Early life history of the daubed shanny (Teleostei: Leptoclinus maculatus) in Svalbard waters
- First Online:
- Cite this article as:
- Meyer Ottesen, C.A., Hop, H., Christiansen, J.S. et al. Mar Biodiv (2011) 41: 383. doi:10.1007/s12526-010-0079-3
- 523 Downloads
The daubed shanny (Leptoclinus maculatus, Family Stichaeidae) is considered to be an ecologically significant species in the arctic waters of Norway because of high abundance and the unique energy storage abilities of its postlarvae. Both postlarvae and adults are found in relative large abundances in Svalbard fjords as well as along the ice edge of north-east Svalbard, even at sub-zero temperatures. The postlarva feeds primarily on Calanus spp. and stores lipids from this high-energy diet in a unique lipid sac on the ventral side of its body. This energy storage enables it to survive pelagically during the arctic winters when food is scarce. The postlarvae are pelagic for the first 2–3 years of their life before they descend to the bottom and transform to a benthic mode of life. Our results indicated that this transition takes place when the postlarva reach about 80 mm in length and an age of 3 years. The relative size of the lipid sac, as well as changes in the morphology of the postlarvae, can be used as indices of the transition from pelagic postlarva to benthic juvenile. The lipid sac index (% of gutted weight) was negatively correlated with length and age and started to decrease when the postlarva transformed to a benthic lifestyle. At this point, the growth in terms of the length–weight relationship shifted from a positive asymptotic growth to a negative asymptotic growth, reflecting the changes in the lipid sac index, while several changes in morphological traits took place. The benthic mode of early juveniles is largely influenced by environmental conditions and prey abundance, since the fish is depleted from their energy stores and need to rapidly switch to benthic prey in their new habitat. In conclusion, this is the first study on the early life history of the daubed shanny and it presents evidence that the first 5 years in the life of this species is divided in two distinct parts, one pelagic and one benthic, in which postlarvae display unique growth and morphological traits adapted to the challenges of these arctic marine environments.
KeywordsLeptoclinus maculatus Postlarva Lipid storage Adaptability to high arctic waters Early life history Morphology
The daubed shanny (Leptoclinus maculatus) Fries, 1838, is one of three teleost fish species in the family Stichaeidae (Gill 1864) that are commonly found in the arctic waters. In the Barents Sea, they are distributed from the polar front and northwards (Byrkjedal and Høines 2007). Unlike their southern relatives, the arctic stichaeids have evolved several physiological and biochemical adaptations that enable them to survive sub-zero temperatures and prolonged periods of food shortage. Among these are the synthesis of antifreeze proteins (C. Ottesen, unpublished data) and a unique lipid sac in postlarval fish which may store large amounts of dietary lipids (Falk-Petersen et al. 1986). The lipid sac is situated on the ventral part of the body under the gut and runs from the pectoral fins to the anal opening. It consists of closely packed lipid vacuoles of different sizes which are surrounded by a simple membrane. The lipids are mainly triacylglycerols (80%) and phospholipids (18%) (Murzina et al. 2008, 2010), and fatty acid signatures originate from feeding on Calanus spp. Although the role of lipid accumulation in buoyancy control of fishes is well known (Clarke et al. 1984; Phleger 1998), Falk-Petersen et al. (1986) concluded that the lipid sac of L. maculatus is primarily an energy storage organ and that its role in buoyancy control is of minor importance. The lipid sac enables the postlarvae to take full advantage of the high secondary production in the arctic summer and store sufficient energy to meet metabolic requirements during periods of food shortage in winter. Furthermore, L. maculatus has a lipid content in the flesh of approximately 40% dw (dry weight) (Falk-Petersen et al. 1986). The lipid sac, together with the energy-rich muscle tissue, obviously represents important adaptations for larval L. maculates to sustain itself in a seasonally variable and extreme arctic marine environment. The pelagic postlarvae are abundant and constitute a valuable food item for predatory fishes, e.g. Atlantic cod (Gadus morhua) and polar cod (Boreogadus saida) (C. Ottesen, unpublished data), sea birds, e.g. common guillemot (Uria aalge) (Watanuki et al. 1992) and Atlantic puffin (Fratercula arctica) (Barrett 2002), and seals, e.g. ringed seal (Phoca hispida) (Labansen et al. 2007) and harp seal (Phoca groenlandica) (Nilssen and Lindstrøm 2005). As such, they are a valuable component in the transfer of energy from Calanus spp. to higher trophic levels in the arctic marine ecosystem of Svalbard.
Leptoclinus maculatus has a circumpolar distribution between latitudes 79°N and 43°N (http://www.fishbase.org) and two sub-species have been recognised with L. m. diaphanocarus in the Pacific Ocean and L. m. maculatus in the Atlantic Ocean (Mecklenburg and Sheiko 2004). Leptoclinus maculatus is a small-sized species with a maximum length of ∼200 mm, but rarely more than 160 mm (Andriyashev 1954). Both postlarvae and adults are found in relatively high numbers along the ice edge and in the fjords of west and north-east Svalbard. In the Barents Sea and Svalbard waters, L. maculatus are found at low temperatures (−1.6 to 2.0°C) and high salinities (33–35‰) (Andriyashev 1954). In the north Atlantic, L. maculatus probably spawns demersally in shallow waters during winter (Pethon 2005). However, the exact times of spawning and hatching are unknown as eggs and prolarval stages have not been reported. It is not known if L. maculatus prolarva has a yolk sac stage where the yolk sac is later transformed to the lipid sac, or if they hatch with the lipid sac as an energy supply instead of a yolk sac. Leptoclinus maculatus has a relatively low fecundity (≤1,000 eggs) (Andriyashev 1954; Pethon 2005) and quite large eggs of 1.50 mm in diameter (Christiansen et al. 1998). Leptoclinus maculatus probably exhibits parental care as has been reported for several species within the Stichaeidae family (Baylis 1981; Gross and Shine 1981). The life history of L. maculatus comprises a pelagic phase of early postlarvae, and a benthic phase of late postlarvae, juveniles and adults that live on soft bottoms from 15 to 400 m depth (Pethon 2005), and in the Barents Sea from 50 to 240 m depth (Andriyashev 1954). Arctic fishes are limited to cold waters and a relative narrow temperature range and, therefore, they are considered particularly vulnerable to climate change (Rose 2005).
In this study, we examined the body morphometrics and pigmentation, growth (i.e. body size-at-age and body weight–length relationships), and lipid sac indices (LSI) of postlarval L. maculatus. Based on these characters, we describe four postlarval stages which are either entirely pelagic or display a transitional phase between a pelagic and benthic mode of life.
Materials and methods
Sampling of fishes
Analyses of postlarvae
Number of postlarval daubed shanny (Leptoclinus maculatus) from each habitat analysed in this study
All statistical treatments and graphical presentation of the data were done in SYSTAT12 (2007 version). As no differences could be found between sampling areas and seasons, data were pooled for analyses. Based on the result of the length distribution of postlarvae, two size-groups were identified: one pelagic consisting mostly of L2 and L3 stages and one transitional consisting of L4 and L5 stages (see “Results”). Rather than using the individual postlarval stages, these groups were used to describe the length–weight relationship and growth of postlarvae. Thus, postlarva stages were pooled into these two groups in these descriptions. A linear regression module was applied for analysing the length–weight relationship and growth of postlarval groups. Log gutted weight and log caudal length were used in the linear regression module. One outlier was removed from the analysis so that the results represent the majority of each postlarval group. These results were tested for significance using ANCOVA. Morphological differences between postlarval stages and postlarval groups were determined by discriminant analysis, with the complete model. Each morphological measurement was tested using Kruskal–Wallis for postlarval stages and two-sample t test for postlarval groups.
Description of postlarval stages
Length and age distribution according to habitat
Size-at-age and LSI of postlarval stages
Length increased moderately for 1–2 years, and then a sharp increase in length was apparent when the postlarvae reached about 60 mm in length at 2 years of age. The length growth slowed down after the postlarvae had reached ≈80 mm and 3 years of age (Fig. 6a). The same pattern was seen for weight with age (Fig. 6b). It increased slightly until the postlarvae reached 2 years of age and ≈0.60 g gutted weight. The weight increase was fast until the postlarvae reached ≈1.5 g and 3 years of age, and subsequently the growth slowed down. This pattern of growth in length and weight was also seen during maturation, which seems to occur when the fish is around 6 years of age (C. Ottesen, unpublished data).
Length–weight relationship and LSI of postlarval groups
Length–weight relationship of L. maculatus postlarval groups based on a linear regression analysis of log gutted weight as a function of log caudal length. One outlier was removed from the analysis
Results of the complete model of the discriminant analyses of postlarval stages (df1/df2 = 42/132, F ratio = 22.2, p = 0.001) and postlarval groups (df1/df2 = 14/44, F ratio = 9.86, p = 0.001) of L. maculatus
Depth of caudal peduncle
Dorsal fin length
Anal fin length
The classification matrix obtained from discriminant analysis of postlarval stages, and postlarval groups of L. maculatus
Morphometric measurements of L. maculatus and their ability to distinguish between different postlarval stages (a) and postlarval groups (b)
A. Postlarval stages (Kruskal–Wallis)
B. Postlarval groups (two-sample t test)
Length of dorsal fin
Length of caudal fin
Depth of caudal peduncle
Length of anal fin
Length and age distribution
We found five different postlarval stages of L. maculatus, and four stages were processed in our analyses. The smallest stages (L1–L3) were only found in the pelagic environment, the largest (L4–L5) were in a transitional phase and found both in the pelagic environment and at the bottom. The length and age distribution of L. maculatus postlarvae showed a relatively large size- and age-range in the pelagic environment. Our results suggest that L. maculatus larvae grow and develop the first 2–3 years in the pelagic environment and then descend to the bottom as late postlarvae or early juveniles. This seems to be in contrast to most other benthic fishes, which only have a short pelagic larval stage in conjunction with the zooplankton bloom. All the benthic specimens of L. maculatus postlarvae were >70 mm and ≥2 years. Although the age distribution was more even in the postlarval samples, there was a notable lack of specimens between 61 and 71 mm in length, resulting in two peaks in the length distribution: one between ≈40 and 60 mm (pelagic postlarvae) and one between ≈70 and 100 mm (transitional postlarvae). Similar size-classes were found in arctic shanny (Stichaeus punctatus) from Nuvuk Islands, Canada, where the samples were dominated by two size-classes (Keats et al. 1993). These size-classes corresponded to specimens that were 2+ and 3+ years and 55 mm long and to fish aged 6 and older at 125 mm in length, although fish aged 4–5 years were absent from the samples (Keats et al. 1993). The biomodal size- and age distributions were attributed to variable and irregular recruitment in the population. Variable recruitment could explain the absence of fishes of 61–71 mm in length from our samples, although there was no concurrent absence of age groups. This may also be a consequence of high predation pressure on this size-range when they first attempt to settle at the bottom, or variation in habitat preference (e.g. depth and temperature) between size groups, as reported for longfin prickleback (Bryozoichthys lysimus) (Tokranov and Orlov 2004), which influences sampling of postlarval size-groups.
Size-at-age and LSI
The results showed a slow increase in size from ages 1 to 2, rapid increase until the age of 3, and then slow increase thereafter. The most plausible explanation for this result is differences between larval developmental stages, feeding success, habitat change and LSI. The slow increase in size from 1 to 2 years of age may be misleading since we only obtained a few specimens of <2 years of age. However, a slow larval growth has been reported in arctic shanny (Stichaeus punctatus) (Keats et al. 1993), long shanny (Stichaeus grigorjewi) (Kyushin 1990) and Opisthocentrus zonope from cold waters (Gnyubkina and Markevich 2008). Our results may reflect hatching during the winter and thus slow growth for the 0- to 1-year-old larvae due to low temperatures and low availability of prey. The LSI of the smallest postlarva (L2 stage) increased from 1 to 2 years of age and between 2 and 3 years (L3–early L4 stages), and as the LSI reached its peak the postlarvae increased rapidly in size with age. When they reached 3 years (80 mm, 1.7 g), the growth in size with age slowed down. At the same age and size, the LSI started to decrease and the lipid sac was absorbed and disappeared as the pelagic postlarvae developed into benthic juveniles. As the lipid sac diminishes in size when the larvae are at the late L4 stage, they become increasingly dependent upon securing enough energy for their metabolic processes and can no longer rely on the energy reserve. At the same time, they settle on the bottom and switch to different types of prey. They will also encounter other predators and have to learn to avoid them. Thus, increased energy expenses associated with the habitat shift are expected to result in reduced growth.
Length–weight relationship and LSI of postlarval groups
In contrast to increase in size with age, it was the smallest size-group (the Pelagic group) that had the fastest growth in weight. The Pelagic group (L2–L3 stages, ≈40–60 mm in length) had a positive asymptotic growth (b ≈ 3.6), thus becoming stouter with length, while the Transitional group (L4–L5 stages, ≈70–100 mm) had a negative allometric growth (b ≈ 2.7), thus becoming thinner with length. The differences in s in length–weight relationship and growth between the Pelagic group and the Transitional group are likely related to the decreasing LSI of the Transitional group, and the different life modes of the two groups. Differences in habitat and abiotic factors (e.g. temperature and salinity) likely influenced the length–weight relationships. Transitional postlarvae might not obtain as much energy for tissue growth as the Pelagic group. As was evident from the size-at-age results, the growth slows down in the oldest and largest postlarvae. It should also be mentioned that eel-like larvae swim slowly, have manoeuvring problems and rarely attack the same prey twice (Froese 1990). Therefore, they face challenges in obtaining sufficient food for rapid growth when they no longer can rely on stored lipid reserves. Comparatively, pelagic postlarvae still have large lipid stores that they can rely on if the prey abundance is low. As we have seen, the increase in size with age seems to be slow for the smallest specimens. However, specimens in the Pelagic group must also allocate some energy for growth as their small size makes them available to a large range of predators. Also, eel-shaped fish larvae do not swim constantly but alternate between active swimming when feeding and passive gliding (Froese 1990). This is considered to be energetically advantageous (Froese 1990) and thus enables the pelagic postlarvae to sustain their lipid reserves through the winter when zooplankton is scarce. The low temperature of the high arctic waters also slows down the metabolism so that the postlarvae can save energy for growth and survival.
Large morphological changes occurred between the L3 and L4 stages. The L2–L3 and the L4–L5 stages, respectively, overlapped in morphological features, but the L2–L3 stages did not overlap much with the L4–L5 stages. The L3 stage is the one with most specimens misclassified to other stages by the discriminant analysis. Half the specimens were misclassified as L2 stages, which indicates that the L3 specimens we obtained were intermediate between L2 and L3 and that the lack of specimens in the 61–71 mm length-group might represent the late L3 stage. The high misclassification of L3 specimens in the discriminant analysis could also be contributed to the subjective classification of the larval stages. It may be that L2 and L3 should have been classified as one stage rather than separate stages. The absence of the 61–71 mm size-group is one probable reason for the gap between the L2–L3 stages and L4–L5 stages. The classification of specimens into postlarval groups was almost 100% correct and thus the groups did not overlap. This supports the idea that the absent 61–71 mm specimens represent a morphological step in the development. It also supports that at around 80 mm (the size-range represented by the Transitional group), L. maculatus changes from a pelagic life mode to a benthic life mode and starts to develop into juveniles. As the postlarvae develop from L2–L3 stages (the Pelagic group) into L4–L5 stages (the Transitional group), they become stouter with a significantly deeper head and body, larger eyes, and jaws. This likely reflects the habitat and life mode change between the Pelagic group and the Transitional group. The L2–L3 stages are translucent, which makes them difficult to detect in the pelagic habitat. As they grow into L4–L5 stages, they become more pigmented and thus develop a benthic camouflage. The wider jaws enable them to exploit a larger range of prey, and the larger eyes might make them more efficient predators in the benthic habitat. The more powerful body enables them to better avoid benthic predators.
The L. maculatus postlarva has a unique lipid sac that enables them to store energy reserves when food is available in the pelagic that they can survive on during the winter. The lipid sac is likely an adaptation for survival in high Arctic waters. It enables the slow growing L. maculatus postlarvae to live in the pelagic zone, where there a fewer predators, during the first 2–3 years of their life. The results of age and size distribution, length-at-age, growth and LSI of postlarval stages indicate that the transition between pelagic and a benthic life mode happens when the postlarva reaches approximately 80 mm in length at 3 years of age. At this time, their lipid sac is absorbed, its body becomes more densely pigmented and changes into a deeper more powerful body with larger jaws and eyes. This presumably increases their survival in the benthic habitat. Although there are descriptions of L. maculatus postlarvae from other areas of their distributional range, the lipid sac does not seem to be present in postlarvae from these areas. It may be that the lipid sac is a trait that has evolved only in some high Arctic populations, such as the population in Svalbard waters.
As L. maculatus is an ecologically important species in arctic waters, more work needs to be done to reveal different aspects of its development and life history. Laboratory rearing of L. maculatus should be done to study and describe its embryonic development, larval size at hatching and morphology, development of its prolarva and development from prolarva to postlarva. If there is a yolk sac stage, the duration from the time of yolk absorption to development of lipid sac should also be studied, as well as the size at first feeding and the impact of feeding on the development of the lipid sac. We further suggest study of morphological traits of postlarvae from the whole distribution range of L. maculatus to examine if the lipid sac of the postlarvae is present in many populations or if it is a trait that have evolved only in some high arctic populations. Such a study should include morphometric measurements of the postlarvae, its lipid sac if present, and lipid composition in different populations, as well as DNA studies.
This publication was originally presented at the Arctic Frontiers Conference in Tromsø, January 2010. The support and initiative of ARCTOS and Arctic Frontiers are gratefully acknowledged. This work is supported by Statoil and ARCTOS. We also acknowledge the TUNU-Programme (University of Tromsø) for the opportunity to sample material. We would like to thank the crew of RV Jan Mayen for their assistance.
This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.