Polar Biology

, Volume 34, Issue 3, pp 363–370

Temperature-driven biogeography of the deep-sea family Lithodidae (Crustacea: Decapoda: Anomura) in the Southern Ocean


    • National Oceanography Centre, Southampton, School of Ocean and Earth ScienceUniversity of Southampton
  • Sven Thatje
    • National Oceanography Centre, Southampton, School of Ocean and Earth ScienceUniversity of Southampton
Original Paper

DOI: 10.1007/s00300-010-0890-0

Cite this article as:
Hall, S. & Thatje, S. Polar Biol (2011) 34: 363. doi:10.1007/s00300-010-0890-0


Species’ distributions are dynamic and are linked to the changing physical environment. Temperature is considered to be a major factor influencing biogeography, especially in ectotherms such as the family Lithodidae. Lithodids are rare amongst decapods in their ability to inhabit the higher latitudes of the Southern Ocean; however, they are usually found in locations where water temperature is above 0.5°C. This study, for the first time, provides a baseline indication of the limits of the lithodid distribution around Antarctica, which will be instrumental in any future work on range extensions in this group. The distribution of lithodids is likely to change as temperatures along the West Antarctic Peninsula continue to rise, and range extensions by durophagous predators, such as the lithodids, are regarded as a potential threat to the unique structure of Antarctic continental-shelf ecosystems.


Cold adaptationKing crabPredatorsClimate changePolar ecosystemsBiogeography


Lithodid crabs are a group of predominantly deep-sea crustaceans, whose predatory trophic position and commercial fishery makes their dynamic global distribution interesting. The family Lithodidae occupies predominantly cold-water habitats (Zaklan 2002; Thatje and Arntz 2004; Hall and Thatje 2009) and temperature is a constraining factor for distribution, particularly because of its effect on early life-history stages (Anger et al. 2003; Thatje et al. 2005). Lineage-wide lower and upper thermal limits have been predicted (approximately 0.5 and 15°C respectively) for deep-sea genera of the Lithodidae (Lithodes, Paralomis, Neolithodes) and their secondarily shallow-water relatives (Hall and Thatje 2009). This study, for the first time, examines the hypothesis of a lineage-wide physiological thermal threshold using a biogeographical dataset.

Temperature affects the most fundamental biological processes; in ectotherms, all aspects of life—from basic chemical reactions to feeding and reproductive behaviour—can be related to environmental temperature (Pörtner 2001). Physiological limitations within the Lithodidae mean that larvae will not successfully develop to maturity at temperatures lower than 0–2°C in laboratory experiments (Nakanishi 1981; Shirley and Shirley 1989; Thatje 2004). The speed of growth and development is reduced after even a small decrease in environmental temperature (Kurata 1960; Nakanishi 1985; Anger et al. 2003, 2004; Calcagno et al. 2003, 2005). This can be particularly important for early life stages that are vulnerable to seasonal fluctuations in food abundance and size-related predation pressures (Holm-Hansen 1985, Thatje 2004). Latitudinal clines in fecundity and egg size (Wägele 1987; Gorny et al. 1992; Fischer et al. 2009) are indications of the increased cost of reproduction at high latitudes and the transition to a reproductive strategy with increased energy investment per offspring (Atkinson et al. 2001; Thatje 2004). The adoption of a lecithotrophic larval mode of development in the deep-sea genera of the Lithodidae allows independence from seasonal primary production and enables tolerance of the protracted development times that are associated with cold waters in polar and deep-sea environments (Shirley and Zhou 1997; Thatje et al. 2003). Optimal levels of growth and survival during larval development occur between 5 and 10°C in laboratory-reared lithodid species (Anger et al. 2003; Jørgensen et al. 2005).

The Southern Ocean

Very few crushing seafloor predators inhabit the waters around Antarctica and even fewer on the continental shelf (Thatje and Arntz 2004; Aronson et al. 2007); brachyuran crabs, which are important community-structuring predators elsewhere in the shallow ocean, are absent from Southern Ocean (Gorny 1999; Frederich et al. 2001). The general lack of crabs, sharks and rays on the high-Antarctic continental shelves may be one of the primary explanations for the rather archaic structure and function of shelf benthos when compared with shallow-water communities elsewhere in the oceans (for review see Aronson et al. 2007). The presence of large anomuran crabs in the Southern Ocean is limited to the Lithodidae, which have been found in several deep-water locations including the continental slope of the West Antarctic Peninsula (Klages et al. 1995; Thatje and Arntz 2004; Ahyong and Dawson 2006; Thatje et al. 2008). Three genera of deep-sea lithodids (Lithodes, Neolithodes and Paralomis) are found at latitudes above 60°S, although species number is lower than in the adjacent sub-Antarctic regions (Lovrich et al. 2005; Thatje et al. 2005). For several reasons, the Southern Ocean is an interesting study area for examining temperature-linked distributional trends in mobile, benthic organisms such as the Lithodidae, whose evolutionary history in the deep sea has been established (Hall and Thatje 2009).
  • The region (40–60°S) is divided by persistent oceanic fronts within the Antarctic Circumpolar Current (ACC), which extend to great depths and are identified by marked temperature changes (Deacon 1937; Nowlin and Klink 1986; Moore et al. 1999; Barker and Thomas 2004). The ACC is bounded to the north by the subtropical front (35–45°S), where cold sub-Antarctic waters meet warm, saline subtropical waters (Orsi et al. 1995; Rintoul et al. 2001). To the south (c. 60°S), the ACC is defined by the Antarctic Divergence, where there is an upwelling of deep waters (Fig. 1).
    Fig. 1

    Map of the Southern Ocean indicating the position of the Antarctic convergence: part of the circumpolar current. This study area is divided by the 60°S line of latitude into sub-Antarctic (45–60°S) and Antarctic (60°S+) regions. Dots indicate the major sites in which lithodids have been found to date. Refer to references for exact locations

  • The Southern Ocean includes some of the coldest waters on earth and spans the physiological lower temperature range for the Lithodidae. The Weddell Sea, for example, has temperatures consistently below 0.5°C (Lamb 1977; Barnes et al. 2006) and would not be expected to support reproductive lithodid populations based on experimental data (Shirley and Shirley 1989; Thatje 2004).

  • Temperature regimes in the Southern Ocean are changing over time and space, with the changing position of the fronts and a steady increase in water temperature. Locations such as the West Antarctic Peninsula (WAP) are among the fastest warming areas on earth as a result of global warming (Meredith and King 2005).

This study aims to establish the baseline for a link between temperature and distribution of the family Lithodidae in a region with many potential thermal ‘barriers’. This dataset will be expanded through further scientific discovery and will be instrumental in the future to illustrate and understand the changing faunal distributions in some of the most rapidly changing marine environments on earth (Meredith and King 2005; Aronson et al. 2007).

Materials and methods

The study area

For this study, the sub-Antarctic/Southern Ocean region (Fig. 1) was divided into two sections based on latitude and oceanographic features. The range 45–60°S includes locations within the ACC, from the sub-Antarctic front to the southern circum-Antarctic front (Antarctic divergence). The ACC itself passes south of Patagonia before diverging across the Argentine continental shelf (Antezana 1999), so fauna in and around coastal South America are not included in this study.

The range 60–75°S approximately includes all waters south of the Antarctic divergence (the southern boundary of the ACC). Scott Island in the Ross Sea, Peter I island in the Bellingshausen Sea, and the West Antarctic Peninsula (WAP) (all south of 60°S) lie south of the Antarctic divergence. Discontinuities in lithodid distribution mean that 60°S is a convenient division of the region for the purpose of this study (Thatje et al. 2005; Ahyong and Dawson 2006; Thatje et al. 2008).

Data sources

Data were gathered from three sources, as follows:
  1. (1)

    Published records of lithodids were sourced from peer-reviewed journals and other literature (including some comprehensive reviews of faunal distributions in different regions of the Southern Ocean Acosta et al. 1989; Ahyong and Dawson 2006; Ahyong 2010; Arana and Retamal 1999; García Raso et al. 2005; Klages et al. 1995; López Abellán and Balguerías 1993; Macpherson 1988, 2004; Miquel et al. 1985; Purves et al. 2003; Spiridonov et al. 2006; Thatje et al. 2005, 2008) Identity was verified from descriptions and pictures, or by inspection where samples were deposited in museums.

  2. (2)

    Specimens (mostly unpublished records) were examined from museum collections in the Natural History Museum, London (NHM); Senckenberg Museum, Frankfurt; Musée National d’Histoire Naturelle, Paris (MNHN); Institut de Ciencies del Mar, Barcelona; United States National Museum of Natural History, Smithsonian Institute, Washington (USNM).

  3. (3)

    Specimens with associated environmental data were obtained courtesy of commercial vessels or scientific cruises; material from the 2007 R.V. James Clark Ross cruise to the Bellingshausen Sea (JCR166) and material from Atlantis ‘08 expedition were used.


For each specimen studied, the depth, location and date of sample collection were noted. The study included all of the 17 species of lithodid described to date from the study area. Data from 82 sample locations (indicated in Fig. 1) south of 45°S (not including South American continent) were used in this study. Water temperature at the time of sampling was obtained from cruise reports where possible. Otherwise, temperatures were estimated based on time of year, depth and location. The majority of the estimated environmental data were taken from the National Oceanographic Data Centre <World Ocean Atlas 2005> (Locarnini et al. 2006) and the <Southern Ocean Atlas> (Olbers et al. 1992) (http://odv.awi.de/en/data/ocean/southern_ocean_atlas/).

Within each latitudinal range outlined above (45–60°S, 60–75°S), the longitudinal and depth distributions were examined within and between species of lithodids (Figs. 3, 5, 6, 7). The southern-most frontier of the lithodid distribution was examined by plotting water temperature variation with depth and longitude at 55°S, 60°S, 65°S, 70°S and 75°S (Figs. 3, 4, 5, 6, 7). These data were sampled from the <Southern Ocean Atlas> (Olbers et al. 1992); each temperature estimate included all data from within two degrees of the stated latitude (e.g. 60 ± 2°S). Temperature at depths of 200, 500, 1,000, and 2,000 m were sampled, because they are relevant to the distribution of lithodids in this region.


Antarctic and Subantarctic distribution

Members of the lithodid genera Paralomis, Lithodes and Neolithodes inhabit some of the relatively warmer waters around Antarctica (Klages et al. 1995; Arana and Retamal 1999; García Raso et al. 2005; Thatje et al. 2008; Ahyong 2010). Most often, the ambient temperatures in which lithodids are found in Antarctic/sub-Antarctic waters are between 1 and 4°C. The coldest waters in which lithodids have been found are between 0.4 and 0.5°C in the Ross Sea (Fig. 2).
Fig. 2

Temperature measured or estimated at sample location for species of Lithodinae south of 45°S (excluding Patagonia). Temperature data obtained either from in situ measurements at the time of capture, or from the <Southern Ocean Atlas> (Olbers et al. 1992). A dashed line at 0.5°C indicates the predicted physiological threshold for some members of the family, coinciding with the coldest waters in which many lithodid species are known to persist

Region 45–60°S

Fourteen species of lithodids are known to occur between 45 and 60°S (Paralomis aculeata, P. anamerae, P. birsteini, P. elongata, P. formosa, P. granulosa, P. spinosissima, Lithodes confundens, L. murrayi, L. santolla, L. turkayi, Neolithodes diomedeae, N. duhameli, N. capensis). Paralomis spinosissima, P. formosa and P. anamerae have overlapping distributions, and numerous specimens have been found around South Georgia (Fig. 3). Exploratory fisheries have shown that P. spinosissima is encountered regularly in waters between 200-800 m around Shag Rocks and South Georgia (López Abellán and Balguerías 1993; Purves et al. 2003). Paralomis formosa is found in deeper waters, typically 600–1,600 m (Purves et al. 2003; Thatje et al. 2005) and adults have been found where ambient temperatures are between 0.5 and 2.1°C (Figs. 2, 3). In the southern islands of the Scotia Arc (60°S, 25–60°W: 0.5 to −1°C; Fig. 4), no populations of Paralomis have been identified above 500 m, despite a significant sampling effort (López Abellán and Balguerías 1993). Paralomis aculeata was originally described at 600 m off the Prince Edward Islands (45°S) and is known also from the Crozet Islands in the southern Indian Ocean (46°S: Arnaud and Do-Chi 1979; Miquel et al. 1985; Macpherson 2004). It has a wide bathymetric range, approximately 150–1,500 m, and a recorded temperature range from 2 to 4°C (Fig. 2). Between 25°W and 30°E, no records of any lithodids exist to date, with the exception of P. elongata at 300 m close to Bouvet Island (54°S, 2°E; 0.5–3°C, Fig. 3; Arntz et al. 2006; Spiridonov et al. 2006). This gap in distribution coincides with low water temperatures in the mid-Atlantic (1.5–0°C at 55°S: Fig. 3), although the sampling effort in this region is unknown.
Fig. 3

a Lithodidae captured between 45 and 60°S shown by sampled depth against longitude. To increase graphical clarity, species Lithodes murrayi, L. santolla, L. confundens and L. turkayi are pooled; as are species Neolithodes diomedeae (50–30°W), N. duhameli (60–80°E) and N. brodiei (170°E). Paralomis elongata (3°E) and P. aculeata (30–60°E) were shown to be synonymous in molecular studies (Thatje et al. 2008); b Temperatures at depths 200, 500, 1,000, 2,000 and 3,000 m below sea level at longitudes between 54 and 56°S. Data obtained from the <Southern Ocean Atlas> (Olbers et al. 1992)

Fig. 4

Temperatures at depths 200, 500, 1,000 and 2,000 m below sea level at longitudes between 59 and 61°S. Data obtained from the <Southern Ocean Atlas> (Olbers et al. 1992). No lithodids are found between these latitudes despite a significant sampling effort around the southern islands of the Scotia Arc (indicated)

Region 60+°S

Within high-Antarctic waters above 60°S, lithodid diversity declines substantially and includes species Paralomis birsteini, P. stevensi, Lithodes murrayi, Neolithodes diomedae, N. yaldwyni and N. capensis (Figs. 5, 6, 7). At 65°S, temperature dips substantially below zero in the Weddell Sea and in the Australian-Antarctic basin, particularly above 500 m (Fig. 5). The areas between 60–70°S in which lithodids are found are those with notably higher temperatures (0.5–2.5°C, Figs. 5, 6, 7), such as the Bellingshausen Sea and the Balleny islands near the Ross Sea. Lithodids are present on the continental slope of the West Antarctic Penninsula (WAP) but have not yet been found on the continental shelf, where temperatures are around 1°C (60°W, Fig. 5). The Southern Ocean below 70°S is divided by the continental landmass into a Weddell Sea section (−2 to 1°C) and a Ross Sea section (0.5–1.5°C). To date, lithodids have only rarely been found south of 70°S and they are absent at all latitudes from the Weddell Sea. Sparse records of adult Neolithodes yaldwyni from more than 1,000 m water depth at 75°S in the Ross Sea sector are taken from waters where the temperature is around 1°C (Fig. 7). The continental shelf above 75°S in both the Weddell Sea and the Ross Sea is colder than 0.5°C, reaching sub-zero temperatures (Fig. 7).
Fig. 5

a Sample depth and longitude of Lithodidae captured between of 62.5 and 67.5°S. b Temperatures at depths 200, 500, 1,000 and 2,000 m below sea level at longitudes between 64 and 66°S. Data obtained from the <Southern Ocean Atlas> (Olbers et al. 1992)

Fig. 6

a Sample depth and longitude of Lithodidae captured between of 67.5 and 72.5°S (none recorded in this range above 70°S). Neolithodes species are N. yaldwyni, N. capensis and N. diomedeae. b Temperatures at depths 200, 500, 1,000 and 2,000 m below sea level at longitudes between 69 and 71°S. Data obtained from the <Southern Ocean Atlas> (Olbers et al. 1992). Grey blocks represent landmass above sea-level

Fig. 7

a Sample depth and longitude of Lithodidae captured between of 72.5 and 76°S. b Temperatures at depths 200, 500 and 1,000 m below sea level at longitudes between 73 and 76°S. Data obtained from the <Southern Ocean Atlas> (Olbers et al. 1992). Grey blocks represent landmass above sea-level


Temperatures in the Southern Ocean are low but stable; seasonal temperatures only fluctuate by a few degrees Celsius (Foster 1984). Diversity of lithodids in the region 45–60°S is higher than 60–70°S, and species Neolithodes yaldwyni and Paralomis stevensi are both endemic to waters south of 60°S. This indicates that some adaptations to very low temperatures are present in lithodids living at the lowest end of the family’s temperature range (Hall and Thatje 2009).

A limit to the currently recognised range of the Lithodidae coincides with regions where water temperature is colder than 0.4°C (Fig. 2), such as the Weddell Sea (Figs. 5, 6, 7) and the Antarctic continental shelf at all longitudes (Figs. 6, 7). Close to the Antarctic continent, particularly in the Weddell and Ross Seas, extremely dense, cold Antarctic Bottom water (AABW) sinks from the surface to the deep sea. The Scotia Arc, between the Antarctic Peninsula and Patagonia (Acosta et al. 1989), has northern (South Georgia, c. 52°S) and southern archipelagoes (South Shetland, South Orkney, c. 59°S), which have different temperature profiles and faunal distributions (e.g. Lovrich et al. 2005; Griffiths et al. 2008). The cold deep water formed in the Weddell Sea makes the temperature in the south-east Scotia Ridge particularly low (Fig. 4). This is a possible reason for the absence of the Lithodidae from this area, despite their abundance in the archipelago to the north (Gorny 1999; Arntz et al. 2005).

Temperatures are greater than 0.5°C in some areas from which lithodids are thought to be excluded, such as 70°S in the Ross Sea (Fig. 6b) and the continental shelf of the WAP. On the WAP continental shelf, for example, temperatures are only slightly lower (1–1.5°C, Fig. 5) than the adjacent continental slope, where lithodids are present (Thatje et al. 2008). Sampling effort is significant in the Southern Ocean, especially in the Weddell Sea and shallower parts of the WAP region (e.g. Arntz et al. 2005, Brandt et al. 2007), but it can not be ascertained to what extent the lack of records is the result of undersampling.

Physiological experiments to date have suggested a lower limit for larval development between 1 and 0.5°C in some Southern Ocean species (Thatje 2004 and references therein). Adult specimens found in the very coldest waters of the Southern Ocean might be migrant rather than reproductive populations—adults may tolerate temperatures lower than do larvae or juveniles, meaning that reproductive populations can’t establish at the frontier of the lithodid range. Contrary to this hypothesis, thirteen specimens of P. birsteini, including juveniles, were recently video-recorded between 1,123 and 1,394 m water depths on the Antarctic continental slope/rise in the Bellingshausen Sea (Thatje et al. 2008; see also García Raso et al. 2005). This, as well as the presence of ovigerous females of P. stevensi and P. birsteini (Ross Sea: Ahyong and Dawson 2006; Bellingshausen Sea: Arana and Retamal 1999) above 60°S, indicates that reproductively active populations of at least some lithodid species do, in fact, exist south of the Antarctic divergence. No records of lithodid larvae have been found within the Southern Ocean; however, deep-sea lithodids are understood to have lecithotrophic larvae that disperse by drifting close to the sea floor (Thatje 2004; Lovrich 1999).

The changing thermal structure of oceans may play an important role in determining patterns of lithodid biogeography. This could be an increasingly important phenomenon in consideration of climate change and oscillations in oceanic upwelling zones (Thatje et al. 2005). Warming of the polar oceans might be gradually opening up new habitats to the Southern Ocean lithodids (Aronson et al. 2007), and the slight mismatch between distributional and physiologically theoretical thresholds could be evidence of a range-expansion in progress.

Species of the genera Paralomis, Lithodes and Neolithodes are among the few anomuran taxa found at high latitudes in the Southern Ocean and it seems likely that a history of deep-sea adaptation (Thiel et al. 1996; Thatje 2004), has been associated with their successful colonization of Polar regions. In the Antarctic, particularly in the Bellingshausen Sea, lithodids have the potential to threaten the isolated shelf communities (Thatje et al. 2005; Aronson et al. 2007), which have evolved in the absence of crushing predators (such as crabs, lobsters, sharks and rays) that would be found in shelf ecosystems at lower latitudes (Feldmann and Tshudy 1989; Dayton et al. 1974; Arntz et al. 1994; Crame 1994; McClintock and Baker 1997). Here, where lithodids of the subfamily Lithodinae seem to be living at the lower boundary of their physiological threshold, even a slight increase in temperature might open up new habitats. Lithodids have the potential to expand into previously uninhabitable regions of polar seas if water temperatures continue to increase, with potentially devastating effects for the Antarctic shelf fauna.


This research is supported through a grant from the Total Foundation (Abyss2100) to Sven Thatje. We would like to thank staff at the National Museum of Natural History, Smithsonian Institute, Washington (USNM), especially Rafael Lemaitre and Karen Reed; Régis Cleva at the Muséum national d’Histoire naturelle, Paris (MNHN); Paul Clark at the Natural History Museum (NHM), London; Enrique Macpherson at the Centro de Estudios Avanzados de Blanes; Michael Türkay at the Senckenberg Museum, Frankfurt; and Ana Ramos and her group at Instituto Español de Oceanografía, Vigo. This study would not have been possible without the information made available by these institutions. This work was supported by the Marine Biodiversity and Ecosystem Functioning Network of Excellence MarBEF (contract no. GOCE-CT-2003-505446) of the FP6, and through a NERC PhD studentship to Sally Hall.

Copyright information

© Springer-Verlag 2010