of the Ganga river, India." European Journal of Experimental Biology, 5(2), pp. 16-21. Maynard, S. D. (1982). “Aspects of the biology of mesopelagic fishes of the genus Cyclothone (pisces: Gonostomatidae) In Hawaiian Waters.” ProQuest Dissertations and Theses Global. McNutt, M. K., R. Camilli, T. J. Crone, G. D. Guthrie, P. A. Hsieh, T. B. Ryerson, O. Savas, and F. Shaffer. (2012). “Review of flow rate estimates of the Deepwater Horizon oil spill.” Proceedings of the National Academy of Sciences, 109(50), pp. 20260-20267. Mehner, T., (2012). “Diel vertical migration of freshwater fishes–proximate triggers, ultimate causes and research perspectives.” Freshwater Biology, 57(7), pp. 1342-1359. Menon, N. G., N. G. K. Pillai, R. Reghu, and K. Balachandran. (1996). “Distribution and abundance of the genus Vincigurria (Gonostomatidae) in the DSL of Indian EEZ with a note on the biology of Vinciguerria nimbaria.” Proceeding of second workshop of scientific results of FORV Sagar Sampada., pp. 271-284. Miya, M. and T. Nemoto. (1985). “Reproduction and ecology. Reproduction, growth and vertical distribution of the mesopelagic fish Cyclothone pseudopallida (family Gonostomatidae).” In: Uyeno, T.; R. Arai, T.Taniuchi and K. Matsuura (Eds.). Indo-Pacific Fish Biology. Proceedings of the Second International Conference on Indo-Pacific Fishes. The Ichthyological Society of Japan, Tokyo, i-xii, pp. 830-837. Miya, M. and T. Nemoto. (1986). “Life history and vertical distribution of the mesopelagic fish Cyclothone alba (family Gonostomatidae) in Sagami Bay, Central Japan.” Deep Sea Research Part A. Oceanographic Research Papers, 33(8), pp. 1053-1068. doi:https://doi.org/10.1016/0198-0149(86)90030-0 Miya, M. and T. Nemoto. (1991). “Comparative life histories of the meso- and bathypelagic fishes of the genus Cyclothone (Pisces: Gonostomatidae) in Sagami Bay, central Japan.” Deep Sea Research Part A. Oceanographic Research Papers, 38(1), pp. 67-89. doi:https://doi.org/10.1016/0198-0149(91)90055-K Moore, J. A. (1999). “Deep-sea finfish fisheries: lessons from history.” Fisheries, 24(7), pp. 16- 21. Olivar, M. P., A. Bernal, B. Molí, M. Peña, R. Balbín, A. Castellón, J. Miquel, and E. Massutí. (2012). “Vertical distribution, diversity and assemblages of mesopelagic fishes in the western Mediterranean.” Deep Sea Research Part I: Oceanographic Research Papers, 62, pp. 53-69. Pauly, D., V. Christensen, J. Dalsgaard, R. Froese, and F. Torres. (1998). “Fishing down marine food webs.” Science, 279(5352), pp. 860-863. Pearre Jr, S. (2003). “Eat and run? The hunger/satiation hypothesis in vertical migration: history, evidence and consequences.” Biological Reviews, 78(1), pp. 1-79.

Proud, R., M. J. Cox, and A. S. Brierley. (2017). “Biogeography of the global ocean's mesopelagic zone.” Current Biology, 27(1), pp. 113-119. doi:10.1016/j.cub.2016.11.003 Quist, M. C. and D. A. Isermann. (2017). “Age and growth of fishes: principles and techniques.” American Fisheries Society. Robison, B. H. (2009). “Conservation of deep pelagic biodiversity.” Conservation Biology, 23(4), pp. 847-858. doi:10.1111/j.1523-1739.2009.01219.x

Robinson, C., D. K. Steinberg, T. R. Anderson, J. Arístegui, C. A. Carlson, J. R. Frost, J. Ghiglione, S. Hernandes-Leon, G. A. Jackson, R. Koppelmann, B. Quéguiner, O. Ragueneau, F. Rassoilzadegan, B. H. Robison, C. Tamburini, T. Tanaka, K. F. Wishner, and J. Zhang. (2010). “Mesopelagic zone ecology and biogeochemistry – a synthesis.” Deep Sea Research Part II: Topical Studies in Oceanography, 57(16), pp. 1504-1518.

Schwarzhans, W. (1978). "Otolith-morphology and its usage for higher systematical units, with special reference to the Myctophiformes sl." Mededelingen van de Werkgroep voor Tertiaire en Kwartaire Geologie ,15(4), pp. 167-185.

Smith, C. L. and E. H. Atz. (1973). “Hermaphroditism in the Mesopelagic Fishes Omosudis lowei and Alepisaurus ferox.” Copeia, 1973(1), pp. 41-44. doi:10.2307/1442355

Stearns, S. C. and J. C. Koella. (1986). “The evolution of pheno-typic plasticity in life-history traits: predictions of reaction norms for age and size at maturity.” Evolution, 40, pp. 893– 913. Stequert, B., F. Menard, and E. Marchal. (2003). “Reproductive biology of Vinciguerria nimbaria in the equatorial waters of the eastern Atlantic Ocean.” Journal of Fish Biology, 62(5), pp. 1116-1136. doi:10.1046/j.1095-8649.2003.00104.x St. John, M. A., A. Borja, G. Chust, M. Heath, I. Grigorov, P. Mariani, A. P. Martin, and R. S. Santos. (2016). “A dark hole in our understanding of marine ecosystems and their services: perspectives from the mesopelagic community.” Frontiers in Marine Science, 3. doi:10.3389/fmars.2016.00031 Sutton, T. T. (2013). “Vertical ecology of the pelagic ocean: classical patterns and new perspectives.” Journal of fish biology, 83(6), pp. 1508-1527. Sutton, T. T., R. M. Clark, D. C. Dunn, P. N. Halpin, A. D. Rogers, J. Guinotte, S. J. Bograd, M. V. Angel, J. A. A. Perez, K. Wishner, R. L. Haedrich, D. J. Lindsay, J. C. Drazen, A. Vereshchaka, U. Piatkowski, T. Morato, K. Blachowiak-Samolyk, B. H. Robison, K. M. Gjerde, A. Pierrot-Bults, P. Bernal, G. Reygondeau, and M. Heino. (2017). “A global biogeographic classification of the mesopelagic zone.” Deep Sea Research Part I: Oceanographic Research Papers, 126, pp. 85-102. doi:10.1016/j.dsr.2017.05.006

Sutton, T. T. and T. L. Hopkins. (1996). “Trophic ecology of the stomiid (Pisces: Stomiidae) fish assemblage of the eastern Gulf of Mexico: Strategies, selectivity and impact of a top mesopelagic predator group.” Marine Biology, 127(2), pp. 179-192. doi:10.1007/bf00942102.

Sutton, T. T., P. H. Wiebe, L. Madin, and A. Bucklin. (2010). “Diversity and community structure of pelagic fishes to 5000m depth in the Sargasso Sea.” Deep Sea Research Part II: Topical Studies in Oceanography, 57(24-26), pp. 2220-2233. doi:10.1016/j.dsr2.2010.09.024 “The mesopelagic: Cinderella of the oceans”. The Economist, 15 April 2017, https://www.economist.com/news/science-and-technology/21720618-one-least- understood-parts-sea-also-one-most-important. Thorpe, J. E. (1994). “Salmonid fishes and the estuarine environment.” Estuaries, 17(1), pp. 76- 93. Tomás, J., and J. Panfili. (2000). “Otolith microstructure examination and growth patterns of Vinciguerria nimbaria (Photichthyidae) in the tropical Atlantic Ocean.” Fisheries Research, 46(1), pp. 131-145. doi:https://doi.org/10.1016/S0165-7836(00)00140-5 Tuset, V., A. Lombarte, and C. Assis. (2008). “Otolith atlas for the western Mediterranean, north and central eastern Atlantic.” Scientia Marina, 72(S1), pp. 7-198. http://dx.doi.org/10.3989/scimar.2008.72s17 Webb, T. J., E. V. Berghe, and R. O'Dor. (2010). “Biodiversity's big wet secret: the global distribution of marine biological records reveals chronic under-exploration of the deep pelagic ocean.” PLoS One, 5(8), pp. 10223. doi:10.1371/journal.pone.0010223

Wiebe, P. H., A. W. Morton, A. M. Bradley, R. H. Backus, J. E. Craddock, V. Barber, and G. R. Flierl. (1985). “New development in the MOCNESS, an apparatus for sampling zooplankton and micronekton.” Marine Biology, 87(3), pp. 313-323. doi:10.1007/bf00397811

Worm, B., R. Hilborn, J. K. Baum, T. A. Branch, J. S. Collie, C. Costello, M. J. Fogarty, E. A. Fulton, J. A. Hutchings, S. Jennings, and O. P. Jensen (2009). “Rebuilding global fisheries. Science, 325(5940), pp. 578-585.

Yamamichi, M., and B. E. Miner. (2015). “Indirect evolutionary rescue: prey adapts, predator avoids extinction.” Evolutionary applications, 8(8), pp. 787-795.

Zahuranec, B. J. (2000). “Zoogeography and systematics of the lanternfishes of the genus Nannobrachium (Myctophidae: Lampanyctini).” Smithsonian Contributions to Zoology 607, pp. 1–69.