Three-year investigations into sperm whale-fall ecosystems in Japan


  • Yoshihiro Fujiwara,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    Search for more papers by this author
  • Masaru Kawato,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    Search for more papers by this author
  • Tomoko Yamamoto,

    1.  Faculty of Fisheries, Kagoshima University, 4-50-20 Shimoarata, Kagoshima, Japan
    Search for more papers by this author
  • Toshiro Yamanaka,

    1.  Department of Evolution of Earth Environments, Graduate School of Social and Cultural Studies, Kyushu University, Ropponmatsu, Chuo-ku, Fukuoka, Japan
    Search for more papers by this author
  • Waka Sato-Okoshi,

    1.  Graduate School of Agricultural Science, Tohoku University, 1-1 Amamiya-machi, Tsutsumidori, Aoba-ku, Sendai, Miyagi, Japan
    Search for more papers by this author
  • Chikayo Noda,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    2.  Graduate School of Biosphere Science, Hiroshima University, 1-4-4 Kagamiyama, Higashi-Hiroshima, Japan
    Search for more papers by this author
  • Shinji Tsuchida,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    Search for more papers by this author
  • Tomoyuki Komai,

    1.  Natural History Museum and Institute, Chiba, 955-2 Aoba-cho, Chuo-ku, Chiba, Japan
    Search for more papers by this author
  • Sherine Sonia Cubelio,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    2.  Tokyo University of Marine Science and Technology, 4-5-7 Konan, Minato-ku, Tokyo, Japan
    Search for more papers by this author
  • Takenori Sasaki,

    1.  The University Museum, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, Japan
    Search for more papers by this author
  • Karen Jacobsen,

    1.  In Situ Scientific Illustration, Ketchum, ID, USA
    Search for more papers by this author
  • Kaoru Kubokawa,

    1.  Center for Advanced Marine Research, Ocean Research Institute, The University of Tokyo, 1-15-1 Minamidai, Nakano-ku, Tokyo, Japan
    Search for more papers by this author
  • Katsunori Fujikura,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    Search for more papers by this author
  • Tadashi Maruyama,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    Search for more papers by this author
  • Yasuo Furushima,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    Search for more papers by this author
  • Kenji Okoshi,

    1.  Faculty of Science and Technology, Ishinomaki Senshu University, 1 Shinmito, Minamisakai, Ishinomaki, Miyagi, Japan
    Search for more papers by this author
  • Hiroshi Miyake,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    2.  Enoshima Aquarium, Katasekaigan, Fujisawa, Kanagawa, Japan
    Search for more papers by this author
  • Masayuki Miyazaki,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    Search for more papers by this author
  • Yuichi Nogi,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    Search for more papers by this author
  • Akiko Yatabe,

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    2.  Tokyo University of Marine Science and Technology, 4-5-7 Konan, Minato-ku, Tokyo, Japan
    Search for more papers by this author
  • Takashi Okutani

    1.  Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa, Japan
    Search for more papers by this author

Yoshihiro Fujiwara, Extremobiosphere Research Center, Japan Agency for Marine-Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa 237-0061, Japan.


We report the first study of sperm whale-fall ecosystems, based on mass sinking of whale carcasses at shelf depths in the northwest Pacific. We conducted three observations over a 2-year period on replicate sperm-whale carcasses implanted at depths of 219–254 m off the southern part of Japan from July 2003 to August 2005. The study was made possible by a mass stranding of sperm whales in January 2002, and the subsequent sinking of 12 carcasses in the waters off Cape Nomamisaki. Dense aggregations of unique chemosynthesis-based fauna had formed around the whale carcasses after 18 months (July 2003). The mytilid mussel Adipicola pacifica was the most abundant macrofaunal species and covered most of the exposed bone surfaces. The general composition of the fauna was similar to that of deep-water reducing habitats, but none of the species appearing in this study has been found at hydrothermal vents, cold seeps or deep-water whale falls. A new species of lancelet, which was the first record of the subphylum Cephalochordata from reducing environments, a new species of Osedax; a rarely encountered benthic ctenophore, and a rare gastropod species were discovered at this sperm whale-fall site. Benthic communities were similar across all the carcasses studied, although the body sizes of the whales were very different. The succession of epifaunal communities was relatively rapid and the sulphophilic stage was considerably shorter than that of other known whale falls.


The first discovery of a whale-fall in situ occurred in 1987 in the Santa Catalina Basin, California at a depth of 1240 m (Smith et al. 1989). A large chemosynthetic assemblage was reported around the whale carcasses (Smith et al. 1989), with similarities to hydrothermal vent and cold-seep communities. Since this discovery, many whale-fall communities have been reported, including modern and fossil assemblages (Smith & Baco 2003). Whale falls have been thought to be ‘stepping stones’ not only for dispersal of deep-sea chemosymbiotic species but also for the introduction over evolutionary time of chemoautotrophy-dependent invertebrates to vent and seep environments (Smith et al. 1989; Distel et al. 2000).

A mass stranding of 14 sperm whales (Physeter macrocephalus Linnaeus, 1758) occurred on the southwestern coast of Kyushu Island, southern Japan, on January 22, 2002. The bodies of 12 whales were sunk by local government authorities in the waters off Cape Nomamisaki, southwestern tip of Kyushu Island, at depths of 200–300 m on February 1, 2002.

Most ecological studies of whale falls have been conducted on baleen whale carcasses off California, at depths of 1000–2000 m (Smith & Baco 2003). No sperm whale falls and related biological assemblages have been discovered before this mass sinking, although this whale species should be sufficiently large to sustain whale-fall-specific biological assemblages. In addition, sperm whales have an oil-rich structure known as the spermaceti organ that takes up 25–33% of the animal's body (Whitehead 2003). This unusual organ might serve as a unique habitat for whale-fall specialists.

The sperm whale falls off Cape Nomamisaki were located in waters shallower than most previously studied whale falls. The only whale-fall community reported shallower was at 125 m in the North Sea (Glover et al. 2005). The whale bone-eating siboglinid worm Osedax mucofloris was the most abundant species on the North Sea skeleton (Glover et al. 2005). It was not clear whether mass aggregation of chemosymbiont-bearing invertebrates occurs at shallow-water whale falls, as on deep-water falls.

While many whale-fall communities have been reported from the northeast Pacific, limited whale-fall information is available from the northwest Pacific. The only whale-fall community reported from this region was on the Torishima Seamount at a depth of 4037 m (Fujioka et al. 1993; Wada 1993). The skeleton had already been eroded heavily by the time of discovery. Unidentified mytilids, tubicolous polychaetes and galatheid crabs were abundant (Naganuma et al. 1996).

There have been no previous observations of multiple large whale carcasses implanted simultaneously in a specific limited area. The largest whale sunk off Cape Nomamisaki was twice as heavy as the smallest. The influence of whale body size on the formation and succession of biological assemblages could thus be investigated.

The aims of this study were to (1) clarify whether sperm whale carcasses, like those of baleen whales, sustain chemosynthetic communities, (2) characterize the macrofaunal assemblages on whale falls in relatively shallower water in the northwest Pacific, (3) investigate the ecological differences between whale-fall communities on carcasses of varying size, and (4) document patterns of ecological succession in shallow-water whale-fall communities.

Material and Methods

Site survey and sampling

The carcasses of 12 sperm whales were loaded on barges and sunk by local government authorities in the waters off Cape Nomamisaki, southwestern tip of Kyushu Island, at depths of 200–300 m on February 1, 2002 (Table 1, Fig. 1). The whales had decomposed internally but the external morphology was not severely damaged. Each whale was wrapped in nylon net (mesh size: approximately 10 cm). As ballast, six to thirteen concrete blocks (approximately 2.5 tons each) were attached to each whale using wire rope. Five of the whales were subsequently studied using the remotely operated vehicle (ROV) Hyper-Dolphin in 2003, 2004 and 2005 (Table 1). Ten dives were conducted in July 2003 (dives no. 189–198), six dives in July 2004 (dives no. 328–333) and eleven dives in July–August 2005 (dives no. 452, 453, 456–459, 462–466). Biological and geochemical samples were collected using manipulators, a scoop-sampler, a suction sampler and sediment corers installed on the ROV. Whale bones were collected using manipulators and stored in a sample box or a sample basket. Most epifaunal species were collected using a suction sampler, and sediment-dwelling infauna were collected using a scoop-sampler. Biological sorting was conducted using three sieves with different mesh sizes (0.5, 1 and 2 mm). Taxonomic identifications were made using collected specimens (except for some large species mentioned in Tables 2–5) by T. Okutani and T. Sasaki for mollusks; W. Sato-Okoshi & K. Fujikura for polychaetes; S. Tsuchida, T. Komai and S. S. Cubelio for crustaceans; H. Miyake for ctenophores; K. Kubokawa for cephalochordates; and T. Okutani and Y. Fujiwara for the remainder. Sediment samples were collected using PVC tube corers (7 cm in diameter, 28 cm in length). Core samples of bones were collected from vertebrae using an electric hole saw (3 cm in diameter, 30 cm in length). Water temperature was measured using the SBE 19 CTD profiler (Sea-Bird Electronics, Washington, USA).

Table 1.   Whale falls investigated in this study. These specimens were dropped off Cape Nomamisaki, Japan, on February 1, 2002.
whale length (m)estimated weight (t)depth (m)latitude (N)longitude (E)dive no./year
July 2003July 2004July–August 2005
  1. Whale numbers indicate the order of dropping to the seafloor. Whale lengths were provided by the local government and their weights were estimated according to Lockyer (1976). Geographic positions were calculated according to the World Geodetic System 1984 (WGS84). Dive numbers of the ROV Hyper-Dolphin are shown at each whale-fall site.

213.2023.021931 °23.865′129 °58.766′198462
616.0039.022831 °20.998′129 °59.158′191, 196329, 331
, 332
452, 456
712.9521.922931 °20.720′129 °59.285′189, 190
, 192, 193,
328, 329
, 331, 332
453, 457
, 464, 465
1113.0522.324531 °18.844′129 °59.520′333458
1213.5024.525431 °18.515′129 °59.374′194, 195330459, 463
, 466
Figure 1.

 Location of whale falls (solid circles: no. 2, 6, 7, 11 and 12) studied off Cape Nomamisaki, Japan. The locations of Cape Nomamisaki and Kagoshima Bay are indicated. The ordinate is in degrees north latitude and the abscissa in degrees east longitude. Open circle indicates the habitat of Lamellibrachia satsuma at the depth of 80 m.

Table 2.   Molluscan fauna associated with all the sperm whale falls investigated.
  1. Plus signs indicate the appearance (observed and/or collected) of each taxon in each year. Specimens collected/observed from all whale carcasses investigated are shown. The total number of taxa appearing in each year is given on the bottom row. All taxa shown in this table were collected unless otherwise stated.

  2. *Identified from high-definition TV video.

BivalviaSolemya pervernicosa+++
Adipicola crypta+++
Adipicola pacifica+++
Atrina teramachii  +
Chlamys empressae + 
Chlamys lemniscata  +
Lucinoma adamsianum  +
Wallucina sp.+  
Nitidotellina soyoae+  
GastropodaDillwynella vitrea+++
Homalopoma sp. ++
Cocculina sp.+++
Tanea magnifluctuata ++
Pisanianura breviaxe  +
Ceratostoma inornata ++
Mitrella bicincta ++
Zeuxis castus+  
Olivella spretoides + 
Pleurobranchella nicobarica + 
ScaphopodaStriodentalium rhabdotum + 
Gadilina sp. + 
CephalopodaOctopus sp.*  +
no. of taxa2281414
Table 3.   Polychaetous fauna associated with all the sperm whale falls investigated.
subclassorderfamily (species)200320042005
  1. Plus signs indicate the appearance of each taxon in each year. The total number of taxa appearing in each year is given on the bottom row. All taxa shown in this table were collected.

UnplacedParaonidae + 
Aphroditidae  +
Goniadidae+ +
Nephtyidae+ +
Onuphidae  +
SabellidaSerpulidae ++
Siboglinidae (Osedax japonicus)+++
Acrocirridae  +
Terebellidae  +
Apistobranchidae ?+  
UnplacedProtodrilidae ++
no. of taxa 23161518
Table 4.   Crustacean fauna associated with all the sperm whale falls investigated.
  1. Plus signs indicate the appearance (observed and/or collected) of each taxon in each year. The total number of taxa appearing in each year is given on the bottom row. All taxa shown in this table were collected unless otherwise stated.

  2. *Identified from high-definition TV video.

Ostracoda  unidentified sp.+  
 /CirripediaPedunculataHeteralepadidaeHeteralepas sp.+++
MalacostracaLeptostracaNebaliidaeunidentified sp.++ 
 AmphipodaGammaridaeunidentified sp.+++
 Cumacea unidentified sp.  +
 Euphausiacea unidentified sp. + 
  /CarideaAlpheidaeAlpheus sp. (macrochirus group?)  +
  HippolytidaeEualus sp. cf. kikuchii  +
  ProcessidaeProcessa philippinensis  +
  PandalidaePlesionika crosnieri  +
   Plesionika grandis  +
  /ThalassinideaCallianassidaeCallianassa s.l. sp. + 
  /AnomuraDiogenidaeCestopagurus sp. nov. ++
   Paguristes albimaculatus  +
  PaguridaeNematopagurus lepidochirus  +
   Nematopagurus spinulosensoris  +
  ChyrostylidaeEumunida sp.  +
  GalatheidaeGalathea sp. 1  +
   Galathea sp. 2  +
   Galathea sp. 3  +
   Galathea sp. 4+  
   Munida sp. 1 ++
   Munida sp. 2  +
   Munida sp. 3+  
   unidentified sp.+  
  /BrachyuraHomolidaeHomola orientalis  +
   unidentified sp. + 
  DorippidaeEthusa sp.  +
  LeucosiidaeCryptocnemus obolus  +
   unidentified sp. 1  +
   unidentified sp. 2 ++
   Merocryptus lambriformis  +
  MajidaeMacrocheira kaempferi* + 
   Pugettia minor  +
   Oxypleurodon stimpsoni  +
  AtelecyclidaeTrachycarcinus sagamiensis  +
  CancridaeCancer gibbosulus+  
   Cancer japonicus+  
  GoneplacidaeCarcinoplax surugensis+ +
  XanthidaeMedaeus serratus  +
   unidentified sp. ++
  PinnotheridaePinnixa sp. ++
  no. of taxa42101231
Table 5.   Other remarkable taxa associated with all the sperm whale falls investigated.
  1. Plus signs indicate the appearance (observed and/or collected) of each taxon in each year. The total number of taxa appearing in each year is given on the bottom row. All taxa shown in this table were collected unless otherwise stated.

  2. *Identified from high-definition TV video.

Poriferaunidentified sp. ++
Cnidariaunidentified sp.  +
 Hydrozoaunidentified sp. ++
 AnthozoaActiniaria: Unidentified sp.++ 
 Edwardsia sp.  +
 unidentified sp.  +
CtenophoraLyrocteis imperatoris  +
Nemerteaunidentified sp.+++
Sipunculaunidentified sp.+++
Echiuraunidentified sp.  +
Entoproctaunidentified sp. ++
Brachiopodaunidentified sp.++ 
 Crinoideaunidentified sp. ++
 Asteroideaunidentified sp. ++
 Ophiuroideaunidentified sp.+ +
 Phrynophiurida: Unidentified sp. ++
 Echinoideaunidentified sp.  +
 Holothuroideaunidentified sp.+  
 CephalochordataAsymmetron inferum+++
 VertebrataCongridae gen. sp.*  +
 Moridae gen. sp.*+++
 Ophidiidae gen. sp.*  +
 Helicolenus hilgendorfi*+++
 Sebastiscus tertius*+++
 Niphon spinosus* ++
 no. of taxa25101522

Quantitative sampling and measurements of shell length of Adipicola species

Quadrat sampling (quadrat size: 25 and 100 cm2) of Adipicola pacifica was conducted quantitatively on the surfaces of recovered whale bones. Three sets of quadrat samples were obtained in July 2003, five sets in July 2004 and four sets in July 2005. The total number and wet weight (including shells) of A. pacifica collected from each quadrat was measured. Shell lengths of all A. pacifica specimens collected for quantitative analyses and all Adipicola crypta collected during all three cruises were also measured.

Measurement of sulfide concentration in sediments

To measure sulfide concentrations in sediments, acid-volatile sulfide (AVS) was liberated by anaerobic acidification of cored sediments in 1 n HCl during active distillation (with pure nitrogen gas) of a ca. 5-cm3 core fraction and collection of liberated H2S in traps containing cadmium acetate solution (2.5%), following which the sulfide was precipitated as CdS. The yellow CdS precipitate was oxidized with a few drops of hydrogen peroxide solution (34.5%). These resulting sulfates were recovered as BaSO4 after adding BaCl2 solution. The resulting BaSO4 precipitates were precisely weighed and calculated as millimoles per kilogram of dry sediment (mm·S2−·kg−1). Analytical error associated with the overall process of the AVS determinations was less than 5%.

Specimen maintenance in aquaria

One vertebra, one ulna and one epiphysis collected in July 2004 and three vertebrae collected in July–August 2005 were maintained at JAMSTEC in five air-bubbled aquaria (approximately 100 l each) containing artificial Rohtomarine (Rei-Sea, Tokyo, Japan) seawater at 12 °C. The aquarium water was filtered using EHEIM classic or professional filters (EHEIM GmbH & Co. KG, Deizisau, Germany) and the seawater was exchanged when the transparency became low. The salinity was manually controlled at approximately 35‰ by adding freshwater. No organic dietary supplements were supplied.


Decomposition of whale carcasses

In July 2003 (1.5 years after emplacement), we visited four whale carcasses off Cape Nomamisaki, Japan, at depths of 219–254 m (Table 1, Fig. 1). The seafloor sediment was sandy and the water temperature was 12 °C. The whale carcasses had been largely skeletonized by this time (Fig. 2a and d). Most soft tissues, such as the viscera and muscle had been consumed, but large amounts of head and blubber tissue remained. Only the vertebrae of the largest whale (number 6, total length = 16 m) were still connected with the soft tissue.

Figure 2.

 A whale-fall community (whale number 12, 13.5 m length) at a depth of 254 m off Cape Nomamisaki, Japan. A dense aggregation of fauna had formed around the whale skeleton by July 2003. However, by July 2005 this community was already beginning to disappear. These images were taken by the ROV Hyper-Dolphin in (a) July 2003, (b) July 2004 and (c) July 2005. (d) Schematic drawing of whale no. 12 in July 2003.

The second whale-fall cruise was conducted 2.5 years after carcass emplacements (July 2004) (Table 1). The lower halves of the vertebrae of all skeletons were mostly buried in the sediment and their spinous processes had nearly disappeared (Fig. 2b). The skulls had broken down considerably (Fig. 2b). The connective tissue between the vertebrae had disappeared, but large amounts of soft tissue from the skulls and blubber were still present. The third cruise was conducted 3.5 years after the carcass sinking (July–August 2005) (Table 1). The bones were buried deeper in the sediments (Fig. 2c) but the cephalic soft tissues and some blubber were still present.

Whale bones were sampled during all three cruises. The vertebrae were relatively solid after 1.5 years but fragile and porous after 3.5 years, with little difference between individual whale skeletons. Core samples collected from vertebrae after 3.5 years were less oily and malodorous than those collected during the first and second sampling times.

Biological assemblages around whale carcasses

Dense biological assemblages occurred around sperm whale carcasses after 1.5 years. A mytilid mussel was the most abundant (Fig. 3) and covered most exposed surfaces of the skulls, ribs, epiphyses and vertebrae. Two other symbiont-harboring bivalves and a new species of lancelet were located in sediments underneath the skeletons. A new species of the polychaete in the genus Osedax inhabited the bones and soft tissues. After 2.5 and 3.5 years, whale carcasses showed higher species richness (44, 56, and 85 taxa at 1.5, 2.5 and 3.5 years respectively) and the biomass was the greatest at 1.5-year-old carcasses. Control sampling of background fauna was conducted more than 10 m from the carcasses. No species were shared between the whale falls and background environment.

Figure 3.

 Living specimens of the mytilid mussel Adipicola pacifica on a whale vertebra. Both inhalent and exhalent siphons were extended into the water, unlike the other mytilids.

Molluscan fauna

The molluscan fauna observed/collected at the whale fall sites is shown in Table 2. The most abundant mollusk was the mytilid mussel Adipicola pacifica (Dall, Bartsch & Rehder, 1938), which covered bone surfaces exposed to seawater. Adipicola pacifica extended its inhalent and exhalent siphons into the water column (Fig. 3), as reported by Okutani et al. (2003). Another whale-fall mussel, Adipicola crypta (Dall, Bartsch & Rehder, 1938) (Fig. 4a), was also found on the same whale carcasses but it was attached only to bone surfaces buried in sediments (Fig. 5). The solemyid clam Solemya pervernicosa Kuroda, 1948 was also collected from sediments beneath the carcasses (Fig. 4b). These three species were the most abundant bivalves observed throughout the 3.5-year study period. The most abundant gastropod species at all three sampling times was Dillwynella vitrea Hasegawa, 1997. Many carnivorous gastropods, such as Ceratostoma inornata (Récluz, 1851), Tanea magnifluctuata (Kuroda 1961) (Fig. 4c) and Mitrella bicincta (Gould, 1860), were present on 2.5- and 3.5-year-old carcasses but were rare on 1.5-year-old carcasses. Ceratostoma inornata was observed to feed on a live A. crypta in our aquaria.

Figure 4.

 Remarkable benthic fauna at the sperm whale site. (a) Adipicola crypta, (b) Solemya pervernicosa, (c) Tanea magnifluctuata, (d) an unidentified protodrilid polychaete, (e) Osedax japonicus, (f) Asymmetron inferum and (g) Lyrocteis imperatoris.

Figure 5.

 An ulna of a sperm whale. Adipicola pacifica (circled) was attached to part of the bone exposed to seawater and Adipicola crypta (squared) was attached to that buried in the sediments. The solid line indicates the boundary between the two areas.

Polychaetous fauna

Polychaete species collected/observed at the whale-fall sites are shown in Table 3. The total number of polychaete families was similar among years, but the family identities varied markedly between the 1.5-year and later sampling times. The Nereididae, Capitellidae and Dorvilleidae constituted more than 70% of total collected polychaete abundance at 1.5-year carcasses. On the other hand, a protodrilid polychaete was obviously more abundant on the 2.5- and 3.5-year-old ones (Fig. 4d), but population sizes could not be accurately estimated because the worm was located in small pores of whale bones and was not quantitatively sampled. The collected Cirratulidae, Lumbrineridae and Dorvilleidae constituted more than 80% of total polychaete abundance (excluding protodrilid and Osedax polychaetes) on the 2.5-year-old carcasses and more than 70% of total polychaetes on 3.5-year carcasses. Osedax japonicusFujikura et al. 2006 was found not only on maxillary bones but also in cephalic soft tissues and blubber at 1.5 years, and many specimens were discovered on vertebrae and in soft tissues at 2.5 and 3.5 years (Fig. 4e), as reported by Fujikura et al. (2006).

Crustacean fauna

Numerous cirripeds identified as Heteralepas sp. were observed clinging to the nylon nets wrapped around the whales but not on rocks and sediments far from the whale carcasses. The total number of crustacean taxa recorded was similar at 1.5 and 2.5 years, but threefold greater at 3.5 years (Table 4). Three specimens of the Japanese spider crab Macrocheira kaempferi (Temminck, 1836) were only found at the no. 12 whale site at 2.5 years (Fig. 2b).

Other main taxa

The remaining taxa recorded around the whale falls are listed in Table 5. The total number of taxonomic groups/species gradually increased from 1.5- to 3.5-year-old carcasses. A new species of lancelet, Asymmetron inferumNishikawa 2004, was discovered in the sediments beneath the whale carcasses at 1.5 years, and was present throughout the 3-year observation period (Fig. 4f). An undescribed sipunculan species was observed in bones and in sediments underneath the bones in all three years. An undescribed entoproct was originally found in our aquarium on land with two whale bones collected in July 2004. Many individuals of that entoproct were found on the aquarium glass and on shells of living A. pacifica. One specimen of the entoproct was also found in an aquarium onboard during the ROV cruise at 3.5 years. Molecular phylogenetic analysis of the species showed a close relationship to the genus Loxosomella (Fujiwara et al., unpublished data). More than 10 specimens of the benthic ctenophore Lyrocteis imperatorisKomai 1941 were discovered around the whale carcasses in July–August 2005 (Fig. 4g). Lyrocteis imperatoris specimens varied substantially in color (yellow, brown, white, white with red stripes and white with red spots). The tentacles were occasionally extended (Fig. 4g). Two species of rockfish, Sebastiscus tertius (Basukov & Chen, 1978) and Helicolenus hilgendorfi (Steindachner and Döderlein, 1884), were the most abundant fish inhabiting areas around the whale bones during the 3-year observation period. Several large specimens of the temperate bass Niphon spinosus Cuvier, 1828 were observed between whale bones in July 2004 and July–August 2005. Other predatory fish, i.e., a conger eel, a morid cod and an ophidiid cuskeel, were only seen on 3.5-year-old carcasses.

Shell length of whale-fall mussels

The shell lengths of the collected whale-fall mussels A. crypta and A. pacifica were measured (Fig. 6). The two Adipicola species showed different trends in shell length. The mean size of A. pacifica was the largest on 1.5-year-old carcasses and gradually decreased thereafter. On the other hand, that of A. crypta was the smallest on 1.5-year-old carcasses and gradually increased. The average shell length of A. crypta was 1.8-fold larger than that of A. pacifica on 1.5-year-old carcasses and 5.5-fold larger on 3.5-year-old ones.

Figure 6.

 Mean shell lengths of Adipicola pacifica (solid circles) and Adipicola crypta (solid squares). Numbers of specimens used for measurements are indicated. Means ± 1 standard error are given.

Density and biomass of Adipicola pacifica

The density of A. pacifica was the greatest on 2.5-year-old carcasses (July 2004) and the total number of the individuals was more than 100,000 per m2 (Fig. 7a). The biomass of A. pacifica was the greatest on 1.5-year-old carcasses (July 2003), exceeding 17 kg·m−2, and gradually decreased during the subsequent two years (Fig. 7b). The biomass on the surface of 1.5-year-old carcasses was more than 20-fold greater than that on 3.5-year-old ones.

Figure 7.

 Time series of density and biomass of Adipicola pacifica inhabiting surfaces of whale skeletons. (a) Number of individuals per square meter. (b) Wet weight (kg) of A. pacifica per square meter. Means ± 1 standard error are given.

Sulfide concentrations in sediments beneath whale bones

Vertical profiles of sulfide concentrations were measured in sediments beneath the whale skeletons during the 3-year observation period (Fig. 8). After 1.5 years, the highest sulfide concentrations were 0.87 mm·S2−·kg−1 at 5 cm below the surface of the sediment. After 2.5 years, the highest concentrations were 40.5 mm·S2−·kg−1 at 8 cm below the sediment surface. At 3.5 years maximum sulfide concentrations of 41.7 mm·S2−·kg−1 occurred at 14 cm below the sediment surface.

Figure 8.

 Concentrations of acid-volatile sulfide (AVS) in sediments beneath the whale carcasses. The ordinate indicates sampling depth (cm) below the surface of sediments and the abscissa the concentration of AVS (mm·S2−·kg−1). Solid triangles show the values sampled in July 2003, solid circles those in July 2004 and solid squares those in July 2005.


To the best of our knowledge, this is the first reported whale-fall ecosystem on sperm–whale carcasses, providing new insights into poorly known, shallow-water whale-fall communities. A sperm whale carcass can sustain a whale-fall-specific biological assemblage for more than 3 years, as has been reported for baleen whales in the deep sea (Smith & Baco 2003). The invertebrate fauna was similar to those in other whale-fall communities at the family level, but not at the species level (Table 6). Mytilid mussels, cocculinid limpets, and dorvilleid and Osedax polychaetes were the most abundant taxa, as in other whale-fall communities. However, none of the species in this study was recorded at any other whale-fall site except for A. pacifica (Smith & Baco 2003), perhaps because of the difference in water depth. Molecular phylogenetic analysis showed that Osedax polychaetes were divided into two groups based on habitat depth, independent of geographic location (Fujikura et al. 2006; Fujiwara et al., unpublished data). In addition, no species conspecific to those at the whale falls off Cape Nomamisaki appeared at the whale fall on the Torishima Seamount (depth =4037 m), which was the nearest previously reported whale-fall location (Naganuma et al. 1996). Further studies of shallow-water whale falls are required to clarify the influence of water depth and geography on the distribution of whale-fall specialists.

Table 6.   Abundant taxa collected at the whale-fall site off Cape Nomamisaki.
taxaoff Nomamisakiknown habitatcounterpart reported off Californiareference
  1. Their known habitat and their counterparts reported off California are shown. nr: taxa not reported.

 MytilidaeAdipicola pacificaSouth-East of Honshu, North-East Japan Sea, Hawaii, attached on sunken whale bones, 150–715 m deepIdas washingtoniaSmith et al. (1989)
, Okutani (2000)
Okutani et al. (2003)
 Adipicola cryptaSouth-East of Honshu, Hawaii, 100–200 m deep Okutani (2000)
Okutani et al. (2003)
 SolemyidaeSolemya pervernicosaSouth-East of Honshu and Hokkaido, 100–1500 m deepnrOkutani (2000)
 CocculinidaeCocculina sp. Cocculina craigsmithiMcLean (1992)
 SkeneidaeDillwynella vitreaNorth-East Japan Sea, South-East of Honshu and Hokkaido, attached on sunken wood, 100–200 m deepnrOkutani (2000)
 MuricidaeCeratostoma inornataJapan, Korean Peninsula, intertidal rocky bottom to 20 m deepnrOkutani (2000)
 ProtodrilidaeProtodrilidae gen. sp. nr 
 NereididaeNereididae gen. sp. nr 
 CapitellidaeCapitellidae gen. sp. nr 
 DorvilleidaeDorvilleidae gen. sp. Ophryotrocha spp.,
Dorvillea sp.
Dorvilleid sp.
Bennett et al. (1994)
Goffredi et al. (2004)
 SiboglinidaeOsedax japonicadiscovered first off Cape NomamisakiOsedax frankpressi
Osedax rubiplumus
Rouse et al. (2004)
Fujikura et al. (2006)
Sipunculaunidentified sp. Golfingia nicolasiSmith et al. (1998)
ThecostracaHeteralepas sp. nr 
CephalochordataAsymmetron inferumdiscovered first off Cape NomamisakinrNishikawa (2004)

Smith & Baco (2003) divided the succession of whale-fall communities into four stages, i.e., the ‘mobile-scavenger’, ‘enrichment opportunist’, ‘sulphophilic’ and ‘reef’ stages. The sperm whale-fall communities in this study were already at the sulphophilic stage in July 2003 (within 1.5 years of reaching the seafloor). Three abundant bivalves, A. pacifica, A. crypta and S. pervernicosa, harbored thioautotrophic symbionts in their gills (Fujiwara et al., unpublished data). Adipicola pacifica, in particular, attained extraordinary abundances (>100,000 m−2) and biomass (∼17 kg·m−2 wet weight); the biomass of A. pacifica at 1.5 years overlaps the extraordinary biomasses of bivalves observed at hydrothermal vents and cold seeps (Gebruk et al. 2000). Smith & Baco (2003) mentioned that the sulphophilic stage might be markedly long-lasting for large whale skeletons. Schuller et al. (2004) suggested that the whale-fall reducing habitat was extremely long-lived and able to support life for many decades, perhaps nearing a century. However, the sperm whale falls in the present study showed rapid ecological succession of epifauna, although the sperm whales investigated here were relatively large; the estimated body weight of the smallest specimen was 22 t and that of the largest 39 t.

The diversity and structure of the epifaunal communities changed markedly from the sulphophilic to the beginning of the reef stages between July 2003 and July 2004, at least on the surfaces of the whale skeletons. The biomass of A. pacifica was the greatest on the 1.5-year-old carcasses and rapidly decreased thereafter. Suspension feeders such as crinoids, basket stars, cnidarians and a benthic ctenophore were recorded on 2.5- and/or 3.5-year-old carcasses, which should be an indication of the reef stage. These taxa were not observed to attach to regions where bacterial mats, A. pacifica or O. japonicus appeared. These suspension feeders might be intolerant of reduced chemicals and/or organic compound effluents from whale skeletons. Indeed, many suspension feeders already existed on the concrete sinkers in 2003. In addition, carnivorous gastropods and predatory fish were abundant around 2.5- and 3.5-year-old carcasses. The reason for such rapid succession was unclear, but one possibility was that the biological decomposition, including bacterial degradation could be faster with the higher water temperature at these sperm whale-fall sites (12 °C) than at the 4 °C sites studied in the deep sea by Smith & Baco (2003). Shallow-water whale falls have rarely been reported, and rapid succession on shallow whale falls may explain their apparent rarity. Another possible cause for more rapid succession in our study might be the difference in whale species (sperm versus baleen whales) between the shallow- and deep-water studies. A new sperm whale carcass discovered at a deeper water depth (925 m) off Japan in January 2006 provides an opportunity to determine whether sperm whale-fall communities generally show rapid succession.

Compared to the rapid epifaunal succession, the diversity of infauna was relatively stable throughout the 3-year observations, i.e., A. crypta, S. pervernicosa, and A. inferum were abundant in each year. Sulfide concentrations in the sediments appeared to be sufficient for sulphophilic infauna at all sampling times, especially at 2.5 and 3.5 years (Fig. 8) (cf. Smith & Baco 2003). Further investigation will clarify whether the infaunal succession of shallow-water whale falls is rapid as well, compared with deep-sea whale falls.

Adipicola pacifica and A. crypta showed different trends in shell length in the present study, which implied a difference in the habitat conditions between epi- and infauna. Mean Adipicola pacifica size became shorter each year, indicating that its life span was less than 1 year and the conditions of its habitat had worsened. Consequently, new generations apparently did not grow as fast as the earlier generations. The whale bones collected in 2005 were very fragile and porous, and core samples collected from vertebrae were less oily and malodorous than those collected during the first and second cruises, which implied that many organic compounds such as energy-rich lipids and proteins had disappeared. Meanwhile, Adipicola crypta became larger with time. It appeared to have a longer life than A. pacifica and to survive for more than 3 years. Several specimens of A. crypta have been living with whale bones in our aquaria since July 2004, few dead shells have been found in the tanks, and the number of living specimens has remained stable. Adipicola crypta lived in the sediments beneath whale carcasses, in which the sulfide concentrations were higher in July 2004 and July 2005 than in July 2003. Energy-rich organic compounds apparently oozed from the whale bones into the sediments, and a suitable reducing environment for infaunal species might have formed through anaerobic bacterial degradation of the compounds just beneath the whale carcasses.

Decomposition of the largest whale carcass (whale no. 6) seemed to be slower than that of the others at the beginning of this study. However, the benthic communities were similar among the carcasses observed, although the body sizes of the whales were not homogeneous. The differences in the sizes of the whales in this study might not be a significant influence on benthic fauna at this stage of whale-fall development.

A shallow-water chemosynthetic community has been reported in Kagoshima Bay at 80 m (Hashimoto et al. 1993), which was close to the sperm whale site (Fig. 1). The vestimentiferan tubeworm Lamellibrachia satsuma was the most abundant species in the bay (Miura et al. 1997) and, based on molecular phylogenetic analyses, the same species was recorded at hydrothermal vents on the Nikko Seamount (470 m) 1500 km south of both Kagoshima Bay and the sperm whale site (Kojima et al. 2001). The whale-fall site off Cape Nomamisaki was thought to be a good candidate for a new distribution site of L. satsuma because it was in both the geographic and water depth ranges of the tubeworm. However, no vestimentiferans (including L. satsuma) were found at the whale-fall site. In addition, no vesicomyid clams, no Bathymodiolus mussels and no chemosymbiotic gastropods were recorded at this site, which implies low species-level similarity between this shallow-water whale-fall and deep-sea vent/seep environments. It is possible that a suitable chemosynthetic environment for tubeworms and other chemosymbiotic taxa would be generated after 3.5 years at the Cape Nomamisaki whale falls.

The new species of lancelet A. inferum was discovered in the sperm whale-fall habitats; this was the first recorded cephalochordate from a chemosynthetic habitat and the deepest recorded branchiostoma (Nishikawa 2004). In general, lancelets inhabit shallow, subtidal tropical, subtropical and temperate sand flats. They prefer coarse sand with fairly fast water flow and do not inhabit silty sediments (Berrill 1987; Nishikawa et al. 1997). However, this whale fall-related lancelet preferred a deeper, reducing, organic compound-rich environment. Physiological experiments will clarify the mechanism of adaptation of this species to such an extreme environment.

The benthic ctenophore L. imperatoris was described as a new species in 1941 (Komai 1941) and has not been reported for more than 60 years. The type specimens were collected in a dredge survey. Therefore, this was the first scientific observation of L. imperatorisin situ. It is not certain, however, whether this benthic ctenophore is specifically a whale fall-related species because many specimens appearing around the skeletons did not attach to the bones.

The naticid gastropod T. magnifluctuata is also a very rare species. The only two reports of this species (Kuroda 1961; Matsumoto 1979) did not describe the habitats, soft tissues and operculum. This species is assumed to be carnivorous but its diet was not clarified.

The Osedax species collected in this study was examined morphologically and phylogenetically and was described as the new species O. japonicus (Fujikura et al. 2006). The most characteristic feature of its ecology was the whale-fall habitat. Unlike other known Osedax species, Osedax japonicus appeared upon cephalic soft tissues and blubber in addition to whale skeletons. It was unclear whether its habitation on the soft tissues was due to the whale species, the Osedax species or both. One possible explanation could be that the large amount of lipids in sperm whales supports Osedax growth. Further sperm whale-fall studies in Sagami Bay may resolve this question.

Shallow-water whale falls appear to be rarely encountered, possibly because they are easily decomposed by biological/microbiological activities at relatively higher water temperatures and/or they are quickly buried in sediments transported from shore. Therefore, a large number of whale-fall dependent species may remain undiscovered. Molecular phylogenetic analyses indicated that A. pacifica diverged prior to other bathymodiolin mussels (Miyazaki et al., personal communication), Osedax is a sister taxon to hydrothermal vent/seep siboglinids (Rouse et al. 2004; Glover et al. 2005; Fujikura et al. 2006), A. inferum diverged early in the evolution of lancelet (Kon et al. 2005) and L. imperatoris occupies a basal position among ctenophores (Fujiwara et al., unpublished data). Shallow-water whale-fall ecosystems might have thus played an important role in the evolution not only of deep-sea chemosynthetic ecosystems but also of all marine life.


Sperm whale-fall communities were investigated for 3 years using an ROV. Five sperm whale carcasses sustained chemosynthesis-based communities at depths of 219–254 m, which were similar to the deeper whale-fall communities in general but with certain unique features. The rate of epifaunal succession was notably more rapid than that of deeper communities on large whale falls. The sulphophilic stage appears to be much shorter, although the whale carcass sizes should have been sufficiently large for long-term support of sulphophilic species. No vent/seep specialists were present at this whale-fall site and many new, poorly described and/or rarely encountered species appeared. Further information on whale carcasses from a wide range of areas and depths will clarify the spatiotemporal dynamics of deep-sea life in such isolated, ephemeral habitats.


We wish to express our sincere thanks to Professor Hidehiro Kato (Tokyo University of Marine Science & Technology) and local inhabitants for their efforts in dropping the whale carcasses. We are very grateful for helpful comments from Drs Hiroshi Senou (Kanagawa Prefectural Museum of Natural History) and Keiichi Matsuura (The National Science Museum) on the identification of fish, to Professor Toshiyuki Yamaguchi (Chiba University) on the identification of the cirriped and to Dr Tohru Iseto (The Kyoto University Museum) and Professor Yoshihisa Shirayama (Seto Marine Biological Laboratory, Kyoto University) on the identification of the entoproct. We are deeply obligated to Prof. Craig Smith (associate editor) and all three reviewers for their careful and considerate review as they have indeed helped us improve this manuscript. We thank Dr Dhugal J. Lindsay for his useful suggestions and Mr Mamoru Sano for creating maps. We thank Mr Hitoshi Tanaka, Ms Misumi Aoki, and Mr Kaoru Tsukuda for active support aboard, Ms Kazuyo Okano for groundwork for the first cruise, Dr Yukiko Fujii, Ms Yoko Sasaki and Ms Hiroko Nakamura (JAMSTEC) for laboratory assistance and Ms Shizue Kanai and Ms Tomomi Nagashima (JAMSTEC) for administrative help. We also thank the operation teams of the ROV Hyper-Dolphin and the JAMSTEC deep tow systems and the captains and crew of the R/V Natsushima and R/V Kaiyo.