Content uploaded by Evgeny Romanov
Author content
All content in this area was uploaded by Evgeny Romanov on Mar 14, 2014
Content may be subject to copyright.
ORIGINAL PAPER
The swimming crab Charybdis smithii: distribution,
biology and trophic role in the pelagic ecosystem
of the western Indian Ocean
Evgeny Romanov ÆMichel Potier Æ
Veniamin Zamorov ÆFre
´de
´ric Me
´nard
Received: 20 August 2008 / Accepted: 27 January 2009 / Published online: 1 March 2009
ÓSpringer-Verlag 2009
Abstract Surface ‘‘swarms’’ of the swimming crabs
Charybdis smithii are still considered as an unusual phe-
nomenon in the open Indian Ocean, although their dense
pelagic aggregations were already reported in waters off
the Indian coast and in the northern Arabian Sea. Based on
an extensive large-scale data series taken over 45 years, we
demonstrate that C. smithii is common in the pelagic
provinces of the western Indian Ocean driven by the wind
monsoon regime. Swimming crabs are dispersed by the
monsoon currents throughout the equatorial Indian Ocean.
They aggregate at night in the upper 150-m layer, where
their estimated biomass derived from pelagic trawling data
can exceed 130 kg km
-2
. Abundance of C. smithii can
reach[15,000 ind. km
-2
in July (i.e. the peak of the south-
west monsoon), declines by 50-fold in March and is neg-
ligible in May. C. smithii is an important prey for more
than 30 species of abundant epipelagic top predators. In
turn, it feeds on mesopelagic species. This swimming crab
is a major species of the intermediate trophic levels and
represents a crucial seasonal trophic link in the open ocean
ecosystem of the western Indian Ocean. Outbursts in
pelagic waters of huge biomasses of ordinarily benthic
crustaceans (C. smithii and Natosquilla investigatoris) are a
remarkable feature of the Indian Ocean, although similar,
but smaller, events are reported in the Pacific and Atlantic
Oceans.
Introduction
The western Indian Ocean is a highly dynamic monsoon-
driven ecosystem exhibiting contrasting oceanographic
regimes that follow seasonal variability in atmospheric
circulation. The impacts of the monsoon on the oceanic
currents and on the productivity of the global ecosystem
are well documented (Bakun et al. 1998; Longhurst 1998;
Schott and McCreary 2001). However, the biotic responses
of intermediate- and high-trophic levels of pelagic food
webs to strong seasonal environmental changes have been
relatively poorly studied until now.
Seasonal alterations in the pelagic ecosystem are visible
from variable fishing patterns and yields of highly mobile
pelagic fisheries such as tuna purse seine fishing and
pelagic longline fishing (Fonteneau 1997), which follow
aggregations of top predators on an ocean-wide scale.
Large pelagic fish have a high energy demand (Korsmeyer
and Dewar 2001) and search continuously for food. Their
aggregations indicate mesoscale zones of high secondary
productivity (Fonteneau et al. 2008). In addition, these
species are widely used as ‘‘biological samplers’’ to study
intermediate trophic-level species recovered in their
stomachs (Cherel et al. 2007; Potier et al. 2008). Oppor-
tunistic feeding habits of pelagic top predators (Me
´nard
et al. 2006) allow us to use the prey occurrence in the
stomach contents as indicators of seasonal changes in the
Communicated by X. Irigoien.
E. Romanov (&)F. Me
´nard
CRH (Centre de Recherche Halieutique Me
´diterrane
´enne et
Tropicale), IRD (Institut de Recherche pour le De
´veloppement),
UR 109 Thetis, BP 171, 34203 Se
`te Cedex, France
e-mail: evgeny.romanov@ird.fr
M. Potier
IRD, UR 109 Thetis, BP 172, 97492 Ste Clotilde Cedex,
La Re
´union, France
V. Zamorov
I.I. Mechnikov Odessa National University (ONU),
2, Dvoryanskaya St., 65000 Odessa, Ukraine
123
Mar Biol (2009) 156:1089–1107
DOI 10.1007/s00227-009-1151-z
pelagic realm, and to investigate biological characteristics
of the most abundant prey species.
The Indian Ocean swimming crab Charybdis smithii
(Portunidae) has been reported as an important part of the
diet for several large pelagic species: yellowfin tuna,
Thunnus albacares; bigeye tuna, Thunnus obesus; and
longnose lancetfish, Alepisaurus ferox (Merrett 1968;
Bashmakov et al. 1991; Potier et al. 2007a,b; Romanov and
Zamorov 2007). Its pelagic ‘‘swarms’’ were observed off
the African coast, in the Arabian Sea and off Hindustan
(Sankarankutty and Rangarajan 1962; Della Croce and
Holthuis 1965; Losse 1969; Rice 1969; Spiridonov 1994;
van Couwelaar et al. 1997), while benthic records are
widely scattered along the slope of the western and northern
Indian Ocean from South Africa to the Andaman Sea (Apel
and Spiridonov 1998;Tu
¨rkay and Spiridonov 2006).
Compared with the majority of portunids, C. smithii
have a light, faintly calcified, non-granulated carapace
(Tu
¨rkay and Spiridonov 2006). Consequently, juvenile C.
smithii may drift and swim in the open ocean after meta-
morphosis and do not need substrate to survive. This ability
to remain in the water column allows juveniles and adults
to colonize the vast monsoon area of the Indian Ocean
using oceanic circulation.
Despite its abundance, little attention has been paid to
the biological studies of this species beyond Indian waters
(Balasubramanian and Suseelan 1990,1998,2001). Cur-
rent knowledge of the pelagic phase of C. smithii is based
on sparse sampling (van Couwelaar et al. 1997)oron
regional surveys (Balasubramanian and Suseelan 2001).
But these investigations have covered only a small part of
its hypothetical distribution area outlined by van Cou-
welaar et al. (1997) and Tu
¨rkay and Spiridonov (2006)
(Fig. 1a). Almost nothing is known on the occurrence,
abundance and biological characteristics of C. smithii in
the high seas, in particular within the equatorial area.
Strong seasonal variability of the pelagic biomass of the
swimming crab has been reported in the Arabian Sea (van
Couwelaar et al. 1997; Baars et al. 1998), however, the
origin and strength of this variability are still unclear.
In this paper, we investigate the pelagic distribution,
abundance, biology and role of C. smithii in the open ocean
ecosystem in relation with environmental forcing and ocean
dynamics. We analyzed an extensive archival dataset com-
piled over 40 years (1961–2001) as well as recent data
collected during the 2000–2007 period over the whole Indian
Ocean area. Data from trawl surveys and samples recovered
from the stomach contents of pelagic top predators were
used. The combination of long-term data series and of two
sampling methods allowed us to obtain new information on
the role of C. smithii in the pelagic trophic pathways and to
develop new hypotheses on the life cycle of C.smithii and its
links with regional environmental processes.
Materials and methods
Study area and sampling
Two sets of data were explored in this study (Table 1):
Soviet/Ukrainian data (YugNIRO
1,2
), covering the histori-
cal period (1961–2001) and French (IRD
3
) data for a recent
period (2000–2007).
Trawl surveys
A total of 7,258 trawl operations (11,307 h of trawling)
were analyzed, with 1,084 tows (1,301 h) in the western
Indian Ocean. Commercial midwater trawls during
YugNIRO cruises and an International Young Gadoid
Pelagic Trawl (IYGPT) during IRD cruises were used. The
horizontal and vertical openings of commercial trawling
nets used by YugNIRO were in the ranges of 25–65 and
20–60 m, respectively. IYGPT has a horizontal opening of
15.8 m and a vertical opening of 8 m. A cod-end of 10 mm
stretched mesh was used. Trawls were towed horizontally
at a speed of 1.5–6.8 knots; fishing depths ranged from the
surface to 1,300 m (YugNIRO) and from 25 to 255 m
(IRD). The duration of the tows varied from 10 to 420 min
(average 74 min) for commercial trawls, while IYGPT
trawl was towed for 30 min (Table 1a).
Stomach sampling
Top pelagic predators (tunas, billfishes, sharks and asso-
ciated species) caught by pelagic longlines and purse seines
were used as biological samplers to collect data on
C. smithii. Stomach content analyses allowed us to estimate
occurrence of the crabs in the environment and its trophic
contribution to predators’ diet (Table 1b, c).
Research fishing operations with pelagic longlines
(4,805 sets) throughout the Indian Ocean (Table 1b) were
the main source of information for crabs from the deep
layers (29,865 non-empty stomachs of top predators). The
estimated depth of longline hooks ranged from 50 to 400 m
in YugNIRO and from 50 to 125 m in IRD cruises.
Commercial purse seine operations were carried out in the
western equatorial waters and southward to the Mozambique
1
YugNIRO—Southern Scientific Research Institute of Marine
Fisheries & Oceanography, Kerch, Crimea, Ukraine.
2
Exploratory cruises were arranged by auxiliary organization, which
carried out fisheries research under general auspice of YugNIRO:
former Department of Searching and Scientific Research Fleet at
the Southern Basin of Ministry of Fisheries of the USSR
(Yugrybpromrazvedka) (till 1988). Starting from 1993 operated as
fishing company. Terminated its activity in 2004–2006.
3
IRD—Institut de Recherche pour le De
´veloppement, France.
1090 Mar Biol (2009) 156:1089–1107
123
30° S
30° E 40° E 50° E 60° E 70° E 80° E 90° E 100° E 110° E
20° S
10° S
0°
10° N
20° N
A
I n d i a n O c e a n
Saya-de-Malha Bank
Saya-de-Malha Bank
Saya-de-Malha Bank
Ind. hour-1
500-1000
10-500
0-10
0-0
30° S
30° E 40° E 50° E 60° E 70° E 80° E 90° E 100° E 110° E
20° S
10° S
0°
10° N
20° N
B
Stomachs
with crabs
50-600
5-50
0-5
0-0
30° S
30° E 40° E 50° E 60° E 70° E 80° E 90° E 100° E 110° E
20° S
10° S
0°
10° N
20° N
C
M O N S
I S S G
E A F R
Ind. 1,000 m
-3
0.05-0.50
0.01-0.05
0.00-0.01
0.00-0.00
Fig. 1 Pelagic distribution of
swimming crab Charybdis
smithii in the Indian Ocean.
aPublished records: crosses are
non-quantitative observations
(Sankarankutty and Rangarajan
1962; Della Croce and Holthuis
1965; Zarenkov 1968; Losse
1969; Rice 1969; Daniel and
Chakrapany 1983; Spiridonov
1994; van Couwelaar et al.
1997; Neumann and Spiridonov
1999; Christiansen and Boetius
2000); black shades are 1°
aggregated data from
Balasubramanian and Suseelan
(2001) in number of crabs per
hour tow adapted from their
original Fig. 1. Upper left
corners are catch of Isaaks-Kidd
midwater trawl (IKMT), lower
right corners are catch by
pelagic trawls, dots shows zero-
catch squares either for IKMT
or for pelagic trawls. Records of
sedimentation from pelagic
realm (Christiansen and Boetius
2000) are given, while bottom
and stranding records are not
presented. Bold line is a border
of the hypothetical species
distribution area from van
Couwelaar et al. (1997).
bThis study, longline samples:
cumulative number of fish
preyed on C. smithii.cThis
study, trawl samples: average
abundance of C. smithii in
ind. 1,000 m
-3
.Horizontal
black polygon lines show a
border between monsoon
provinces (MONS) of the
northern Indian Ocean outlined
by Longhurst (1998) (including
coastal provinces) and Indian
South Subtropical Gyre
Province (ISSG). Limits of the
Eastern Africa Coastal Province
(EAFR, vertical black polygon)
are also shown. On all panels,
approximate position of 200-
mile Exclusive Economic Zones
(EEZs) of coastal states (dotted
line) and 200 m isobath are
shown. 1 nautical
mile =1,852 m. Coastline and
bathymetry data are from
GEBCO (IOC, IHO, BODC
2003)
Mar Biol (2009) 156:1089–1107 1091
123
Channel (681 sets), providing information on crabs near the
surface (1,652 non-empty fish stomachs) (Table 1c).
During the YugNIRO program, most of the stomach
contents were analyzed in the field. Some non-empty
stomachs were randomly selected, preserved in 10%
formaldehyde solution and transported to shore for further
processing in the laboratory. In IRD cruises, the sampled
stomachs were preserved frozen and processed in the lab-
oratory (Potier et al. 2007a).
Predator classification
Schools of tunas and tuna-like species at the surface are
exploited by purse seine fishery, while longliners target
individual fish in smaller loose shoals at greater depths.
Considerable feeding partitioning between fishes caught by
purse seines (surface predators) and longlines (subsurface
predators) has been demonstrated in several studies
(Bashmakov et al. 1991; Potier et al. 2004,2008) in spite of
active vertical migratory behavior of tuna (Dagorn et al.
2000; Schaefer et al. 2007). Surface predators often
aggregate in big schools seeking out swarms of favoured
prey in the upper ocean layer (Me
´nard and Marchal 2003),
therefore sharing the same feeding history. Subsurface
predators usually prey on individual targets, therefore
exhibiting more individualized feeding behavior. We used
the occurrence of C. smithii in the stomachs of surface and
subsurface predators as additional information for analyz-
ing the vertical distribution of crabs in the water column.
Data analysis
Identification
Most of the 200 species of Indo-Pacific portunid crabs are
mostly shallow-water inhabitants, but some species are
Table 1 List of the data used in the present study
(a) Pelagic (midwater) trawls
Data type, origin No. of
cruises
Period and year
of sampling
Trawl hauls/hours
of trawling
Total catch
(kg)
Catch of
C. smithii
(kg)
Trawl hauls
with C. smithii
(%)
Research cruises,
YugNIRO, Ukraine
90
27
a
1976–2001
1976–1990
a
7,217/11,286
1,043/1,280
a
19,281,613
398,089
a
658
648
a
2.9
18.1
a
Research cruises, IRD,
France,
R/V ‘La Curieuse’
2 July–October 2002 41/21 77.7 1.5 19.5
(b) Pelagic longlines
Data type, origin No. of
cruises
Period and year
of sampling
No. of longline
stations/hooks
Sampled stomachs/
non-empty stomachs
Non-empty
stomachs
with crabs (%)
Predator’ species
(taxa) sampled/
preyed on crabs
SIOTLLRP
b
, YugNIRO 111 1961–1989 4,700/2,054,722 55,753/29,196
1,349
c
12.7 123/30
Research cruises, IRD,
France, R/V ‘Amitie
´’,
by SFA, Seychelles
9 October 2001–
July 2003
105/*44,700 791/669
c
12.4 24/7
(c) Purse seines
Data type, origin No. of
cruises
Period and year
of sampling
No. of PS
sets sampled
Sampled
stomachs/non-empty
stomachs
Non-empty
stomachs with
crabs (%)
Predator’ species
(taxa) sampled/
preyed on crabs
Scientific observers, Soviet
commercial
fisheries, YugNIRO
25 1984–1992 612 23,403/1,208
195
c
3.0 30/4
Scientific observers, French
commercial
fisheries, IRD
7 October 2001–
January 2007
69 698/444
c
4.7 22/6
Sampling periods, areas and seasons, data stratification and origin are presented by fishing gear and research program
a
Operations within western equatorial area, Arabian Sea and Gulf of Aden
b
SIOTLLRP—Soviet Indian Ocean tuna longline research program; list of vessels and cruises available in Romanov et al. (2006)
c
Laboratory-processed samples
1092 Mar Biol (2009) 156:1089–1107
123
found on the continental slope and on deepwater sea-
mounts. Their taxonomy is still being revised and
identification to the species level is complicated in the field
(Stephenson 1972; Neumann and Spiridonov 1999; Spiri-
donov and Tu
¨rkay 2001).
In YugNIRO cruises, swimming crabs recovered from
predators’ stomachs as well as trawls were usually identi-
fied in the field to family level, but rarely to species level.
Identifications to the species level were done in laboratory-
processed stomachs (3,309 individual crabs) and during
several pelagic and bottom trawl surveys carried out in the
Gulf of Aden.
In the pelagic zone of the Indian Ocean, only one spe-
cies (C. smithii) was identified both in YugNIRO and IRD
samples and reported to date in the literature (Zamorov
et al. 1991a; van Couwelaar et al. 1997; Balasubramanian
and Suseelan 2001; Potier et al. 2004,2007a,b). Similarly,
the only portunid species recorded in the pelagic and bot-
tom trawls of the Gulf of Aden was identified as C. smithii
(Yu.V. Korzun, personal communication
4
). However, to
ensure that our records belong to the species under con-
sideration, only ‘‘true pelagic operations’’, i.e. midwater
trawl hauls beyond the 1,000 m isobath were used in this
study.
Trawl surveys
To investigate the swimming crab’s distribution and to
estimate its abundance, trawl catches were adjusted to the
filtered volume Vcalculated as
V¼DS;
where Dis the distance covered by the trawl (m) and Sis
the trawl mouth area (m
2
).
Dis estimated from the exact positions of the start and
the end of each haul
D¼60 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðLat1Lat2Þ2þðLon1Lon2Þ2cos2ð0:5ðLat1Lat2ÞÞ
q
Sis estimated assuming an elliptic trawl mouth shape,
S=p0.5h0.5w, with hand wthe theoretical vertical and
horizontal trawl openings, respectively.
The index of abundance Awas expressed both in kg and
in number per 1,000 m
3
as
A¼qCV11000
where Cis the catch in kg or in number, respectively, and q
is the catchability coefficient assumed to be equal 1.
Our observations and earlier studies demonstrate that
crab distribution is very patchy (Della Croce and Holthuis
1965; Losse 1969; van Couwelaar et al. 1997). Crab
abundance in the patches and its importance for top pre-
dators were the primary goals of this study. Therefore, we
estimated crab abundance and biomass from crab-positive
hauls only, and we treated hauls without crab as indicators
of a lack of crab aggregations.
Abundance and vertical distribution of crabs were ana-
lyzed separately for ‘day’ (07:00–15:59 local time),
‘sunset’ (16:00–18:59), ‘night’ (19:00–04:59) and ‘sunrise’
samples (05:00–06:59).
Night hauls in the upper 150 m layer yielded 80% of the
crab total catch. Hence, abundance estimates per area
(km
-2
) were calculated for 0–150 m depth.
Maps used data aggregated by 1°trapeziums of latitude
and longitude.
Stomach contents
All sampled predators were measured with precision to
1 cm, weighed and dissected onboard. Fork length (FL)
was measured for all species except bathoids (disk width,
DW) and billfishes (lower jaw-fork length, LJFL in
YugNIRO program and eye-fork length, EFL in IRD
sampling).
Crabs recovered from the stomach contents and trawl
catches were measured and weighed. Total carapace width
(CW), total carapace length (CL), anterior-lateral length
(ALL), posterior-lateral length (PLL), length of the manus
of the cheliped (LM), length of the dactylus of the natatory
leg (DL) were measured to 0.1 mm. Fresh specimens
caught by trawls were weighed (to 0.01 g). For individuals
recovered from fish stomachs, reconstituted weight was
estimated based on our own allometric equation calculated
from morphometric measurements of trawl-caught crabs:
TW ¼0:1289 103CW3:0646;n¼102;
R2¼0:8996;p\0:001
Sex was recorded and females were checked for the
presence of eggs.
The CW was used to estimate the size frequency and the
predator–prey size ratios. When necessary, CW was esti-
mated from our allometric equations (all variables are in
mm):
CW ¼3:3261þ3:1455 DL;n¼150;R2¼0:9105;p\0:001
CW ¼4:3836þ1:5671CL;n¼70;R2¼0:9757;p\0:001
The importance of C. smithii was investigated through
two indices: (a) frequency of occurrence (%F =number
of stomachs with crabs divided by the total number of
non-empty stomachs) and (b) the reconstituted weight
(%RW =total reconstituted weight of C. smithii divided
by the total reconstituted weight of prey items). We
classified swimming crabs as the ‘‘main prey’’ when it
4
Korzun (2006) Personal communication. YugNIRO, 2, Sverdlov
Str., 98300, Kerch, Crimea, Ukraine. e-mail: korzuny@mail.ru.
Mar Biol (2009) 156:1089–1107 1093
123
ranked first in %RW among other individual prey species/
taxa. It was classified as ‘‘important secondary prey’’ when
ranked second or third.
Information on the number of crabs consumed by each
individual fish was absent for the stomachs processed in the
field. Therefore, we used the cumulative occurrence of
C. smithii in the stomachs to map its distribution. %F of
crabs in the 1°cells was analyzed for strata with 10 or more
non-empty stomachs.
Every stomach was classified according to the season in
which it was recovered, i.e. north-east monsoon (NEM,
December–March), spring inter-monsoon (SIM, April–
May), south-west monsoon (SWM, June–August), and
autumn inter-monsoon (AIM, September–November).
Because of the seasonality of crab occurrence, long-term
variability was analyzed by pooling samples recorded
during the ‘‘crab season’’, i.e. from June of year nto
February of year n?1.
Monsoon winds during SWM generate upwelling along
the coasts of Somalia and Arabia and offshore advection
(Schott and McCreary 2001). A proxy of upwelling
intensity was computed by averaging scalar wind strength
from June to August in the area 0°–15°N, 50°–60°E. Data
came from GAO (Marsac 2008)
5
and ICOADS
6
(Worley
et al. 2005).
Mann–Whitney, Kruskal–Wallis and post hoc multiple
comparison tests were performed to take into account the
non-normality of the data. Average values are arithmetic
mean ±standard error (SE).
Results
Spatial and temporal distribution
Geographic area
Most of the pelagic C. smithii were observed in the tropical
western Indian Ocean from the Arabian Sea to 7°S,
and from the eastern coast of Africa to the Maldives
archipelago (Fig. 1b). Southward from 7°S, the only crab
‘‘hotspot’’ is found around Saya-de-Malha bank. Eastward
from the Maldives, C. smithii is occasionally recorded in
the open ocean within a latitudinal band from 0°to 5°N
eastward to approximately 90°E. Few crabs are observed in
the Mozambique Channel or in oceanic waters off Java and
Sumatra.
Both stomach and trawl records show patchy C. smithii
distributions: 1°cells with high crab abundance are com-
monly bordered by cells with zero crab occurrence
(Fig. 1b, c). The highest %F in the stomachs of predators
(50% or more) was observed within several ‘‘hotspots’’
along the equator: the Travin Bank seamount (00°260N,
56°000E; 187 m depth), in the Exclusive Economic Zone
(EEZ) of the Seychelles, in the high seas westward and
eastward from the Seychelles EEZ, and off the coast of
India. Highest trawl catches ([0.1 ind. 1,000 m
-3
) were
recorded along the 200-mile EEZs of Kenya, Somali,
Yemen and Oman, in the Seychelles EEZ and eastward
from the Seychelles.
Within its principal distribution area (western Indian
Ocean northward from 7°S), C. smithii occurred in 18.1/
19.5% of midwater trawl hauls and represented 1.9/0.2% of
total catch (in wet weight) for IRD and YugNIRO samples,
respectively (Table 1a). Catch of this species in the ‘‘crab-
positive hauls’’ varied from 0.1 to 41.3 kg, i.e. from a
single specimen to 19,538 individuals. Maximum recorded
biomass was 798.9 910
-5
kg 1,000 m
-3
eastward from
the Seychelles (NEM, mean weight of crabs 24.0 g) and
Table 2 Estimates of abundance (mean ±SE) of Charybdis smithii in the pelagic zone of the western and eastern Indian Ocean
Data origin Gear Western Indian Ocean Eastern Indian Ocean
ind. h
-1
A
b
A
n
ind. h
-1
A
b
A
n
This study
(crab-positive hauls)
Commercial midwater
trawls
383.9 ±146.8 52.6 ±8.3 0.041 ±0.009 28.2 ±5.3 8.1 ±1.7 0.005 ±0.001
IYGPT 34.9 ±13.3 50.2 ±17.7 0.046 ±0.018 – – –
Both gears – 52.5 ±8.0 0.042 ±0.009 – – –
van Couwelaar et al. (1997) RMT, MOCNESS – – 4.7–15.6
a
–––
Balasubramanian and
Suseelan (2001)
Midwater trawl 393 – – 10 – –
IKMT 11 – – – – –
A
b
abundance as biomass (910
-5
kg 1,000 m
-3
), A
n
abundance in numbers (ind. 1,000 m
-3
)
a
For van Couwelaar et al. (1997), maximum recorded densities are given
5
Data source: Center for Ocean-Atmospheric Prediction Studies,
The Florida State University, http://www.coaps.fsu.edu/woce/html/
ndnwinds.htm.
6
ICOADS data provided by the NOAA/OAR/ESRL PSD, Boulder,
CO, USA, from their web site at http://www.cdc.noaa.gov/.
1094 Mar Biol (2009) 156:1089–1107
123
maximum density was 1.19 ind. 1,000 m
-3
off Yemen
EZZ (SWM, mean weight of crabs 4.0 g), over the northern
part of the sub-marine Carlsberg Ridge. The mean biomass
and density were (52.5 ±8.0) 910
-5
kg 1,000 m
-3
and
0.042 ±0.009 ind. 1,000 m
-3
, respectively (Table 2). In
the eastern Indian Ocean, crab abundance was six to eight
times lower (Table 2).
Most of the crab catches occurred at night: 72% in
YugNIRO and 93% in IRD expeditions. Mean density
was 0.053 ±0.014 ind. 1,000 m
-3
at night and 0.011 ±
0.003 ind. 1,000 m
-3
during daytime (Table 3). The
difference in abundance between night and day was
significant both in biomass and in number (Mann–Whitney
Utest, p\0.01) (Table 3). Abundance estimated for
the two periods of data collection (YugNIRO and IRD) is
very close (Tables 2,3) (no significant differences both
for biomass and numbers; p=0.236 and p=0.113,
respectively).
In the upper 150 m layer, mean abundance in the pat-
ches was estimated at 134 kg km
-2
for biomass and
7,286 ind. km
-2
for numbers.
Vertical distribution
Most of crab catch (95.7%) was obtained from the water
layer down to a 150 m depth (Fig. 2). Crabs were not
recorded deeper than 350 m, except one haul (400–450 m)
close to the Travin Bank. The highest abundance was
observed within the upper 100 m (Fig. 2).
Subsurface predators with crabs in their stomachs were
caught from 50 to 400 m.
The diurnal vertical distribution pattern was investigated
using trawl survey data. Overall fishing effort (Fig. 2) and
fishing operations during the night and twilight (Fig. 3) are
concentrated in the upper 0–150 m. However, the number
of fishing operations below 150 m during daytime (72
hauls) and at night or twilight (49) lies within a similar
order of magnitude, allowing us to detect vertical move-
ment of crab aggregations.
Trawl catches indicate most likely diel migration of
C. smithii (Fig. 3). From late afternoon (16:00) and through
the night, swimming crab aggregated in the upper water
layer, usually not deeper than 150 m. At dawn, it was
caught from 50 to 250 m depth. During the day, individuals
were scattered in the water column from the upper layer to
350 m depth (one positive haul was obtained at 400–450 m
depth) with a peak of abundance at 50–150 and 300–350 m
depths. Crab aggregations were never observed in the deep
layers during the night either by trawling or by acoustic
means.
Seasonal distribution
In the trawl catches, C. smithii was recorded all year round,
except August. The overall sampling effort in August–
October was low (no data in September, two hauls in
August and three in October). Therefore, we did not con-
sider trawl data for this period as representative.
The monthly occurrence of crabs was high during NEM
Table 3 Catches of Charybdis smithii (mean ±SE) for night and day hauls and significance level of nonparametric Mann–Whitney Utest
Period YugNIRO IRD Both sources
Crab-positive
hauls
A
b
A
n
Crab-positive
hauls
A
b
A
n
Crab-positive
hauls
A
b
A
n
Day 34 27.1 ±12.0 0.011 ±0.003 2 9.3 0.008 36 26.1 ±11.3 0.011 ±0.003
Night 125 62.6 ±11.3 0.053 ±0.014 5 61.8 ±25.3 0.056 ±0.026 130 62.6 ±10.9 0.053 ±0.013
pvalue \0.01 \0.01 Not estimated \0.01 \0.01
Abundance indexes abbreviations and units as in Table 2
Effort, %
50
0
50
100
150
200
250
300
350
400
450
500
800
Depth, m
025 0 0.1 0.4 0.8 1.2
ind. 1,000 m-3
Fig. 2 Distribution of sampling effort and catch by depth strata. Left
panel: relative fishing effort (in % of total filtered volume,
12.4 910
9
m
3
) by midwater trawls in the pelagic zone of the Indian
Ocean; right panel: abundance, ind. 1,000 m
-3
, positive sets only.
Dots are means, boxes are standard errors (SE), and whiskers are
ranges
Mar Biol (2009) 156:1089–1107 1095
123
(18–74% of trawl sets) and lower during SWM (12–35%).
However, the highest abundance was observed during SWM:
mean monthly abundance reached 0.19 ind. 1,000 m
-3
(July). It steadily decreased from November to May to
0.00003 ind. 1,000 m
-3
(Fig. 4).
The same trend was observed in the stomach contents of
three subsurface predators (yellowfin and bigeye tunas and
lancetfish). They preyed on C. smithii all year round and all
three species showed a synchronous seasonal cycle
(Fig. 4). With the onset of the SWM, crab occurrence
sharply increased. The highest crab occurrence in
predators’ stomachs was observed in August–November
(late SWM and AIM). During NEM, the crab occurrence
gradually decreased. From March to May (SIM) the
swimming crab is rarely prey for yellowfin tuna and
lancetfish (less than 5% in %F) and is absent in the
stomachs of bigeye tuna. Crabs were rarely observed in the
stomachs of surface predators (mainly yellowfin tuna;
Table 4) from June to December (SWM–AIM).
The peak of C. smithii occurrence is 1.5–2 months after
the peak of summer monsoon winds and upwelling
(Fig. 4). The onset of coastal upwelling and the offshore
Day
07:00-16:00
Sunset
16:00-19:00
Night
19:00-05:00
Sunrise
05:00-07:00
Filtered volume,
% of total 2.8×10
9
m
3
0
50
100
150
200
250
300
350
400
450
500
800
Depth, m
Filtered volume, Filtered volume, Filtered volume,
0 0.05 0.1
0
50
100
150
200
250
300
350
400
450
500
800
Depth, m
0 0.05 0.1
ind. 1,000 m-3
0 20 40 0 20 40 0 20 40 0 20 40
00.1 0.4 0.8 1.2 0 0.05 0.1
ind. 1,000 m-3 ind. 1,000 m-3 ind. 1,000 m-3
% of total 1.6×10
9
m
3
% of total 7.1×10
9
m
3
% of total 0.9×10
9
m
3
Fig. 3 Diurnal vertical distribution of relative fishing effort (upper
panel) in % of total values and abundance of Charybdis smithii (lower
panel), ind. 1,000 m
-3
, in the midwater trawls in the pelagic zone of
the western Indian Ocean. Note:910 scale magnitude for the night-
time abundances. Dots are means, boxes are standard errors (SE), and
whiskers are ranges
1096 Mar Biol (2009) 156:1089–1107
123
advection during SWM correspond to an increasing trend
of crab occurrence in predators’ stomachs and of its
abundance in trawl catches.
Long-term trend
The annual occurrence of C. smithii in the stomachs of
yellowfin and bigeye tunas and lancetfish exhibited a
strong inter-annual variability (Fig. 5). Although spatial
and temporal irregularity in the sampling effort may affect
the results, at least one period of high crab occurrence from
SWM 1978 until NEM 1987–1988 and one period of low
occurrence (from 1970 till 1978) can be distinguished.
Sampling during 1963–1969 and 2001–2004 produced
inconsistent indicators of crab occurrence for both species
of tunas, while analysis of lancetfish stomachs showed high
crab occurrence.
Contribution to the diet of large pelagic fish
The swimming crab was recovered from the stomachs of
34 predatory species; 30 predator species caught by long-
line and 8 predators caught by purse seines (Table 4). Most
of the crab predators were abundant epipelagic oceanic
species; four (Seriola rivoliana,Lobotes surinamensis,
Lethrinus harak and Lagocephalus sceleratus) were
demersal or reef-associated species.
Both in archival data and in recent sampling, C. smithii
was frequently recorded in the stomachs of four subsurface
predators: yellowfin tuna (%F =9.3–14.4%), bigeye tuna
(6.4–14.3%), longnose lancetfish (23.5–45.2%), and pela-
gic stingray (Pteroplatytrygon violacea) (16.7–27.1%)
(Table 4). It was a common prey of silky shark (Carcha-
rhinus falciformis) (28.2%), blue shark (Prionace glauca)
(16.6%), dolphinfish (Coryphaena hippurus) (7.0%),
unidentified carcharhinid sharks (genus Carcharhinus)
(6.8%) and great barracuda (Sphyraena barracuda) (6.4%)
in the YugNIRO data. It was rare (%F equal 3.3% or less)
in the stomachs of skipjack (Katsuwonus pelamis) and
albacore (Thunnus alalunga) tunas, in billfishes (Makaira
spp., Istiophorus platypterus and Xiphias gladius), silvertip
sharks (Carcharhinus albimarginatus), oceanic white-tip
shark (Carcharhinus longimanus), sandbar shark (Car-
charhinus plumbeus), mako sharks (Isurus spp.), and
wahoo (Acanthocybium solandri) (Table 4).
Among surface predators, crab was a rare prey item. The
highest occurrence was recorded in the stomachs of yel-
lowfin tuna (2.7–7.2%), silky shark (11.8%),
7
rainbow
runner (Elagatis bipinnulata) (12.5–25.0) (see footnote 7)
and spotted oceanic triggerfish (Canthidermis maculata)
(up to 100%) (see footnote 7), while only one specimen
was recorded for skipjack and albacore tuna. Bigeye tuna
and barracuda caught at the surface had no crabs in their
stomachs, although they commonly preyed on crabs at
greater depth. C. smithii was only observed in the stomachs
of surface predators caught around natural flotsam or fish
aggregating devices (FAD). A sole individual was recov-
ered from yellowfin tuna in a free-swimming school caught
above Travin Bank.
IMI, ms-1
0.0
0.4
0.8
1.2
ind. 1,000 m-3
-10
-5
0
5
10
NEM SWMSIM AIM
YFT
1500
544
695
838
669
640
772
552
904
667 953
1465
0
10
20
30
40
50
60
70
BET
184
64
82
112
324 380 434 283
168 129
295
108
0
10
20
30
40
50
60
70
ALX
38
81
57
43 77
61 10 53 84
99
53
71
0
10
20
30
40
50
60
70
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Frequency of occurrence, %
TRAW
Fig. 4 Annual cycle of Indian monsoon circulation index (IMI) and
seasonal variability in Charybdis smithii abundance. Upper panel:
IMI (open circles) and monthly abundance. Other panels shows
occurrence (%F) of C. smithii in the stomachs of three principal
predators: T. albacares (YFT), T. obesus (BET), A. ferox (ALX). IMI
is in ms
-1
; abundance is ind. 1,000 m
-3
(night positive hauls only).
Dots are means, boxes are standard errors (SE), and whiskers are
ranges. Sample sizes (non-empty stomach) are given in numbers for
each month in the figure. Shaded areas correspond to monsoon
periods of the Indian Ocean: north-eastern monsoon (NEM), south-
western monsoon (SWM); intermonsoon periods are non-shaded:
boreal spring intermonsoon (SIM) and boreal autumn intermonsoon
(AIM). IMI values are adapted from Cherchi et al. (2007)
7
Sample size is low, n\30 ind. (Table 4).
Mar Biol (2009) 156:1089–1107 1097
123
Table 4 List of top predators preying on Charybdis smithii in the western tropical Indian Ocean grouped by sampling gear and program
No Species/taxa Pelagic longlines Purse seines
YugNIRO IRD YugNIRO IRD
n%F Mean size
(range)
n%F Mean size
(range)
n%F Mean size
(range)
n%F Mean size
(range)
1Alopias superciliosus 28 3.6 256 1 0.0
2Alopias vulpinus 57 1.8 176
3Isurus oxyrinchus 205 2.4 187.5 (156–223)
4Isurus spp. 15 6.7 139.5 (130–149)
5Carcharhinus albimarginatus 596 0.2 141 1 0.0
6Carcharhinus falciformis 110 28.2 137.9 (70–208) 3 0.0 4 0.0 17 11.8
7Carcharhinus longimanus 82 2.4 128.5 (120–137) 1 0.0 1 0.0
8Carcharhinus plumbeus 304 0.7 169.0 (165–173)
9Carcharhinus spp. 1085 6.8 170.4 (81–305) 2 0.0
10 Galeocerdo cuvier 266 3.4 208.6 (122–310) 1 0.0
11 Prionace glauca 145 16.6 221.1 (185–268) 1 0.0
12 Pteroplatytrygon violacea
a
59 27.1 53.7 (42–63) 6 16.7 59.3 (54–62)
13 Alepisaurus ferox 638 23.5 125.2 (54–175)139 45.3 74–163 (138.2)
14 Sphyraena barracuda 65 6.2 117.3 (100–144) 2 0.0 3 0.0 4 0.0
15 Caranx sexfasciatus 1 100.0 117.5
16 Elagatis bipinnulata 3 0.0 8 12.5 75 28 25.0 87
17 Seriola rivoliana 7 14.3
18 Coryphaena hippurus 71 7.0 109.4 (91–120) 13 0.0 2 0.0 20 0.0
19 Brama spp. 2 50.0 83
20 Lobotes surinamensis 4 25.0 39
21 Lethrinus harak 2 50.0 67
22 Lepidocybium flavobrunneum 6 16.7 71
23 Acanthocybium solandri 58 1.7 150 3 0.0 1 0.0 14 0.0
24 Katsuwonus pelamis 257 2.7 62.6 (50–71) 4 0.0 628 0.2 66 81 0.0
25 Thunnus alalunga 409 2.2 98.2 (59.5–147) 10 10.0 90 4 0.0
26 Thunnus albacares 11,993 14.4 123.8 (48–173)118 9.3 138.8 (116–156)391 7.2 82.1 (60–133)186 2.7 41
27 Thunnus obesus 2,990 6.4 131.7 (60–193) 21 14.3 131.0 (110–138) 48 0.0 43 0.0
28 Xiphias gladius
b
121 3.3 150.7 (119–175)135 1.5 119.3 (118–120)
c
29 Istiophorus platypterus
b
412 0.7 186 16 0.0 1 0.0
30 Makaira indica
b
45 2.2 164 1 0.0 2 0.0
31 Makaira spp.
c
144 1.4 164 4 0.0 1 0.0
32 Tetrapturus audax
b
94 3.2 195 1 0.0
1098 Mar Biol (2009) 156:1089–1107
123
Overall importance as prey for tunas and lancetfish
Laboratory-processed samples allowed us to estimate the
importance by reconstituted weight (RW) of swimming
crab in the diet of yellowfin, bigeye tuna and lancetfish
(Table 5). Samples were pooled into the four periods (with
respect to principle processor of the samples): 1970–1978
(corresponds to phase of low crab occurrence), 1979–1981
and 1986–1987 (phase of high crab occurrence), and 2001–
2004 (unclear trend in occurrence). During 1986–1987,
C. smithii constituted more than 50% of the stomach
content in RW for lancetfish and yellowfin tuna. A similar
quantity was found again during the 2001–2004 period for
YFT
1
42
71
6
57
1
21
241
5914
2154
35
103
164
724 502
525 441
72
292
68 246 95
32
900
406
0
20
40
60
80
100
BET
4
413
5
20
228
104
149
24
715
44
50 45
105
18
86
76
151 82
143
93
0
20
40
60
80
100
ALX
13
45
78
2
75
4
63
5
3
47 16
49 62
55
1
246
87
13
0
20
40
60
80
100
1961
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
2001
2003
2005
Frequency of occurrence, %
No sampling in
1990-2000
SWM scalar wind anomalies
-2.5
-1.5
-0.5
0.5
1.5
ms
-1
2.5
Fig. 5 Long-term climatic variability in the north-western Indian
Ocean and abundance of Charybdis smithii.Upper panel shows scalar
wind anomalies during south-west monsoon (SWM) in the area 0°–
15°N, 50°–60°E calculated from GAO 2008 (columns) and ICOADS
(black line). Other panels shows seasonal (June–February) occurrence
(%F) of C. smithii in the stomachs of three principal predators:
T. albacares (YFT), T. obesus (BET), and A. ferox (ALX). Sample
sizes (non-empty stomach) are given in numbers for each year
Table 4 continued
No Species/taxa Pelagic longlines Purse seines
YugNIRO IRD YugNIRO IRD
n%F Mean size
(range)
n%F Mean size
(range)
n%F Mean size
(range)
n%F Mean size
(range)
33 Canthidermis maculata 5 100.0
34 Lagocephalus sceleratus 2 50.0 55
No. of species (taxa)
preyed on crabs
30 5 4 6
No. of species (taxa)
sampled for stomach
content
122 25 30 22
Sample sizes (non-empty stomachs, n), frequency of occurrence of crabs (%F) in the stomachs, and size of predators with crabs in the stomachs (range and mean) are given. Figures for samples
with more than 30 stomachs analyzed are in bold
Size of predators with crabs in the stomachs is fork length (FL) except:
a
disk width (DW),
b
lower jaw-fork length (LJFL) and
c
eye-fork length (EFL)
Mar Biol (2009) 156:1089–1107 1099
123
lancetfish (53%). During these periods, the contribution of
C. smithii to the diet was equal or higher than the contri-
bution of composite prey groups, such as fishes and
cephalopods (Table 5). During all periods of study,
swimming crabs were the main prey of yellowfin tuna and
lancetfish. For bigeye tuna, it was the main prey in 1970–
1978, an important secondary prey in 1986–1987, but had
low importance in 2001–2004.
Biological aspects
Size and predator–prey size ratio
The CW of C. smithii varied from 23 to 70 mm (mean
45.1 ±0.3 mm) (Fig. 6a). The individuals recovered from
the stomachs of yellowfin, bigeye tuna and lancetfish, and
from trawl catches had similar size ranges and medians
(Fig. 6b).
The length of predators that preyed on swimming crab
varied from 55 to 310 cm (Table 4). Observed crab–pred-
ator size ratio was very low: 0.01–0.09.
Seasonal size variability
The monthly weight variations of C. smithii were used in
analysis of the seasonal variability of crab sizes because of
the lack of CW data in YugNIRO trawl samples. Small crabs
(0.7–3.8 g, mean weight 1.2 g; estimated CW 15–25 mm)
were recorded in June (Fig. 7). In July, individuals with a
mean weight of 11.4–11.7 g (estimated CW 41 mm) were
recovered from both predator stomachs and trawls. The mean
Table 5 Importance of the Charybdis smithii and various prey groups for three top predators, Thunnus albacares (YFT), Thunnus obesus (BET)
and Alepisaurus ferox (ALX), estimated from laboratory-processed stomachs
Period Samples processor Predator nNo. of prey species/taxa Prey groups
PCR Cr F Ce
1970–1978 Kornilova GN YFT 43 106 24.3 (1) 2.2 56.3 16.4
BET 23 30 27.7 (1) 3.5 54.4 14.4
ALX
1979–1981 Kornilova GN YFT 46 84 29.9 (1) 2.3 55.7 11.9
BET
ALX
1986–1987 Zamorov VV YFT 173 48 51.3 (1) 0.6 31.0 17.1
BET 78 45 17.9 (3) 2.4 69.4 10.3
ALX 38 20 79.7 (1) 0.8 13.6 5.8
2001–2004 Potier M YFT 114 59 28.0 (1) 4.9 49.5 17.6
BET 21 25 2.0 (9) 0.2 77.1 20.7
ALX 136 58 52.7 (1) 18.0 21.1 6.6
Prey groups are as follows: C. smithii (PCR), crustaceans (Cr), fish (F), and cephalopods (Ce). Importance is expressed in percentage of
reconstituted weight, %RW. Rank of swimming crab in %RW among individual prey species/taxa is presented in parentheses. nis sample size,
non-empty stomachs. Numbers of prey species/taxa identified in the stomach content are given
47.6
47.0
43.8
41.5
47.6
47.0
43.8
41.5
15 20 25 30 35 40 45 50 55 60 65 70 75
Carapace width, mm
YFT
BET
ALX
Trawls B
Carapace width, mm
0
5
10
15
20
25
30
Percentage
A
15 20 25 30 35 40 45 50 55 60 65 70 75
Fig. 6 Size of Charybdis smithii sampled in the western Indian
Ocean (carapace width, CW). aPooled size frequency distribution
from all sampling gears (n=722). bSize of crabs recovered from the
stomachs of YFT, BET, ALX and trawls in the western Indian Ocean.
Dots are means, boxes are standard errors (SE), and whiskers are
ranges
1100 Mar Biol (2009) 156:1089–1107
123
crab weight increased from August to May and reached 40 g
(CW *60 mm) in May. It should be noted that the mean
weight of crabs in November was lower than in October
(stomach samples); a slow increase was recorded in
November and December. We attributed this fact to multiple
recruitment events during SWM and appearance of new
cohorts in the samples. Small individuals (\5 g) that were
recorded during SWM were absent in the pelagic waters
during other seasons. Some big individuals occurred in
October reaching 60 g in weight (CW *70 mm).
Discussion
Geographic distribution
Our results complete considerably past studies of C. smithii
pelagic distributions (van Couwelaar et al. 1997; Bala-
subramanian and Suseelan 2001;Tu
¨rkay and Spiridonov
2006). For the first time, the pelagic distribution of
C. smithii throughout the western Indian Ocean has been
demonstrated from the Arabian Sea and coasts of India to
approximately 7°S. The pelagic occurrence of C. smithii in
the eastern Indian Ocean observed by Balasubramanian
and Suseelan (2001) is confirmed by our records for the
Bay of Bengal and open equatorial waters till 90°E at least.
Therefore, C. smithii cannot be considered as endemic of
the western Indian Ocean (Zamorov et al. 1991b; Spiri-
donov 1994; van Couwelaar et al. 1997; Balasubramanian
and Suseelan 2001). However, most of the crabs are found
in the western basin.
The density recorded in our study was consistent for
the two type of trawls (Tables 2,3) and was similar to the
results obtained by Balasubramanian and Suseelan (2001)
both in the Arabian Sea and the eastern Indian Ocean
(Table 2). These studies showed lower densities of
C. smithii in the eastern part of the ocean. The highest
density of crabs observed during this study was five to ten
times lower than maximum densities reported by van
Couwelaar et al. (1997) for the north Arabian Sea during
SWM. We attribute this difference to patchiness of the
crab distribution, the proximity of van Couwelaar et al.
(1997) sampling efforts to upwelling cells, and to natural
variability.
Visual observations (ER, personal observations; Della
Croce and Holthuis 1965; Losse 1969) suggest that patches
of C. smithii exceed tens of kilometers. Hence, our esti-
mates of abundance within the patches provide an
important insight on the scale of crab invasion into the
open ocean.
Early records and even some recent observation of
abundant patches of C. smithii in the pelagic zone of the
western Indian Ocean were considered as an unusual
phenomenon (Sankarankutty and Rangarajan 1962; Della
Croce and Holthuis 1965; Marine Observer, cit. from Losse
1969; Rice 1969; Billett et al. 2006). Our data show that the
pelagic juvenile-adult stage is characteristic of a peculiar
life pattern developed in the monsoon regions of the Indian
Ocean, as previously mentioned by van Couwelaar et al.
(1997) and Balasubramanian and Suseelan (2001).
Pelagic individuals of C. smithii are commonly recorded
in the four biogeographic provinces of Longhurst (1998):
Indian Monsoon Gyres Province (MONS), Northwestern
Arabian Upwelling Province (ARAB) and western and
eastern India Coastal Province (INDW, INDE). It is rare in
the nutrient-poor waters of Indian South Subtropical Gyre
Province (ISSG) beyond the South Tropical Front (STF)—
a natural border between the MONS and the ISSG (Neiman
and Burkov 1989; Longhurst 1998). Therefore, oceanic
currents, offshore advection and productivity play key roles
in the swimming crab’s distribution and life cycle.
The only crab ‘hotspot’ within ISSG is found around the
Saya-de-Malha Bank (Fig. 1b). Increased primary pro-
duction was reported there as a product of the interaction of
the South Equatorial Current (SEC) with the Mascarene
Ridge (Gallienne et al. 2004). Low occurrence of crab
westward from the bank despite strong water transport
by SEC suggests the existence of a settled population of
C. smithii and/or high mortality in the pelagic waters.
Vertical distribution
From our observations, the pelagic habitat of the swimming
crab extends from the mixed layer to upper mesopelagic
waters within a wide range of temperatures and oxygen
concentrations. At night, C. smithii aggregates from the
surface down to a 150-m depth within a temperature range
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
5
10
15
20
25
30
35
Total weight, g
40
45
Fig. 7 Monthly variations in weight of Charybdis smithii sampled in
the pelagic zone of the western Indian Ocean (total weight). Dots are
mean weight from predators stomachs, open diamonds are mean
weight from trawls, whiskers are standard errors (SE). Three outliers
(two in June and one in January) were removed
Mar Biol (2009) 156:1089–1107 1101
123
of 15–30°C. After sunrise, it is found scattered in the water
column, with the highest densities registered both in the
thermocline and below at 300–350 m (10–17°C). van
Couwelaar et al. (1997) reported that C. smithii avoids
areas of low oxygen concentrations. We recorded crabs
during the day below the thermocline in the oxygen min-
imum layer at concentrations as low as 0.62 mL L
-1
.This
is consistent with observations from Indian waters
(Karuppasamy et al. 2007). However, since we used non-
closing gears, these records need further investigations.
Underwater observations from submersibles in the
Arabian Sea reported a similar distribution: swimming
crabs were detected at the surface during twilight and in the
deeper layer down to 200-m depth during the day (Anon-
ymous 1986,1987). During the daytime, some individuals
stayed in the mixed layer (20–50 m).
van Couwelaar et al. (1997) reported crabs in the towed
nets down to 600 m. Similar results (400–450-m depth)
were observed from our data close to the Travin Bank.
However, these results must be regarded with caution due
to limitations in the fishing gears and low sampling reso-
lution (200–300 m) below 300-m depth used by van
Couwelaar et al. (1997).
The diel migration of the swimming crab is similar to
species found in the Sound Scattering Layer (SSL), such as
mesopelagic fish, crustaceans and squids (Herring et al.
1998; Luo et al. 2000; Ashjian et al. 2002). In the late
afternoon, the migration of the swimming crab toward
upper layers starts earlier than other SSL species: it is often
recorded at the surface before sunset (ER, personal obser-
vations; Losse 1969; Rice 1969). With such behavior, C.
smithii can prey on small schooling epipelagic fish during
twilight (Losse 1969) and on mesopelagic migrants (fish
and cephalopods) at night (Balasubramanian and Suseelan
1998; Mincks et al. 2000).
Daily migration to deep layers reduces the vulnerability
of C. smithii for top predators at the surface: fish, cetaceans
and seabirds. Crab is rare in the stomachs of surface pre-
dators and longline-caught fish commonly inhabiting upper
mixed layer: skipjack tuna, marlins, wahoo, oceanic white-
tip shark and mako shark (Table 4). It is also absent in the
diet of seabirds (Ramos 2000; Asseid et al. 2006). Seabirds
readily prey on crustaceans including crabs spotted at the
surface or stranded, which may constitute a major part of
the bird’s diet (Stewart et al. 1984; Munilla 1997; Aurioles-
Gamboa et al. 2003). Indian Ocean stomatopod Natosquilla
investigatoris, which occurs at the surface during daytime,
is a common prey of seabirds (Losse and Merrett 1971;Le
Corre and Jaquemet 2005; Asseid et al. 2006). C. smithii
stay beyond bird diving range during the day (Le Corre
1997; Weimerskirch et al. 2005) and birds seem to be
unable to detect crabs during twilight, i.e. the evening peak
of foraging activity (Weimerskirch et al. 2004,2009).
The abundance of C. smithii in the stomach contents of
subsurface predators is evidence that crabs are vulnerable
in the morning and during the day when they migrate to the
thermocline and deeper layers.
Only FAD-associated surface predators occasionally
have crabs in their stomachs. Anchored and drifting FADs
(natural logs, anthropogenic flotsam or specially con-
structed devices) are widely used by industrial and
recreational fishermen because they attract and aggregate
many pelagic species (Fonteneau et al. 2000; Castro et al.
2002; Taquet et al. 2007). Presence of FADs may change
the behavior of associated species (Taquet et al. 2007;
Hallier and Gaertner 2008). Two possible alterations may
explain occurrences of crabs in the stomachs of surface
predators associated with FADs: night feeding of fish or/and
daytime aggregation of crabs around FADs at the surface.
During an extensive underwater census of a FAD-
associated communities in the tropical western Indian
Ocean (FADIO project, 2003–2006; http://www.fadio.ird.fr),
C. smithii were never recorded during the daytime
(M. Taquet, personal communication
8
). Small fish species
associated with drifting FADs (mainly Carangidae, Balis-
tidae, Monacanthidae and Kyphosidae fam.) are rarely
observed in tuna stomachs (Bashmakov et al. 1991; Potier
et al. 2004). However, night foraging activity on SSL
species that had migrated to the surface was reported for
FAD-associated tuna (Schaefer and Fuller 2005). These
observations suggest that crabs in the stomachs of FAD-
associated fish are rather nocturnal migrants than prey
consumed during the day.
Role in the pelagic ecosystem
The swimming crab C. smithii is the main prey (or
important secondary prey) of yellowfin tuna, bigeye tuna
and lancetfish from SWM to NEM, and supports the pop-
ulations of the most abundant top predators of the western
Indian Ocean. Although data limitation did not allow us to
estimate crab importance for the silky shark, pelagic
stingray and blue shark, the high occurrence of C. smithii in
their stomachs shows that crabs are a common dietary
component for these fishes as well. Therefore, the swim-
ming crab represents an important species of the
intermediate trophic levels in the open ocean ecosystem of
the western Indian Ocean.
Subsurface predators demonstrate remarkable dietary
shift during periods of low crab abundance. During SIM,
fish, purpleback squid Sthenoteuthis oualaniensis or other
crustaceans prevail in the tuna stomachs. In the ISSG
8
Taquet (2008) Personal communication. IFREMER, CRH, Av. Jean
Monnet, BP 171, 34203 Se
`te Cedex, France. e-mail: marc.taquet@
ifremer.fr
1102 Mar Biol (2009) 156:1089–1107
123
waters, epi- and mesopelagic fish are the principal prey for
yellowfin tuna and lancetfish (Zamorov 1993; Romanov
and Zamorov 2007). Longnose lancetfish demonstrate a
high degree of forced cannibalism in the absence
of abundant non-evasive prey such as swimming crab
(Romanov et al. 2008).
The seasonal monsoon upwelling of the northern Indian
Ocean generates high primary production during the short
period of SWM (Smith 1995). Abundant biomass of short-
cyclic mesopelagic species usually not consumed by epi-
pelagic predators is a major result of the upwelling-
generated productivity (Gjøsaeter 1984; Ashjian et al.
2002; Chesalin and Zuyev 2002). The swimming crab
actively preys on macrozooplankton and micronekton
organisms such as mesopelagic fish, cephalopods, euphau-
siid and decapod shrimps (Losse 1969; Balasubramanian
and Suseelan 1998; Mincks et al. 2000) representing a
crucial link in the energy transfer from low- to high-trophic
levels.
Beside C. smithii, few crab species are known as pelagic
invaders: Euphylax dovii,Portunus xantusii,Pleuroncodes
planipes,Pleuroncodes monodon (Pacific Ocean) and
Polybius henslowii (Atlantic Ocean) (Alverson 1963; Jerde
1967,1970; Allen 1968; Longhurst 2004; Gutie
´rrez et al.
2008). All of them inhabit highly productive upwelling
areas and are often transported as far as 1,000 miles off-
shore (Norse and Estevez 1977; Longhurst 2004; Gutie
´rrez
et al. 2008). As with C. smithii, these species represent
major intermediate trophic pathways in their respective
ecosystems (Alverson 1963; Jerde 1967; Sund et al. 1981;
Munilla 1997; Longhurst 2004; Gutie
´rrez et al. 2008).
However, other crab species dominate within narrow zones
of upwelling activity, while high pelagic biomass
([130 kg km
-2
)ofC. smithii is observed in the open
ocean.
Invasion of swimming crab into pelagic waters is not
unique to the western Indian Ocean. Periodic outbursts of
another crustacean of benthic origin, the stomatopod
N. investigatoris, which dominated pelagic crustacean
communities in 1965–1967, 1971–1974, and 2000–2005,
are known (Losse and Merrett 1971; Demidov 1972; Potier
et al. 2007a,b). Both species demonstrate similar features
reaching adult stages in the pelagic waters, but reproducing
only in benthic ecosystems. Pelagic abundance of swim-
ming crab usually decreased during stomatopod outbursts,
but the underlying mechanisms and degree of potential
interaction among the two species are not clear and need
further investigation.
Life cycle and long-term variability
Two hypotheses have been presented to describe the life
cycle of C. smithii in the western Indian Ocean: pelagic
spawning during SWM in order to facilitate the dispersion
of the pelagic larvae (Zamorov et al. 1991b) and benthic
reproduction (van Couwelaar et al. 1997; Balasubramanian
and Suseelan 1998,2001) during NEM on the shelf areas of
the northern Indian Ocean.
Although no observations of natural crab spawning have
been reported, numerous data support the hypothesis of the
benthic reproduction of C. smithii. No ovigerous females
have been recorded in the pelagic environment during
historical and present studies (Daniel and Chakrapany
1983; van Couwelaar et al. 1997; Balasubramanian and
Suseelan 1998). Norse and Fox-Norse (1977) have shown
that the reproduction phase of portunids is linked with their
native benthic environment: these crabs are unable to
attach fertilized eggs to the female’s abdomen independent
of the substratum.
Despite assumed seasonality in the reproduction of
C. smithii, combined benthic occurrences of ovigerous
females and pelagic larvae (Daniel and Chakrapany 1983;
van Couwelaar et al. 1997; Balasubramanian and Suseelan
1998,2001; Thomas and Kurup 2001) did not show a
seasonal pattern.
In contrast, recruitment of the pelagic population and
abundance of the juvenile-adult crabs (Figs. 4,7) is sea-
sonal and apparently monsoon driven. The smallest crabs
(CW 5–6 mm) were observed in the northern Arabian Sea
in mid-May with the onset of the Somali-Arabian upwell-
ing (van Couwelaar et al. 1997). In this study, crabs of 15–
25 mm CW were collected in June 1–1.5 months after the
upwelling onset, while no new recruits were observed
during AIM. Highest crab abundance was recorded in July
and highest occurrence in the stomachs 1.5–2 months after
the peak of upwelling (August–September) (Fig. 4).
We hypothesize that the rapid increase of swimming
crab abundance in open waters is related with the mass
metamorphosis of C. smithii pelagic larvae triggered by
active upwelling. The larvae of portunids are freely dis-
persed by oceanic currents (Sulkin 1984; Pechenik 1999).
They pass several development phases until they reach
metamorphic competence stage (Pechenik 1999; Hadfield
et al. 2001) when they become ready for metamorphosis
into juveniles and settlement. However, as has been shown
for many crustacean species, metamorphosis is triggered by
specific chemical and/or physical cues, mainly associated
with the adult habitat (Forward et al. 2001; Hadfield et al.
2001; Gebauer et al. 2003). In the absence of such cues,
competent larvae can delay their metamorphosis by a few
days up to several months, accumulating in the environ-
ment, or dying if they do not receive the appropriate
metamorphosis stimulus (Pechenik 1999; Hadfield et al.
2001; Gebauer et al. 2003).
The onset of upwelling is marked by intensive pumping
to the surface and offshore transport (up to 750 miles from
Mar Biol (2009) 156:1089–1107 1103
123
the coast; Smith 1995) of colder nutrient-rich subsurface
waters. These waters have specific chemical properties
related to the adult benthic habitat of C. smithii.We
postulate that these waters act as a trigger of mass
metamorphosis of competent larvae of C. smithii drifting
along upwelling fronts, eddies and meanders of the Somali
Current (SC) and in upwelling cells off the Arabian and
Malabar coasts during SWM. Metamorphosis takes place
beyond the shelf and slope within water masses transported
offshore. Juvenile crabs cannot reach the bottom, so are
unable to settle, and drift to the open ocean within gyres of
SC and surface currents of the western Indian Ocean.
Offshore transport of upwelling planktonic community by
monsoon-generated eddies has been reported in the region
(Ashjian et al. 2002).
The degradation of monsoon winds and upwelling in
September results in the collapse of offshore transport and
the interruption of pelagic recruitment.
The circular surface currents developed during SWM in
the western Indian Ocean (Schott and McCreary 2001)
apparently facilitate retention of pelagic crabs within large-
scale oceanic gyres, preventing their return transport to
shelf waters. Reversal of oceanic circulation with the onset
of NEM seems to play the same role, transporting growing
crabs into the South Equatorial Contercurrent where they
drift along the equator and the Seychelles bank toward the
Maldives Archipelago.
C. smithii abundance in the pelagic realm may exceed
15,000 ind. km
-2
in July (i.e. the peak of the south-west
monsoon), decreases during AIM and NEM, and declines
by 50-fold in March. Swimming crabs almost disappear in
April–May (SIM) due to predation, natural mortality
(Christiansen and Boetius 2000) and settlement for physi-
ological maturation and reproduction. Despite an active
vertical migratory behavior, the swimming crab is con-
sidered here as a passive drifter within an ocean-wide
scale. We believe that only a minor part of the pelagic
stock is advected to shelf areas or oceanic seamounts.
Therefore, the fast decline in abundance shows that the
survival rate in the hostile pelagic environment is low. The
presence of a few large crabs (up to 65–70 mm CW) in
July (1 ind. observed) and October (2 ind.) suggests than
some individuals can spend more than 1 year in pelagic
waters.
Wind forcing of long-term fluctuations in pelagic pro-
ductivity of upwelling systems has been known for years
(Goes et al. 2005; Rykaczewski and Checkley 2008).
Positive monsoon wind strength anomalies in the western
Indian Ocean (1970–2006 derived from GAO) correspond
to higher crab occurrence and are consistent with our
hypothesis of pelagic recruitment (Fig. 5). However, major
discrepancies in the climatic data series starting from the
mid-1990s, and gaps in biological data (Fig. 5), complicate
analyses of long-term trends in the Indian Ocean pelagic
communities. Pelagic productivity, survival rates of
planktonic larvae of C. smithii and interspecific interactions
may also play an important role in the variability of
swimming crab abundance in the western Indian Ocean.
Our study shows that C. smithii is an abundant pelagic
species of the monsoon region of the Indian Ocean, rep-
resenting an important trophic link in the epipelagic
ecosystem. Its pelagic abundance and recruitment are dri-
ven by contrasting monsoon regimes and upwelling, while
other factors affecting its annual variability in abundance
are still unknown. Further studies of the crab’s biological
cycle and pelagic dispersion (both at the larvae and juve-
nile-adult stages) involving sampling experiments and
ocean circulation modeling are needed to better understand
the biology of this important species and the interconnec-
tivity between its benthic and pelagic populations.
Acknowledgments This work, a part of the THETIS program of the
IRD (Institut de Recherche pour le De
´veloppement), is also supported
by the BIOPS project funded by Institut Franc¸ais de la Biodiversite
´
(no. CD-AOOI-07-013). The authors gratefully thank the Seychelles
Fishing Authority (SFA), and the crews of the longliners ‘‘Amitie
´’’
and ‘‘Cap Morgane’’ for helping us to collect the samples. We would
like to acknowledge the enormous contribution of YugNIRO scien-
tists in the long-term sampling of data in the Indian Ocean and the
development of the databases used in this paper. One of the authors
(Evgeny Romanov) was a member of that research team in 1985–
2005. Databases for the SIOTLLRP were created within the frame-
work of the data rescue project developed between YugNIRO and US
National Marine Fisheries Service (US NMFS) supported though a
grant from NFWF: NOAA Award No. NA87AC0200. We are
thankful to Anna Kornilova, who made available for analysis paper
archive of stomach contents data for Indian Ocean top predators
developed by her mother, Soviet scientist Galina N. Kornilova
9
who
worked in the YugNIRO from 1958 to 1990. We are indebted to
Natalya Romanova for her database management and programming
efforts, input of primary data of G. Kornilova and V. Zamorov and for
her patience with our numerous requests for data. We are thankful to
Dr. Walther W. Ku
¨hnhold (vTI, Germany) for his extremely valuable
help in obtaining important publications on biology of C. smithii. Our
thanks to Herve
´Demarcq and Francis Marsac for assistance in pro-
cessing netCDF climatic data. Critics and suggestions of three
anonymous reviewers greatly improved manuscript clarity. Elise
Bradbury and David Kaplan provided assistance with the English of
the manuscript. Pierre Lopez polished maps and provided advice with
graphic work.
References
Allen JA (1968) The surface swarming of Polybiuis henslowi
(Brachyura: Portunidae). J Mar Biol Assoc UK 48:107–111
Alverson FG (1963) The food of yellowfin and skipjack tunas in the
eastern Pacific Ocean. IATTC Bull 7:295–389
Anonymous (1986) Protocols of submersible ‘‘TINRO-2’’ diving, R/V
‘‘Gidrobiolog’’ Cruise 2/86. Sevastopol Experimental Construc-
tor Bureau for Underwater Research. Non-paginated (in Russian)
9
Deceased in 2000.
1104 Mar Biol (2009) 156:1089–1107
123
Anonymous (1987) Protocols of submersible ‘‘TINRO-2’’ diving, R/V
‘‘Gidrobiolog’’ Cruise 2/87. Sevastopol Experimental Construc-
tor Bureau for Underwater Research. Non-paginated (in Russian)
Apel M, Spiridonov VA (1998) Taxonomy and zoogeography of the
portunid crabs (Crustacea: Decapoda: Brachyura: Portunidae) of
the Arabian Gulf and adjacent waters. In: Krupp F, Mahnert F
(eds) Fauna of Arabia. National Commission for Wildlife
Conservation and Development (NCWCD) Riyadh, Saudi Ara-
bia; Pro Entomologia c/o Natural History Museum, Basle,
Switzerland, pp 159–331, Figs 1–117
Ashjian CJ, Smith SL, Flagg CN, Idrisi N (2002) Distribution, annual
cycle, and vertical migration of acoustically derived biomass in the
Arabian Sea during 1994–1995. Deep Sea Res Part II Top Stud
Oceanogr 49:2377–2402. doi:10.1016/S0967-0645(02)00041-3
Asseid BS, Drapeau L, Crawford RJM, Dyer BM, Hija A, Mwinyi
AA, Shinula P, Upfold L (2006) The food of three seabirds at
Latham Island, Tanzania, with observations on foraging by
masked boobies Sula dactylatra. Afr J Mar Sci 26:109–114
Aurioles-Gamboa D, Marin A, Aguiniga S (2003) Assimilation of
pelagic red crabs, Pleuroncodes planipes, by western gulls,
Larus occidentalis. Calif Fish Game 89:146–151
Baars MA, Schalk PH, Veldhuis MJW (1998) Seasonal fluctuations in
plankton biomass and productivity in the ecosystems of the
Somali Current, Gulf of Aden, and Southern Red Sea. In:
Sherman K, Okemwa E, Ntiba M (eds) Large marine ecosystems
of the Indian Ocean. Assessment, sustainability and manage-
ment. Blackwell, UK, pp 143–174
Bakun A, Roy C, Lluch-Cota S (1998) Coastal upwelling and other
processes regulating ecosystem productivity and fish production in
the western Indian Ocean. In: Sherman K, Okemwa E, Ntiba M
(eds) Large marine ecosystems of the Indian Ocean. Assessment,
sustainability and management. Blackwell, UK, pp 103–141
Balasubramanian CP, Suseelan C (1990) Preliminary observations on
the distribution and abundance of the swarming crab Charybdis
(Goniohellenus)smithii MacLeay in the deep scattering layers
along the west coast of India. In: Mathew K (ed) First workshop
on scientific results of FORV Sagar Sampada, Cochin, 5–7 June
1989. Central Marine Fisheries Research Institute, Cochin, pp
385–391
Balasubramanian CP, Suseelan C (1998) Natural diet of the deep
water crab Charybdis smithii McLeay (Brachyura: Portunidae)
of the seas around India. Indian J Fish 45:407–411
Balasubramanian CP, Suseelan C (2001) Distribution and population
structure of deep-water crab Charybdis smithii (Decapoda:
Portunidae) in the Arabian Sea and Bay of Bengal. Bull Mar
Sci 68:435–449
Bashmakov VF, Zamorov VV, Romanov EV (1991) Diet composition
of tunas caught with long lines and purse seines in the western
Indian Ocean. In: Workshop on stock assessment of yellowfin
tuna in the Indian Ocean. IPTP, Colombo, pp 53–59
Billett DSM, Bett BJ, Jacobs CL, Rouse IP, Wigham BD (2006) Mass
deposition of jellyfish in the deep Arabian Sea. Limnol Oceanogr
51:2077–2083
Castro JJ, Santiago JA, Santana-Ortega AT (2002) A general theory
on fish aggregation to floating objects: an alternative to the
meeting point hypothesis. Rev Fish Biol Fish 11:255–277. doi:
10.1023/A:1020302414472
Cherchi A, Gualdi S, Behera S, Luo JJ, Masson S, Yamagata T,
Navarra A (2007) The influence of tropical Indian Ocean SST
on the Indian summer monsoon. J Clim 20:3083–3105. doi:
10.1175/JCLI4161.1
Cherel Y, Sabatie R, Potier M, Marsac F, Me
´nard F (2007) New
information from fish diets on the importance of glassy flying
squid (Hyaloteuthis pelagica) (Teuthoidea: Ommastrephidae) in
the epipelagic cephalopod community of the tropical Atlantic
Ocean. Fish Bull (Washington) 105:147–152
Chesalin M, Zuyev G (2002) Pelagic cephalopods of the Arabian Sea
with an emphasis on Sthenoteuthis oualaniensis. Bull Mar Sci
71:209–221
Christiansen B, Boetius A (2000) Mass sedimentation of the
swimming crab Charybdis smithii (Crustacea: Decapoda) in
the deep Arabian Sea. Deep Sea Res Part II Top Stud Oceanogr
47:2673–2685. doi:10.1016/S0967-0645(00)00044-8
Dagorn L, Bach P, Josse E (2000) Movement patterns of large bigeye
tuna (Thunnus obesus) in the open ocean, determined using
ultrasonic telemetry. Mar Biol (Berlin) 136:361–371. doi:
10.1007/s002270050694
Daniel A, Chakrapany S (1983) Observations on the swarming,
breeding habits and some larval stages of the deep-sea portunid
crab Charybdis (Goniohellenus)edwardsi Leene et Buitenijk,
1949 in the northern Arabian Sea in January–February 1974 and
off the Madras coast during January to March 1976 to 1979. Rec
Zool Surv India 81(12):101–108
Della Croce N, Holthuis LB (1965) Swarming of Charybdis
(Goniohellenus)edwardsi Leene et Buitendijk in the Indian
Ocean (Crustacea, Decapoda, Portunidae). Boll Musei Ist Biol
Univ Genova 33(199):33–38
Demidov VF (1972) Report of the scientific group of Soviet fisheries
experts operated in the Southern Yemen. YugNIRO unpublished
field report, 320 p (in Russian)
Fonteneau A (1997) Atlas des pe
ˆcheries thonie
`res tropicales. Captures
mondiales et environnement [Atlas of tropical tuna fisheries,
world catches and environment]. ORSTOM editions, Paris
Fonteneau A, Pallares P, Pianet R (2000) A worldwide review of
purse seine fisheries on FADs. In: Le Gall J-Y, Cayre
´P, Taquet
M (eds) Pe
ˆche thonie
`re et dispositifs de concentration de
poissons. IFREMER, Plouzane (France), Actes Colloques 28, pp
15–34
Fonteneau A, Lucas V, Tewkai E, Delgado A, Demarcq H (2008)
Mesoscale exploitation of a major tuna concentration in the
Indian Ocean. Aquat Living Resour 21:109–121. doi:10.1051/
alr:2008028
Forward RB Jr, Tankersley RA, Rittschof D (2001) Cues for
metamorphosis of brachyuran crabs: an overview. Am Zool
41:1108–1122. doi:10.1668/0003-1569(2001)041[1108:CFMOBC]
2.0.CO;2
Gallienne CP, Conway DVP, Robinson J, Naya N, William JS, Lynch
T, Meunier S (2004) Epipelagic mesozooplankton distribution
and abundance over the Mascarene Plateau and Basin, south-
western Indian Ocean. J Mar Biol Assoc UK 84:1–8. doi:
10.1017/S0025315404008835h
Gebauer P, Paschke K, Anger K (2003) Delayed metamorphosis in
decapod crustaceans: evidence and consequences. Rev Chil Hist
Nat 76:169–175
Gjøsaeter J (1984) Mesopelagic fish, a large potential resource in the
Arabian Sea. Deep Sea Res Part I Oceanogr Res Pap 31:1019–
1035. doi:10.1016/0198-0149(84)90054-2
Goes JI, Thoppil PG, Gomes H do R, Fasullo JT (2005) Warming of
the Eurasian landmass is making the Arabian Sea more
productive. Science 308:545–547. doi:10.1126/science.1106610
Gutie
´rrez M, Ramirez A, Bertrand S, Moron O, Bertrand A (2008)
Ecological niches and areas of overlap of the squat lobster
‘munida’ (Pleuroncodes monodon) and anchoveta (Engraulis
ringens) off Peru. Prog Oceanogr 79:256–263. doi:10.1016/j.
pocean.2008.10.019
Hadfield MG, Carpizo-Ituarte EJ, del Carmen K, Nedved BT (2001)
Metamorphic competence, a major adaptive convergence in
marine invertebrate larvae. Am Zool 41:1123–1131. doi:
10.1668/0003-1569(2001)041[1123:MCAMAC]2.0.CO;2
Hallier J-P, Gaertner D (2008) Drifting fish aggregation devices could
act as an ecological trap for tropical tuna species. Mar Ecol Prog
Ser 353:255–264. doi:10.3354/meps07180
Mar Biol (2009) 156:1089–1107 1105
123
Herring PJ, Fasham MJR, Weeks AR, Hemmings JCP, Roe HSJ,
Pugh PR, Holley S, Crisp NA, Angel MV (1998) Across-slope
relations between the biological populations, the euphotic zone
and the oxygen minimum layer off the coast of Oman during the
southwest monsoon (August, 1994). Prog Oceanogr 41:69–109.
doi:10.1016/S0079-6611(98)00019-6
IOC, IHO, BODC (2003) Centenary edition of the GEBCO Digital
Atlas, published on CD-ROM on behalf of the Intergovernmental
Oceanographic Commission and the International Hydrographic
Organization as part of the General Bathymetric Chart of the
Oceans. British Oceanographic Data Centre, Liverpool
Jerde CW (1967) On the distribution of Portunus (Archelous)affinis
and Euphylax dovii (Decapoda, Brachyura, Portunidae) in the
eastern tropical Pacific. Crustaceana 13:11–22. doi:10.1163/
156854067X00026
Jerde CW (1970) Further notes on the distribution of Portunus
xantusii affinis and Euphylax dovii (Decapoda, Brachyura,
Portunidae) in the eastern tropical Pacific. Crustaceana 19:84–
88. doi:10.1163/156854070X00653
Karuppasamy PK, Balu S, George S, Persis V, Sabu P, Menon NG
(2007) Distribution and abundance of the swarming crab
Charybdis (Goniohellenus)smithii MacLeay in the deep
scattering layer of the eastern Arabian Sea. Indian Hydrobiol
10:165–170
Korsmeyer KE, Dewar H (2001) Tuna metabolism and energetics. In:
Block B, Stevens ED (eds) Tuna: physiology, ecology, evolu-
tion. Academic Press, San Diego, pp 35–78
Le Corre M (1997) Diving depth of two tropical pelecaniformes: the
red-tailed tropicbird and the red-footed booby. Condor 99:1004–
1007. doi:10.2307/1370157
Le Corre M, Jaquemet S (2005) Assessment of the seabird community
of the Mozambique Channel and its potential use as an indicator
of tuna abundance. Estuar Coast Shelf Sci 63:421–428. doi:
10.1016/j.ecss.2004.11.013
Longhurst A (1998) Ecological geography of the sea. Academic
Press, San Diego
Longhurst A (2004) The answer must be red crabs, of course.
Oceanography (Washington) 17:6–7
Losse GF (1969) Notes on the portunid crab Charybdis edwardsi
Leene & Buitendijk, 1949, from the Western Indian Ocean. J Nat
Hist, 145–152. doi:10.1080/00222936900770151
Losse GF, Merrett NR (1971) The occurrence of Oratosquilla
investigatoris (Crustacea: Stomatopoda) in the pelagic zone of
the Gulf of Aden and the equatorial western Indian Ocean. Mar
Biol (Berlin) 10:244–253. doi:10.1007/BF00352813
Luo J, Ortner PB, Forcucci D, Cummings SR (2000) Diel vertical
migration of zooplankton and mesopelagic fish in the Arabian
Sea. Deep Sea Res Part II Top Stud Oceanogr 47:1451–1473.
doi:10.1016/S0967-0645(99)00150-2
Marsac (2008) Gestionnaire d’Applications Oce
´anographiques
(GAO). Oceanographic data analysis and retrieval software.
v. 4.2. IRD 1998-2008
Me
´nard F, Marchal E (2003) Foraging behaviour of tuna feeding on
small schooling Vinciguerria nimbaria in the surface layer of the
equatorial Atlantic Ocean. Aquat Living Resour 16:231–238.
doi:10.1016/S0990-7440(03)00040-8
Me
´nard F, Labrune C, Shin Y-J, Asine A-S, Bard F-X (2006)
Opportunistic predation in tuna: a size-based approach. Mar Ecol
Prog Ser 323:223–231. doi:10.3354/meps323223
Merrett NR (1968) Tuna longline survey in the equatorial Western
Indian Ocean. East Afr Agric For J 34:17–65
Mincks SL, Bollens SM, Madin LP, Horgan E, Butler M, Kremer PM,
Craddock JE (2000) Distribution, abundance, and feeding
ecology of decapods in the Arabian Sea, with implications for
vertical flux. Deep Sea Res Part II Top Stud Oceanogr 47:1475–
1516. doi:10.1016/S0967-0645(99)00151-4
Munilla I (1997) Henslow’s swimming crab (Polybius henslowii)asan
important food for yellow-legged gulls (Larus cachinnans)inNW
Spain. ICES J Mar Sci 54:631–634. doi:10.1006/jmsc.1997.0249
Neiman VG, Burkov VA (1989) Large-scale water circulation in the
Indian Ocean. In: Parin NV, NovikovNP (eds) Biological resources
of the Indian Ocean. Nauka, Moscow, pp 20–66 (in Russian)
Neumann V, Spiridonov VA (1999) Shallow water crabs from the
Western Indian Ocean: Portunoidea and Xanthoidea excluding
Pilumnidae (Crustacea, Decapoda,Brachyura). Trop Zool 12:9–66
Norse EA, Estevez M (1977) Studies on portunid crabs from the
Eastern Pacific. I. Zonation along environmental stress gradients
from the coast of Columbia. Mar Biol (Berlin) 40:365–373. doi:
10.1007/BF00395729
Norse EA, Fox-Norse V (1977) Studies on portunid crabs from the
Eastern Pacific. II. Significance of the unusual distribution of
Euphylax dovii. Mar Biol (Berlin) 40:374–377. doi:10.1007/
BF00395730
Pechenik JA (1999) On the advantages and disadvantages of larval
stages in benthic marine invertebrate life cycles. Mar Ecol Prog
Ser 177:269–297. doi:10.3354/meps177269
Potier M, Marsac F, Lucas V, Sabatie R, Hallier J-P, Me
´nard F (2004)
Feeding partitioning among tuna taken in surface and mid-water
layers: the case of yellowfin (Thunnus albacares) and bigeye (T.
obesus) in the western tropical Indian Ocean. West Indian Ocean
J Mar Sci 3(1):51–62
Potier M, Marsac F, Cherel Y, Lucas V, Sabatie R, Maury O, Me
´nard
F (2007a) Forage fauna in the diet of three large pelagic fishes
(lancetfish, swordfish and yellowfin tuna) in the western
equatorial Indian Ocean. Fish Res 83:60–72. doi:10.1016/
j.fishres.2006.08.020
Potier M, Me
´nard F, Cherel Y, Lorrain A, Sabatie R, Marsac F
(2007b) Role of pelagic crustaceans in the diet of the longnose
lancetfish Alepisaurus ferox in the Seychelles waters. Afr J Mar
Sci 29:113–122. doi:10.2989/AJMS.2007.29.1.10.75
Potier M, Romanov E, Cherel Y, Sabatie R, Zamorov V, Me
´nard F
(2008) Spatial distribution of Cubiceps pauciradiatus (Percifor-
mes: Nomeidae) in the tropical Indian Ocean and its importance
in the diet of large pelagic fishes. Aquat Living Resour 21:123–
134. doi:10.1051/alr:2008026
Ramos JA (2000) Characteristics of foraging habitats and chick food
provisioning by tropical roseate terns. Condor 102:795–803. doi:
10.1650/0010-5422(2000)102[0795:COFHAC]2.0.CO;2
Rice AL (1969) Swarming of swimming crabs. Mar Observ 223:16–20
Romanov EV, Zamorov VV (2007) Regional feeding patterns of the
longnose lancetfish (Alepisaurus ferox Lowe, 1833) of the
western Indian Ocean. West Indian Ocean J Mar Sci 6:37–56
Romanov EV, Sakagawa G, Marsac F, Romanova N (2006) Historical
database on Soviet tuna longline tuna research in the Indian and
Atlantic oceans (first results of YugNIRO-NMFS data rescue
project). In: Paper presented at the eighth session of the IOTC
working party on tropical tunas. IOTC, Seychelles, 32 p
Romanov EV, Me
´nard F, Zamorov VV, Potier M (2008) Variability
in conspecific predation among longnose lancetfish Alepisaurus
ferox in the western Indian Ocean. Fish Sci 74:62–68. doi:
10.1111/j.1444-2906.2007.01496.x
Rykaczewski RR, Checkley DM Jr (2008) Influence of ocean winds
on the pelagic ecosystem in upwelling regions. Proc Natl Acad
Sci USA 105:1965–1970. doi:10.1073/pnas.0711777105
Sankarankutty C, Rangarajan K (1962) On a record of Charybdis
(Goniohellenus)edwardsi Leene and Buitendjik. J Mar Biol
Assoc India 6(2):311
Schaefer KM, Fuller DW (2005) Behavior of bigeye (Thunnus
obesus) and skipjack (Katsuwonus pelamis) tunas within aggre-
gations associated with floating objects in the equatorial eastern
Pacific. Mar Biol (Berlin) 146:781–792. doi:10.1007/s00227-
004-1480-x
1106 Mar Biol (2009) 156:1089–1107
123