Limnol. Oceanogr.

, 54(4), 2009, 1197–1209

E 2009, by the American Society of Limnology and Oceanography, Inc.

Mass deposition event of Pyrosoma atlanticum carcasses off Ivory Coast (West Africa)
M. Lebrato1,* and D. O. B. Jones
National Oceanography Centre, University of Southampton, Southampton, UK

Abstract
Thousands of moribund thaliacean carcasses (Pyrosoma atlanticum) were deposited between February and
March 2006 at the seafloor in the Ivory Coast area (West Africa). Remotely operated vehicle surveys were
conducted in a continuous depth gradient between 20 and 1275 m along an oil pipeline. Video and still
photography were used to estimate the carcass distribution, density, and size on the seabed, as well as recording
the local megafauna interactions with the gelatinous material. Large patches of dead pyrosomids covered
extensive areas on the continental slope, whereas minor aggregations were found on the shelf. The carcasses were
in many instances trapped along the pipelines, accumulating extensively in troughs and furrows in the slope, and
especially in soft sediment channels. Pyrosoma atlanticum dried samples were used to calculate the carbon content,
enabling the extrapolation to the densities and sizes recorded in the video surveys. The average standing stock of
organic carbon associated with the carcasses was .5 g C m22 in the whole slope and canyon, with values as high
as 22 g C m22 in certain areas. Eight megafaunal species from three different phyla were observed 63 times
directly feeding on the decomposing carcasses. The gelatinous carbon may have contributed substantially to the
detrital food web including microbes at the seabed, and certainly to the diet of larger benthic organisms. The
organism-level carbon measurements and documented fate of pyrosomid organic carbon is new evidence of the
importance of gelatinous material in large-scale processes and elemental cycling.

Carbon inputs to the deep sea—The deep sea can production (Lallier-Verges et al. 1993). The important
experience rapid inputs of organic carbon from the vertical transport of larger carbon parcels, such as
overlying surface waters with rapid responses by micro- gelatinous zooplankton carcasses, has been widely over-
and macrofaunal taxa (Gooday 2002). Deep-sea benthic looked (Billett et al. 2006).
ecosystems rely mainly on primary production from the
surface waters and material originating from primary Gelatinous zooplankton carbon export—Gelatinous zoo-
consumers (fecal pellets, dead bodies, discards; Robison plankton populations do not only affect pelagic ecosystem
et al. 2005). Other less-studied sources of organic carbon functioning (Mills 1995), but may also alter the ultimate
include gelatinous zooplankton carcasses (Billett et al. organic matter flux to the seafloor in both coastal and
2006), detrital particles (Beaulieu 2002), episodic deliveries deep-sea areas (Billett et al. 2006). Data on episodic export
of large organic parcels (Smith and Baco 2003), and of carbon from the carcasses of gelatinous zooplankton
macrophyte detritus (Vetter 1996). The significance of these populations are sparse. The majority of the literature has
organic carbon pathways to biogeochemical processes and focused on ‘‘discarded organogenic’’ inputs, such as
benthic communities vary between habitat and distance mucous sheets (Robison et al. 2005) and fecal pellets
from neritic waters, although reliable estimations remain (Perissinotto and Pakhomov 1998). Gelatinous animals
still in their infancy (Pilskaln et al. 1996). have been considered not the preferred prey item for other
Continental shelves and slopes effectively comprise less species (Moline et al. 2004). However, there is evidence that
than 20% of the oceans’ area, but are net sinks of inorganic they form part of the diet of many fish (Harbison 1998),
and especially organic carbon (Walsh et al. 1981). Carbon and probably are important diet items in benthic commu-
not buried or accumulated in the continental shelf is nities (Roe et al. 1990; Yamamoto et al. 2008). There is also
exported to the open ocean and either redeposited, or evidence that their carbon may be channeled by bacteria in
potentially converted to dissolved organic carbon (DOC), the water column and in the benthos, contributing greatly
entering the water column (Bauer and Druffel 1998). For to the microbial detrital food web (Riemann et al. 2006;
understanding of biogeochemical cycles it is necessary to Titelman et al. 2006). Since the biomass of gelatinous
analyze the collective contributions of all possible active organisms has been reported to increase in past decades
carbon inputs, particularly those that are poorly enumer- (Mills 2001), and new studies have appeared considering
ated. A major problem in this respect is that the settling the importance of organism-level measurements at the
flux of organic carbon and the active accumulation in ecosystem and carbon cycling levels, their role in biogeo-
sediments is assumed to be correlated directly with primary chemical cycles needs to be reconsidered. Billett et al.
(2006) stated that even if a major input and deposition of
* Corresponding
material derived from dead gelatinous zooplankton has
author: [email protected]
been suggested, no direct quantification has been made.
1 Presentaddress: IFM-GEOMAR. Department of Marine Cacchione et al. (1978) observed salp carcasses rolling in
Biogeochemistry, Leibniz Institute of Marine Science, Kiel, the Hudson Canyon at more than 3000-m depth, whereas
Germany. jellyfish bodies have been recorded on continental slopes
1197
Limnology limn-54-04-40.3d 24/4/09 14:42:05 1197 Cust # 08-493
1198 Lebrato and Jones

Fig. 1. (A) The Gulf of Guinea in the equatorial Atlantic and the location of the study area.
(B) Detailed bathymetric chart of the Ivory Coast and the adjacent areas (Sierra Leone and
Ghana). The black box indicates the study area represented in C. Bold bathymetric lines indicate
the 100-, 500-, 1000-, 2000-, and 4000-m zones. (C) Detailed bathymetric chart of the surveys in
the continental shelf and slope areas. Gridded data taken from the general bathymetric chart of
the oceans (GEBCO) Digital Atlas (Centenary Edition).

(Miyake et al. 2005). One of first records of pyrosomids at importance of organism-level measurements in carbon
the seabed was a patch recorded by Monniot and Monniot budget calculations regionally and globally.
(1966) off the Cape Verde Islands. Pyrosoma atlanticum
carcasses have been observed on the Madeira Abyssal Plain Methods
(Roe et al. 1990) and Crambionella orsini bodies were
recorded in canyon systems on the continental slope and Study site—The Ivory Coast occupies a critical position
rise off Oman in the Arabian Sea (Billett et al. 2006). for the Atlantic Equatorial Current system (Hardman-
Yamamoto et al. (2008) provided the most recent evidence Mountford 2000) (Fig. 1A). The Guinea Current and the
of the deposition of jellyfish carcasses (Nemopilema Guinea Undercurrent dominate the physical environment,
nomurai) in the Japan Sea. mediating coastal upwelling and the development of warm
The observation of large patches of dead pyrosomids on and saline waters. Sea surface temperature is reduced
the seabed off Ivory Coast (West Africa) motivated a during these periods, as well as oxygen, whereas nutrients
detailed quantification of this P. atlanticum mass deposi- increase, triggering primary productivity in the area
tion event. This study aims to (1) quantify in detail carcass (Mensah and Koranteng 1988). There is no continuous
densities and sizes from a remotely operated vehicle (ROV) oxygen minimum zone development, such as found in the
video, with relation to depth and topography on the Arabian Sea (Billett et al. 2006), but oxygen levels
continental shelf and slope, (2) calculate the associated approach suboxic values at around 300 m.
carbon standing stock on the basis of video data and The continental shelf off Ivory Coast is characterized by
organism-level carbon measurements, and (3) assess the its narrowness, covering some 10,000 km2, with the slope
fate of the pyrosomid carcasses at the seabed, and their normally starting at about 120 m, gaining depth relatively
contribution to the food web. This study highlights the fast (Fig. 1B). From the coast down to 2000 m, in the area
limited understanding of major processes associated with between Sassandra and Grand Bassam (near Abidjan),
gelatinous zooplankton blooms and the carbon cycle, there is a predominantly gently sloped seafloor, with few
especially when mass deposition events seem to be a major irregularities except for a marine canyon, known as ‘‘le
feature of the world oceans. The work stresses the Trou-sans-fond’’ (the bottomless hole). In the vicinity, a

Limnology limn-54-04-40.3d 24/4/09 14:42:06 1198 Cust # 08-493

 Pyrosoma atlanticum deposition event 1199

Table 1. ROV transect data in the depth zones referred to in the text.

Depth (m) Position

Depth zone (m) Date na Min. Max. Start–end (uN) Start–end (uW) Area sampled (m2)
0–200 Feb–Mar 06 55 26.3 139.5 5.155–5.028 4.519–4.466 5500
200–400 01 Mar 06 5 200.6 385.5 4.942–5.009 4.468–4.474 500
400–900 Feb–Mar 06 11 510.8 899.6 5.000–4.973 4.478–4.532 1100
900–1100 Feb–Mar 06 30 901.3 1095.7 4.973–4.968 4.533–4.497 3000
1100+ Feb–Mar 06 29 1133.1 1275.7 4.969–4.947 4.499–4.491 2900
a
Each n represents one ROV transect (100 m2).

series of submarine canyons and structures dissect the comparing it with taxonomic guides (e.g., Koranteng 2001).
underwater landscape from the outer edge of the shelf and Expert advice (see Acknowledgments) was also used in the
slope down to 2000 m (Fig. 1B,C). case of a few echinoderms and crustaceans. For quantita-
Thaliaceans only represent a minor part of the zoo- tive analyses, data were averaged from every 100-m depth
plankton biomass in the Gulf of Guinea. However, they interval (n in Table 1). Owing to the nature of the survey
bloom periodically, increasing the living zooplankton there were different total areas sampled and hence different
biomass significantly and dominating the standing stock. numbers of sample units at each 100-m depth interval
Roger (1982) reported on the high densities of P. (Table 1).
atlanticum, accounting for 34% of the total zooplankton
biomass in the water column, whereas Goy (1977) recorded Data analyses: Statistical analyses of the carcass density
P. atlanticum populations around the shallow thermocline and size and carbon standing stocks at different depth
dominating in the Guinea Dome. Le-Borge (1983) estimat- intervals were performed. Data processing was carried out
ed that salp blooms off Ivory Coast may occasionally with Minitab 14.0 and graphs were plotted with SigmaPlot
occupy more than 100 km2. 10.0. Pyrosoma atlanticum average carcass size and density
were recorded for each sampling unit. These were
ROV video surveys—Data collection: The Baobab oil subsequently averaged at each depth interval. Statistical
field (discovered 2001) lies in 1484 m of water off Ivory analyses of depth interval medians were performed using
Coast. ROV surveys conducted between 18 February and the nonparametric Kruskal–Wallis test. The Miller (1981)
19 March 2006 were used for this quantitative investigation post hoc nonparametric Tukey-type test (Zar 1999) was
(Table 1). The surveys made several traverses of the used for further comparisons.
continental shelf and slope along seabed oil pipelines, from
26- to 1276-m water depth (Table 1; Fig. 1C). Data were Pyrosoma atlanticum carbon analyses—Pyrosoma atlan-
collected by a Sonsub work-class ROV operated from the ticum carcasses were obtained from the Catalan Sea
vessel DP Reel. The original data had a geodetic datum of (northwest Mediterranean) because no samples were
Port Bouet (Abidjan, Ivory Coast 1987 data). Data were available from the Ivory Coast. The carcasses were
recorded from the videos as universal transverse Mercator collected from 650- to 1000-m depth in trawling surveys
northings and eastings; these were transformed to latitude (OTSB-14 and commercial) during the spring of 2007 (J.
and longitude (WGS84) with FRANSON CoordTrans. Cartes pers. comm.). Carcasses were already at the seabed
when collected (freshly deposited), which maximizes the
ROV video analyses: A total of 309 videos with a mean extrapolation potential with the Ivory Coast carcasses
duration of 20 min were analyzed. Of these, 130 were used observed in video. Pyrosoma atlanticum samples were
in subsequent analysis. The survey videos were visualized stored at 220uC after capture, and subsequently dried at
and analyzed with VisualReview 6.5 (VisualSoft Limited). 80uC for 24 h, and then dried again upon arrival for
Transects were divided into segments representing 100 m2 another 24 h. Samples were disintegrated to powder with a
of seafloor area (precision 6 1 m2) using ROV position metal blade before analysis. Carbon analyses were per-
data; this represented an average of 7 min of video. Size formed for inorganic carbon (IC) in an IC CO2 coulometer
estimations were made by comparison with known and for total carbon (TC) using model 5012 UIC
dimensions of the pipeline. The field of view was oblique, coulometers in triplicate (Johnson et al. 1998). The amount
so size estimations were always performed at the base of the of pyrosomid homogenized material used in the subsample
screen, ensuring the highest possible resolution. Biological measurements was 12.31 6 1.51 mg (n 5 5) for IC, and
and pyrosomid carcass data (megafaunal densities, carcass 11.43 6 2.31 mg (n 5 12) for TC. Calibration of the
densities, carcass size, carcass condition, and observations coulometers was verified with pure, dry CaCO3 standards.
of feeding on carcasses) were recorded from videos within IC was determined to be 11.99% 6 0.33% (cf. 12.00%),
each 100-m2 sampling unit (Table 1). equivalent to 99.90% 6 2.80% CaCO3 (n 5 8) for IC. TC
All megafaunal organisms that could be identified content of the standard was determined to be 11.98% 6
visually and repeatedly (.50 mm) (Jones et al. 2007) were 0.01% (cf. 12.00%), equivalent to 99.85% 6 0.81% of
recorded per 100-m2 sampling unit. Taxonomic uncertain- CaCO3 (n 5 8). Organic carbon (OC) was estimated as the
ties were resolved by taking a still frame of the animal and difference between TC and IC.

Limnology limn-54-04-40.3d 24/4/09 14:42:12 1199 Cust # 08-493

1200 Lebrato and Jones

Table 2. Pyrosoma atlanticum sample morphometrics and carbon component estimations. TC is total carbon, IC is inorganic
carbon, and OC is organic carbon. SD is the standard deviation on the basis of triplicate measurements.

Weight (wt, g) Carbon %

Sample cruise Date Depth (m) Wet Dry Dry wt : wet wt (%) Length (mm) TC (SD) ICa OC (SD)
BIOMARE BOU5 28 Dec 07 800 1.801 0.206 11.43 45.9 41.05 (0.71) 0.17 40.88 (0.71)
BIOMARE3B BOU10 17 Sep 07 650 1.046 0.087 8.31 27.8 31.14 (5.71) 0.16 30.98 (5.71)
BIOMARE3 OTSB6 01 Jul 07 650 0.362 0.041 11.32 17.8 29.60 (0.55) 0.18 29.42 (0.55)
BIOMARE2 OTSB2 28 Apr 07 650 1.007 0.135 13.40 35.9 40.15 (0.68) 0.19 39.96 (0.68)
a
No triplicate measurements for IC due to lack of sample material.

An average of the sample carbon components (n 5 4) on the seafloor (Fig. 2A). The average bottom current
was used to estimate the carbon standing stock from the speed on the continental shelf was ,0.5 m s 21, as
carcasses in the ROV videos: IC 5 0.17%, TC 5 35.10%, determined by video observations of current flow. Some
OC 5 34.93%. The calibration against four individuals is areas had current speeds up to 2 m s21. Near-bed currents
limiting, but literature data from Davenport and Balazs resuspended Pyrosoma carcasses in the water column and
(1991) revealed OC% data comparable with our results (a actively rolled them along the bottom. No carcasses were
dry weight of 39.2%). An equation was used to translate trapped along the pipelines on the shelf. The seabed was
the total length (TL) of the carcasses recorded in the characterized by a mixture of fine and coarse sediment with
videos to the dry weight: ‘‘log dry weight 5 21.804 + 1.692 biogenic material accumulating along the pipelines
log TL’’ (Mayzaud et al. 2007). Subsequently, the dry (Fig. 2A).
weights were converted to carbon per 100 m2. The values
were converted to standing stocks per square meter in each Upper continental slope and canyon: Many P. atlanti-
depth interval, to be comparable with other literature cum carcasses were recorded at the shelf break and the
data. An underestimation of the P. atlanticum standing upper slope between 300 and 400 m (.2000 carcasses per
stock is possible because when they accumulated in the 100 m2 in some cases) (Fig. 2B). The carcasses appeared to
sediment they formed piles, sometimes to 1 m in height, be degrading on both sides of the pipeline or in the vicinity.
with potentially greater numbers in channels and troughs. The seafloor environment was characterized by fine and
Density counts from video were carried in the two visible coarse sediments. In many areas the pipeline was buried,
dimensions; the actual standing stock is potentially higher and numerous channels and troughs were observed, with
than calculated. Densities and the estimated carbon substantial quantities of carcasses accumulating there. The
standing stocks were influenced by the presence of the average current speed was ,0.5 m s21, varying between 0
pipelines in all ROV videos. The pipelines act as an and 1 m s21 in the majority of cases. At midslope depths
artificial trap for drifting carcasses, and they also may and along the walls of the canyon (ca. 400–800 m) there
attract a higher number of scavengers as a result of was high variability in the number of carcasses, with ,100
accumulation of organic matter and an artificial reef to .2000 carcasses per 100 m2 recorded (Fig. 2C). The
effect. pyrosomids were in some cases degrading, forming
gelatinous aggregates (coalescence). The current was less
than 1 m s21. The bottom topography was characterized
Results by fine sediment with the sporadic presence of channels and
troughs. Toward the bottom of the steep continental slope
Sample morphometrics and carbon estimations—Pyro- (ca. 900 m) the number of rolling carcasses decreased.
soma atlanticum wet weight ranged from 1.007 to 1.801 g, Where the pipeline was buried, many carcasses were
with dry weights of 0.041 to 0.206 g (Table 2). The dry observed accumulating in piles and in the troughs, and in
weight as a percentage of wet weight varied between 8.31% much of the video the carcasses were transported across the
and 13.40% (an average of 11.21%). The length varied from field of view by a strong current between 1 and 2 m s21
a small specimen of 18 mm to a larger one of 46 mm (Fig. 2D,E).
(Table 2). TC ranged from 29.60% and 41.05% of the dry
weight, IC from 0.16% and 0.19% of the dry weight, and Lower continental slope: A relatively tranquil seabed
OC from 29.42% and 40.88% of the dry weight (Table 2). environment with a very patchy distribution of the
The dry weight, on average, had 35.10% TC, 34.93% OC, pyrosomid bodies was recorded at the lower continental
and 0.17% IC. slope (ca. 1000 m). The density varied between ,20 to
.2000 carcasses per 100 m2 in some cases, and the
Pyrosoma atlanticum carcass observations at the carcasses were in advanced degradation, starting decom-
sea floor—Continental shelf: Few carcasses were observed position, with an increase in the number of coalescent
rolling or accumulating at shelf depths from 26 to 200 m. gelatinous aggregates rolling along the bottom. In the
Some of the carcasses between 60 and 100 m appeared to majority of cases current speeds were very low. Carcasses
have been deposited recently, owing to their smooth were not moving and formed patches close to the pipelines
appearance. Detritus and macrophyte detritus occurred or rolled very slowly along the seabed in small groups

Limnology limn-54-04-40.3d 24/4/09 14:42:12 1200 Cust # 08-493

 Pyrosoma atlanticum deposition event 1201

Fig. 2. Pyrosoma atlanticum observations in the continental shelf and slope areas in the surveys
at (A) 105 m, (B) 379 m, (C) 744 m, (D) 952 m, (E) 954 m, and (F) 1244 m. The carcasses were
observed accumulating in patches along the pipelines, and in troughs and mounds in the sediment.
Megafaunal taxa were observed feeding on P. atlanticum carcasses: (G) Diadema sp. at 107 m, (H)
Colossendeis sp. at 1198 m, (I) Phormosoma sp. and a regular echinoid at 827 m, and (J) Actinostola
sp. and A. aurelia at 889 m. Scale bars refer to 25 cm from (A) to (F) and to 10 cm from (G) to (J).

Limnology limn-54-04-40.3d 24/4/09 14:42:13 1201 Cust # 08-493

1202 Lebrato and Jones

Table 3. Pyrosoma atlanticum carcass density and size, along with the estimated carbon standing stock per unit area in the depth
zones referred to in the text. TC, total carbon; IC, inorganic carbon; OC, organic carbon.

Mean carcasses (100 m22) Mean carbon standing stock (g C m22)

Depth Density Size (cm) TC IC OC
zone (m) (min–max) (min–max) (min–max) (min–max) (min–max)
0–200 1.87 (0–19) 7.9 (4–14) 0.01 (831024–0.70) 6.531025 (4.531026–3.731024) 0.01 (8.431024–0.07)
200–400 1077 (25–4198) 14.4 (13–16) 5.40 (0.10–20.35) 2.931022 (5.731024–0.10) 5.37 (0.10–20.24)
400–900 908.18 (80–3011) 14.72 (12–18) 5.29 (0.38–22.24) 2.831022 (231023–0.11) 5.26 (0.38–22.12)
900–1100 836.66 (80–3020) 15.1 (11–19) 5.08 (0.43–22.74) 2.731022 (2.331024–0.12) 5.05 (0.43–22.63)
1100+ 707.41 (11–2800) 14.37 (11–18) 4.23 (0.04–20.75) 2.331022 (231024–0.11) 4.21 (0.04–20.65)

(Fig. 2F). A fine sediment environment with many small pyrosomid density decreased slightly (Table 3), but patch-
seabed ripples (wavelength ,200 mm) composed the iness of the carcass distribution increased (Fig. 3). Beyond
landscape. Occasional terrigenous and neritic plant mate- these depths, at the bottom of the steep upper slope and on
rial was seen. the lower slope (.900 m), the density decreased further
with increasing depth (Table 3). In the canyon wall regions
Pyrosoma atlanticum carcasses—Density and size: Den- Pyrosoma density was low and immediately outside the
sity of pyrosomid carcasses was found to be significantly canyon no carcasses were found (Fig. 3).
different between the five depth zones (Table 3; Kruskal– Carcass size, as with density, was significantly different
Wallis, H 5 92.64, df 5 4, p , 0.01). The density of the between depth zones (Kruskal–Wallis, H 5 42.62, df 5 4, p
moribund pyrosomids was significantly lower between 26 , 0.01). Pyrosoma were significantly smaller (post hoc
and 200 m than the deeper zones (post hoc analysis p , analysis p , 0.05) shallower than 200 m (Table 3). The
0.05). Carcass density increased significantly from 200 to average size of the carcasses almost doubled at deeper than
400 m, coinciding with the shelf break (Table 3). The 200 m and increased slightly, but not significantly, with
maximum density was recorded at 385 m, with 4200 depth to 1100 m (Table 3).
carcasses per 100 m2 (note the clear contrast between the
shelf and the start of the slope in Fig. 3). At midslope and Carbon standing stock: The associated OC was signif-
toward the bottom of the upper slope (400–900 m) the icantly different between depth bands (OC: Kruskal–

Fig. 3. Pyrosoma atlanticum carcass density (per 100 m2) distribution, and the estimated
carbon standing stock in grams of organic carbon per 1 m2 (g C m22) in the seabed surveys of the
continental shelf and slope.

Limnology limn-54-04-40.3d 24/4/09 14:42:28 1202 Cust # 08-493

 Pyrosoma atlanticum deposition event 1203

species were present. Phormosoma sp. (Echinoid) was

abundant (Table 4). The Ophiolepadidae (ophiuroidea)
occurred on the soft sediment (Table 4). The 900–1000-m
zone was populated by a similar suite of species as in the
previous zone but at lower densities. The Cnidarian
Actinostola sp. attached to the pipeline and sediment and
was found at notably higher densities in this depth band
(Table 4). Parapenaeus sp. were again common. In the
deepest transect (1100+ m) no species were particularly
abundant, but almost every phylum was represented.
Actinostola sp. had the greatest density (Table 4).

Direct feeding on Pyrosoma atlanticum carcasses—Four

echinoderms (Cidaris sp., Diadema sp., ‘‘white Echinoid,’’
and Phormosoma sp.; Fig. 2G,I), three cnidarians (‘‘white
Actinaria,’’ Actinoscyphia aurelia, and Actinostola sp.;
Fig. 2J), and one arthropod (Colossendeis sp.; Fig. 2H)
were observed directly consuming carcasses 63 times.
Fig. 4. The trend in the calculated carcass carbon standing Scorpaeniformes and a few elasmobranchs were always
stock (g C m22) against depth for the five depth zones in the observed in the vicinity of carcasses. Within the echinoder-
surveys (y-axis in log scale). mata, echinoids used their spines to trap carcasses
(Fig. 2G,I). The asteroid Solaster sp. was not seen
Wallis, H 5 48.53, df 5 4, p , 0.01). The OC standing scavenging directly, but there was an instance where an
stock estimations over the month of the survey were very individual appeared to be engulfing a P. atlanticum corpse.
small, from 26 to 200 m (Table 3). A significant difference Actinaria (cnidarians) were regularly observed feeding or
was observed between 200 m and the deeper depth bands attempting to feed on carcasses (engulfing whole pyroso-
(post hoc analysis p , 0.05). In the 200–400-m depth band mids) wherever they were found (Fig. 2J). Colossendeis sp.
the average OC standing stock significantly increased intercepted and manipulated the carcasses with its legs
(Table 3), with a maximum at 385 m (Table 3; Fig. 3 for (Fig. 2H). No microbial assemblages were directly quanti-
location and Fig. 4 for trend). From 400 to 900 m the OC fied, but in many instances carcasses seemed to have a
decreased slightly, but the maximum was greater than in white matlike covering on their surface. This was partic-
the previous depth band (Table 3; Figs. 3, 4). There was a ularly observed in the deeper carcasses, which have been
decrease in the average OC standing stock in the 900–1100- decomposing for a longer time.
m depth band, but the greatest localized calculation
occurred between these depths (Table 3; Figs. 3, 4). The Discussion
deepest parts of the survey (1100+ m) showed a further
decrease in the average OC standing stock, although dense Gelatinous detritus at the seafloor—Spatiotemporal
patches were still seen (Table 3; Figs. 3, 4). patterns in carcasses: Jellyfish and thaliaceans are reported
to be distributed patchily in their environment (Graham
Pyrosoma atlanticum carcasses in the benthic system— 2001), but may rapidly increase in many oceanic areas,
Associated megafaunal community: A total of 57 species altering ecosystem balances (Mills 1995). Here, mass
from 8 phyla were recorded on and around hard (pipelines) blooming of pyrosomids has led to significant transport
and soft substrata (Table 4). In the 26- to 200-m depth zone of corpses from the coastal margins to the deep sea. This
the Echinodermata had the highest density, with the finding indicates that gelatinous zooplankton populations
echinoid Diadema sp. found commonly on the pipeline have a major effect on large-scale processes, adding new
(Table 4). Comatulid crinoids were also abundant in vast evidence into the fate of their biomass once dead and their
aggregations on the pipeline. Fish were present in high importance in the carbon cycle. Future changes in
numbers from 26- to 200-m depth; unfortunately the most abundance, and biomass linked to the projected global
abundant fish could not be identified from the video. anthropogenic and climatic influence, may change their
Trachurus trachurus was highly abundant (Table 4). In the role and importance in nutrient and element cycling.
Cnidaria, Pennatula sp. was common on sediment and a Coastal margins tend to accumulate gelatinous zooplank-
red alcyonacean was observed regularly attached to the ton owing to physical barriers being compressed along the
pipeline. Within the Crustacea only the shrimp Parape- shoreline. For example, the widely distributed Salpa
naeus sp. occurred at this shallow depth. Polychaetes were democratica has occurred in dense aggregations in Aus-
also present, observed as mounds in the sediment (Table 4). tralia, New Zealand, Japan, and South Africa (Gibbons
In the 200–400-m depth zone fish had reduced richness, 1996). Populations of the same species and Salpa fusiformis
with Scorpaeniformes most abundant (Table 4). The have been reported in the Mediterranean Sea, and Salpa
crustaceans Galathea sp. and Parapenaeus sp. dominated aspera in the Carolina Slope (Madin et al. 2006). Dense
the benthic community. Crinoids were also particularly populations of Pegea confoederata have been widely
abundant (Table 4). From 400 to 900 m only three fish recorded in the Arabian Sea (Ramaswamy et al. 2005).

Limnology limn-54-04-40.3d 24/4/09 14:42:33 1203 Cust # 08-493

1204 Lebrato and Jones

Table. 4. Megafaunal taxa observed off Ivory Coast at each depth zone, along with the substratum type and the density per unit
area. Also included are the observed feeding events of megafauna on Pyrosoma atlanticum carcasses.

Depth zone (m) 0–200 200–400 400–900 900–1100 1100+

Pipeline fine + Pipeline fine Pipeline fine sediment Pipeline fine sediment Pipeline fine
Substratum type coarse sediment sediment + deep channels + deep channels sediment
--------------------------------------------------------------------------------Density (per 100 m2) ------------------------------------------------------------------------------
Chordata
Balistes sp. 1.72 0 0 0 0
Bathyraja sp. 0.01 0 0.09 0.03 0
Bathysolea sp. 0.01 0 0 0 0
Branchiostegus semifasciatus 0.03 0 0 0 0
Cephalopholis taeniops 0.01 0 0 0 0
Chimaera 0.01 0 0 0 0
Dentex sp. 2.36 0.40 0 0 0
Epinephelus sp. 0.12 2.20 0 0 0
Hoplostethus atlanticus 0 2.20 0 0 0
Muraena sp. 0.14 0 0 0 0
Mycteroperca sp. 0.49 0.20 0 0 0
Oxynotus centrina b 0 0.20 0 0 0
Pagrus sp. 0.07 0 0 0 0
Pseudupeneus prayensis 0.69 0 0 0 0
Psychrolutes sp. 0 0 0.18 0 0
Sargocentron sp. 0.56 0 0 0 0
Sciaenidae 3.49 0.20 0 0 0
Scorpaeniformesb 11.50 7.40 1.00 0 0
Serranus cabrilla 1.38 0 0 0 0
Sparus sp. 0.20 0 0 0 0
Trachurus trachurus 25.45 0 0 0 0
Zeus faber 0.03 0 0 0 0
Indet. fish 1 86.69 0 0 0 0
Indet. fish 2 0 0 0 0.03 0.10
Indet. fish 3 0 0 0 0 0.24
Indet. fish 4 0 0 0 0.10 0.10
Echinodermata
Marthasterias glacialis 0.02 0 0 0 0
Indet. red asteroid 0.33 0.20 0 0 0
Henricia sp. 0 0 0.64 0 0
Crinoidea 16.11 58.40 0 0 0
Cidaris sp.a 2.31 0 0.09 0 0
Diadema sp.a 106.40 2.00 0 0 0
Indet. white echinoid a 0 0 0.82 0.53 0.83
Phormosoma sp.a 0 0 193.64 20.77 2.69
Mesothuria sp. 0 0 0.36 0.47 0.41
Ophiolepadidae 0 0 14.82 11.10 0
Indet. red ophiuroid 0.15 0 0 0 0
Solaster sp.b 0 0 0.09 0.17 0.03
Cnidaria
Indet. white Actinariaa
Indet. red Actinaria 1.04 0.60 0.09 0 0
Actinoscyphia aureliaa 0 0 0.36 1.47 2.66
Actinostola sp.a 0 0 1.91 4.37 3.28
Pennatula sp. 13.15 0 0 0.03 1.38
Indet. red soft coral 11.55 0 0 0 0
Crustacea
Galathea sp.b 0.04 71.20 0 0 0
Majidae 0.18 0 0 0 0
Neolithodes grimaldib 0 0.20 0.09 0 0
Neolithodes sp.b 0 0 0 0 0.03
Pagurus sp. 0 0 0 0 0
Paralomis sp. 0 0 0 0.03 0.07
Parapenaeus sp.b 14.76 160 2.18 33.70 3.66
Polychaeta
Indet. annelid 10.13 0.6 0.73 0 0

Limnology limn-54-04-40.3d 24/4/09 14:42:33 1204 Cust # 08-493

 Pyrosoma atlanticum deposition event 1205

Table. 4. Continued.

Depth zone (m) 0–200 200–400 400–900 900–1100 1100+

Pipeline fine + Pipeline fine Pipeline fine sediment Pipeline fine sediment Pipeline fine
Substratum type coarse sediment sediment + deep channels + deep channels sediment
Mollusca
Gastropods 0.09 0 0 0 0
Octopus sp. 0.20 0.04 0.27 0.03 0
Sepia sp. 1.02 0 0 0 0
Arthropoda
Colossendeis sp.a 0 0 0.09 0.13 0.86
Urochordata
Ascidians 0.07 0 0 0 0
a
Observed directly feeding on Pyrosoma atlanticum carcasses.
b
Potential feeders on the carcasses.

As seen here and in Le-Borgne (1983), pyrosomatidae may continental rise after a time period), or to differences in
accumulate in dense numbers in African waters. At present, the topographic complexity of the Ivory Coast and Oman
when these extensive blooms collapse and sink, they slope areas. There is also the possibility that the pipelines
transfer large amounts of organic carbon to both coastal retard the transport to the continental rise at depths where
and deep-sea environments. The observation of vast currents are slow, and the carcasses just stay at the seabed.
quantities of Pyrosoma atlanticum moribund bodies at the It is likely that greater quantities of Pyrosoma would be
seabed adds new evidence toward a major ‘‘gelatinous transported deeper than observed here because many
pathway’’ of carbon to the ocean’s interior (Billett et al. individuals were seen being transported downslope below
2006). 950 m. The video data only cover a limited area. The
Off Ivory Coast, few pyrosomid carcasses were recorded results support the hypothesis that canyon systems are
on the continental shelf (26–200 m). The shelf environment important in the transfer of organic matter to the deep
is highly dynamic, especially shallower than 50 m. The few ocean (Harrold et al. 1998). The patchy distribution of the
carcasses observed were not transported laterally along the gelatinous detritus on the Ivory Coast margin was a
shelf, but downward. They were directed toward the shelf consequence of the bottom complexity at fine and broad
break, in some cases along the pipelines (acting as an scales, which includes the presence of pipelines and
artificial trapping structure), being rolled along the bottom channels. Cacchione et al. (1978) commented on the
or suspended in the water column close to the seafloor. The accumulation of S. aspera carcasses and salp-like material
slope area represented a major change in topographic in hollows and furrows in the slope area off New England
complexity. As a consequence, large accumulations of (NW Atlantic). They also observed ‘‘string-like aggrega-
carcasses occurred. The greater decomposition of the tions of mucous material,’’ which could well resemble the
pyrosomids with increasing depth led to the formation of coalescent material off Ivory Coast.
elongated gelatinous aggregates. The coalescence of several
carcasses and formation of gelatinous patches were similar Carbon export: The estimated organic carbon standing
to the jelly lakes observed on the seabed of the Arabian Sea stocks during the month of the survey were highly
and caused by the decomposition of the scyphomedusan C. dependent on the density and size of the pyrosomid
orsini (Billett et al. 2006). The pipeline was present above aggregations recorded. They were very patchy and in
the seabed in many instances along the slope. Structures ,100-m distance, the carbon stocks changed from about
such as pipelines were observed to intercept and accumu- 1 g C m22 to .20 g C m22. The greatest amounts occurred
late pyrosomids, translating to elevated carbon standing on the upper and lower continental slope within channels
stocks. However, in areas recorded without pipelines, vast and areas of high topographic complexity (also associated
numbers of pyrosomids were observed on the seafloor or in with the pipelines). The carbon standing stock in carcasses
depressions. There were no data from adjacent canyons of gelatinous zooplankton is an important feature of ocean
without any seabed features, but it is highly possible that biogeochemical cycles that has generally been overlooked.
they trap vast numbers of carcasses at sites of high Compared with sediment trap flux data in the area (1.09 g
topographic complexity along the whole slope. C m22 yr21 at 696 m) (Wefer and Fischer 1993) (Table 5),
The maximum density of Pyrosoma carcasses occurred in the average gelatinous carbon standing stock estimated in
canyon settings and on the open slope (especially in deep this study, most likely from one single deposition event,
channels with no pipeline laid). This was in contrast to C. was always higher (.4 g C m22 in the upper and lower
orsini, where the greatest amounts of jelly detritus occurred slope as a whole). At the same depth of the sediment trap
on the continental rise in the Arabian Sea (Billett et al. data, carcass standing stocks were above 20 g C m22 in less
2006). This may be related to spatiotemporal variation than 2 months. Gelatinous zooplankton also release DOC,
(carcasses may eventually have accumulated on the contributing to the pool of total OC. Hansson and

Limnology limn-54-04-40.3d 24/4/09 14:42:35 1205 Cust # 08-493

1206 Lebrato and Jones

Table 5. A selection of organism-level (gelatinous zooplankton) organic carbon (OC) standing stocks, fluxes, and sediment trap
fluxes in the world oceans.

Organism level Depth (m) Carbon origin OC standing stock flux Reference
Arabian Sea 3196 Crambionella orsini carcasses 1.5 g C m22 Billett et al. 2006
Arabian Sea 3188 ˝ 31.3 g C m22 ˝
Arabian Sea 3314 ˝ 78.0 g C m22 ˝
Monterey Bay 200–2979 Bathochordaeus charon houses 7.6 g C m22 yr21 Robison et al. 2005
Sargasso Sea .1500 Salpa aspera carcasses 909 mg C m22 Wiebe et al. (1979)
Shetland Islands - Salpa thompsoni carcasses 560 mg C m22 Nishikawa et al. (1995)
Sediment trap
Cape Blanc 730 Open ocean 3.76 g C m22 yr21 Fischer et al. 2000
Guinea Basin 859 ˝ 2.94 g C m22 yr21 ˝
Guinea Basin 3965 ˝ 2.18 g C m22 yr21 ˝
Guinea Basin 696 ˝ 1.09 g C m22 yr21 Wefer and Fischer 1993
Guinea Basin 3921 ˝ 2.19 g C m22 yr21 ˝
Arabian Sea 828 Coastal 13.5 g C m22 yr21 Honjo et al. 1999
Arabian Sea 903 ˝ 17.2 g C m22 yr21 ˝
Arabian Sea 1974 ˝ 17.4 g C m22 yr21 ˝
Arabian Sea 800 Open ocean 5.7 g C m22 yr21 ˝
Walvis Ridge 1648 ˝ 3.83 g C m22 yr21 Wefer and Fischer 1993
Walvis Ridge 599 ˝ 5.07 g C m22 yr21 ˝

Norrman (1995) calculated 0.012 mg C g wet wt21 d21 Organism-level carbon measurements importance: Rap-
release in the jellyfish Aurelia aurita decomposing carcasses. idly sinking, carbon-enriched vectors such as pyrosomids,
These DOC data, coupled with major OC inputs from the larvacean houses, and jellyfish carcasses (Robison et al.
decomposing carcasses as well as fecal pellets (Wiebe et al. 2005; Billett et al. 2006) are not detected and measured by
1979), will have contributed significant quantities to the traditional sampling methods in the water column (sedi-
carbon pool and to export from continental shelves and ment traps). Yet, they are important contributors to
slopes globally. regional and global carbon inventories, and may have been
Pyrosoma atlanticum has one of the highest carbon ignored in vertical nutrient fluxes and budgets calculations.
contents (see Table 6) recorded in gelatinous animals West et al. (2009) presented data on the influence of
studied to date (Davenport and Balazs 1991). This has decomposing jellyfish on nutrients in the water column,
major implications in the role of pyrosomatidae as vectors with increases in dissolved organic nitrogen and phospho-
of carbon transport in the global oceans because they have rus between 8 and 25 times greater than in controls. There
the potential to export more carbon per unit area than were significant differences when comparing the Ivory
other bloom-forming species. In the study of the elemental Coast Pyrosoma carbon measurements over a period of 2
composition of gelatinous animals, inter- and intraspecific months and the sediment trap data in the area (Table 5).
variation is a common feature (e.g., owing to feeding Average standing stocks above 5 g C m22 over 2 months
preferences, environment, and life history). This may add seen here were similar to that estimated by Robison et al.
uncertainties about the carbon flux at any time and space. (2005) of 7.6 g C m22 yr21 from larvacean houses off
The available literature used conversions from published Monterey Bay, which matched Pilskaln et al. (1996) 7.2 g C
standard observations (Billett et al. 2006), which should be m22 yr21 sediment trap benthic flux at the midslope
avoided if it is possible to measure carbon directly from the (Table 5). Billett et al. (2006) in the Arabian Sea reported
species under investigation. jellyfish standing stocks of carbon between 1.5 and

Table 6. Pyrosoma atlanticum percentage of dry weight to wet weight (dry wt : wet wt %), inorganic carbon percentage (IC), and
organic carbon percentage (OC) composition along with other thaliacean and jellyfish species values from the literature.

Species Dry wt : wet wt (%) IC (%) OC (%) Reference

Pyrosoma atlanticum 11.21 0.17 34.9 This study
˝ 6.31 - 39.2 Davenport and Balazs 1991
Thalia democratica 8.04 - 18 Heron et al. 1988
Salpa fusiformis 4.7 - 11.9 Clarke et al. 1992
˝ 3.8 - 17–22 Dusbischar et al. 2006
Cyclosalpa bakeri - - 3.2 Madin and Purcell 1992

Limnology limn-54-04-40.3d 24/4/09 14:42:35 1206 Cust # 08-493

 Pyrosoma atlanticum deposition event 1207

78.0 g m22, which in some places were substantially higher indicate that their carbon is channeled through bacteria to
than Honjo et al. (1999) sediment trap flux estimations benthic communities. Titelman et al. (2006) have shown
(between 3.3 g C m22 yr21 and 17.4 g C m22 yr21) experimentally that gelatinous detritus originating in
(Table 5). The evidence that dead carcasses of animals jellyfish can be decomposed within a week by bacterial
effectively provide major carbon pathways to the seabed activity. If this short time frame applied to pyrosomids is
has been gathered very slowly (Billett et al. 2006). In uncertain, but based on the state of the carcasses at the
addition, direct quantification of the fluxes and the origins bottom, it will seem that their footprint will last much
are difficult and may rely on advanced technology and longer than the jellyfish. Over large areas, the ultimate fate
effective spatiotemporal sampling. Particulate carbon of gelatinous carbon will certainly be its incorporation to
fluxes from the photic zone (understanding by ‘‘particu- the detrital food web and the sediments through microbial
late’’ any vertically directed flux of particles, organism pathways.
discards, or organisms themselves) effectively sink while
being remineralized on their way down, releasing nutrients Acknowledgments
and elemental components in the water column. We thank Canadian Natural Resources (CNR), the DP Reel
vessel for collecting the data, and Paul Bennion from Sonsub for
Gelatinous carcasses in the food web—Off Ivory Coast enabling the raw data usage. We also thank the scientific and
several species belonging to different phyla were recorded environmental ROV partnership using existing industrial technol-
directly feeding on P. atlanticum carcasses. However, it is ogy project (SERPENT) for enabling access to the data. We are
very grateful for the samples provided by Joan Cartes from the
difficult to assess if animals are attracted to the pipeline as
Centre Mediterrani d’Investigacions Marines I Ambientals in
a hard substratum or by a substantial source of food from Barcelona (Spain) in the project BIOMARE, ref. CTM2006-
carcasses. It is most likely a combination of the two. An 13508-CO2-02/MAR and Darryl Green and Belinda Alker from
artificial structure, combined with a large input of organic the geochemistry group in Southampton, for the use of the
matter (in this case from carcasses), will attract scavengers coulometer system to calculate carbon values. Thanks to Debora
and detritivores that dominate at times the benthic Iglesias in Southampton for funding carbon analyses. Thanks also
community (see Fig. 2C). to specialists for help with species identifications: Paul Tyler for
Rapidly sinking pyrosomid bodies off the Ivory Coast echinoderms, Sven Thatje for crustaceans, and David Billett for
will have made a substantial contribution to benthic energy holothurians and providing useful comments on the final
supply. Pyrosomids appeared to be actively fed on by manuscript. We also thank the reviewers, whose comments helped
crustaceans, echinoderms, cnidarians, and arthropods. a lot in bringing together a final version of the manuscript.
This work was funded by the grant Becas mineras. exp. 19997
Fish, mainly Scorpaeniformes, were always close to the
to M.L.
carcasses, but they were not observed feeding directly on
the Pyrosoma. The limited literature suggests that feeding
opportunities are created by carcasses in other areas. A References
crustacean (Munidopsis sp.) and an echinoderm (Hyphala- BAUER, J. E., AND E. R. M. DRUFFEL. 1998. Ocean margins as a
ster inermis) were observed close to a moribund pyrosomid significant source of organic matter to the deep open ocean.
in the Madeira Abyssal Plain at 5440 m (Roe et al. 1990). Nature 392: 482–485.
Duggins (1981) also recorded a sea urchin feeding on salp BEAULIEU, S. E. 2002. Accumulation and fate of phytodetritus on
carcasses in Alaska. Fish have also been recorded feeding the sea floor. Oceanogr. Mar. Biol. Annu. Rev. 40: 171–232.
on pyrosomids (see Harbison 1998 and references therein). BILLETT, D. S. M., B. J. BETT, C. L. JACOBS, I. P. ROUSE, AND B. D.
The notion of gelatinous carcasses as rare items in WIGHAM. 2006. Mass deposition of jellyfish in the deep
trophic webs (Moline et al. 2004) is not supported by the Arabian Sea. Limnol. Oceanogr. 51: 2077–2083.
CACCHIONE, D. A., G. T. ROWE, AND A. MALAHOFF. 1978.
observations in the benthic system off Ivory Coast.
Submersible investigation of the outer Hudson submarine
Gelatinous zooplankton corpses are food parcels to benthic canyon. In D. J. Stanley and G. Kelling [eds.], Sedimentation
communities, with many animals observed feeding on this in submarine canyons, fans and trenches. Dowden, Hutch-
resource (Harbison 1998). Yamamoto et al. (2008) ob- inson and Ross.
served dense beds of ophiuroids (Ophiura sarsii), anemones, CLARKE, A. L., L. J. HOLMES, AND D. J. GORE. 1992. Proximate
shrimps (Pandalopsis japonica), and molluscs (Chionoecetes and elemental composition of gelatinous zooplankton from
opilio) close to the carcasses of N. nomurai in the Sea of the Southern Ocean. J. Exp. Mar. Biol. Ecol. 155: 55–68.
Japan. Billett et al. (2006) saw no evidence that the jelly DAVENPORT, J., AND G. H. BALAZS. 1991. ‘Fiery pyrosomas’: Are
detritus was a direct food source for benthic invertebrates. Pyrosomas an important items in the diet of leatherback
In the Arabian Sea, and probably in all systems, the high turtles? Brit. Herp. Soc. Bull. 37: 33–38.
organic carbon standing stocks are eventually channeled DUBISCHAR, C. D., E. A. PAKHOMOV, AND U. V. BATHMANN. 2006.
through bacteria. Protozoans may also channel part of the The tunicate Salpa thompsoni ecology in the Southern Ocean.
II. Proximate and elemental composition. Mar. Biol. 149:
organic carbon, since they are known to enter thaliacean 625–632.
fecal pellets, reducing the bacterial populations (Pomeroy DUGGINS, D. O. 1981. Sea urchins and kelp: The effects of short-
et al. 1984). If the same applies for carcasses, then both term changes in urchin diet. Limnol. Oceanogr. 26: 391–394.
bacteria and protozoa will consume and metabolize the FISCHER, G., V. RATMEYER, AND G. WEFER. 2000. Organic carbon
available carbon (Yoon et al. 1996). Off Ivory Coast, fluxes in the Atlantic and the Southern Ocean: Relationship
Pyrosoma atlanticum carcasses were observed covered at to primary production compiled from satellite radiometer
times by a mat-like substance of a white color, which may data. Deep Sea Res. II 47: 1961–1997.

Limnology limn-54-04-40.3d 24/4/09 14:42:36 1207 Cust # 08-493

1208 Lebrato and Jones

GIBBONS, M. J. 1996. Vertical distribution and feeding of Thalia MENSAH, M. A., AND K. A. KORANTENG. 1988. A review of the
democratica on the Agulhas Bank during February, 1994. J. oceanography and fisheries resources in the coastal waters of
Mar. Biol. Assoc. UK 77: 493–506. Ghana. Ministry of Agriculture, Tema (Ghana). Fisheries
GOODAY, A. J. 2002. Biological responses to seasonally varying Department. Research and Utilization Branch Rep. 8.
fluxes of organic matter to the ocean floor: A review. J. MILLER, R. G. 1981. Simultaneous statistical inference. 2nd ed.
Oceanogr. 58: 305–332. Springer-Verlag.
GOY, J. 1977. Vertical migration of zooplankton. Res. Camp. MILLS, C. E. 1995. Medusae, siphonophores and ctenophores as
Mer. CNEXO 13: 71–73. planktivorous predators in changing ocean ecosystems. ICES
GRAHAM, W. M. 2001. Numerical increases and distributional J. Mar. Sci. 52: 575–581.
shifts of Chrysaora quinquecirrha (Desor) and Aurelia aurita ———. 2001. Jellyfish blooms: Are populations increasing
(Cnidaria: Scyphozoa) in the northern Gulf of Mexico. globally in response to changing ocean conditions? Hydro-
Hydrobiologia 451: 97–111. biologia 451: 55–68.
HANSSON, L. J., AND B. NORRMAN. 1995. Release of dissolved MIYAKE, H., M. KITAMURA, AND S. NISHIDA. 2005. Occurrence of
organic carbon (DOC) by the scyphozoan jellyfish Aurelia the scyphomedusa Parumbrosa polylobata Kishinouye, 1910
aurita and its potential influence on the production of in Suruga Bay, Japan. Plank. Biol. Ecol. 52: 58–66.
planktonic bacteria. Mar. Biol. 121: 527–532. MOLINE, M. A., H. CLAUSTRE, T. K. FRAZER, O. SCHOFIELD, AND
HARBISON, G. R. 1998. The parasites and predators of Thaliacea, M. VERNET. 2004. Alteration of the food web along the
p. 187–214. In Q. Bone [ed.], The biology of pelagic Antarctic Peninsula in response to a regional warming trend.
Tunicates. Oxford. Global Change Biol. 10: 1973–1980.
HARDMAN-MOUNTFORD, N. J. 2000. Environmental variability in MONNIOT, C., AND F. MONNIOT. 1966. A benthic pyrosoma: Pyrosoma
the Gulf of Guinea large marine ecosystem: Physical features, benthica n. sp. French Academic Sciences 263: 368–370.
forcing, and fisheries.Ph.D. thesis. Univ. of Warwick. NISHIKAWA, J., M. NAGANOBU, T. ICHII, H. ISHII, M. TERAZAKI, AND
HARROLD, C., K. LIGHT, AND S. LISIN. 1998. Organic enrichment K. KAWAGUCHI. 1995. Distribution of salps near the South
of submarine-canyon and continental-shelf benthic commu- Shetland Islands during austral summer, 1990–1991 with
nities by macroalgal drift imported from nearshore kelp special reference to krill distribution. Polar Biol. 15: 31–39.
forests. Limnol. Oceanogr. 43: 669–678. PERISSINOTTO, R., AND E. A. PAKHOMOV. 1998. Contribution of
HERON, A. C., P. S. MCWILLIAM, AND G. DALPONT. 1988. Length- salps to carbon flux of marginal ice zone of the Lazarev Sea,
weight relation in the salp Thalia democratica and potential Southern Ocean. Mar. Biol. 131: 25–32.
of salps as a source of food. Mar. Ecol. Prog. Ser. 42: 125– PILSKALN, C. H., J. B. PADUAN, F. P. CHAVEZ, R. Y. ANDERSON,
AND W. M. BERELSON. 1996. Carbon export and regeneration
132.
in the coastal upwelling system of Monterey Bay, Central
HONJO, S., J. DYMOND, W. PRELL, AND V. ITTEKKOT. 1999.
California. J. Mar. Res. 54: 1–31.
Monsoon-controlled export fluxes to the interior of the
POMEROY, L. R., R. B. HANSON, P. A. MCGILLIVARY, B. F.
Arabian Sea. Deep-Sea Res. 46: 1859–1902.
KIRCHMAN, AND D. DEIBEL. 1984. Microbiology and chemistry
IOC, IHO, and BODC. 2003. Centenary edition of the GEBCO
of fecal products of pelagic tunicates: rates and fates. Bul.
Digital Atlas. British Oceanographic Data Centre, Liverpool,
Mar. Sci. 35: 426–439.
UK.
RAMASWAMY, V., M. M. SARIN, AND R. RENGARAJAN. 2005.
JOHNSON, K. M., AND oTHERS. 1998. Coulometric total carbon Enhanced export of carbon by salps during the northeast
dioxide analysis for marine studies: Assessment of the monsoon period in the northern Arabian Sea. Deep-Sea Res. I
quality of total inorganic carbon measurements made during 52: 1922–1929.
the US Indian Ocean CO Survey 1994–1996. Mar. Chem. 63: RIEMANN, L., J. TITELMAN, AND U. BÅMSTEDT. 2006. Links between
21–37. jellyfish and microbes in a jellyfish dominated fjord. Mar.
JONES, D. O. B., B. D. WIGHAM, I. R. HUDSON, AND B. J. BETT. Ecol. Prog. Ser. 325: 29–42.
2007. Anthropogenic disturbance of deep-sea megabenthic ROBISON, B. H., K. R. REISENBICHLER, AND R. E. SHERLOCK. 2005.
communities investigated using remotely operated vehicles Giant larvacean houses: Rapid carbon transport to the deep
(Faroe–Shetland Channel, NE Atlantic). Mar. Biol. 151: sea floor. Science 308: 1609–1611.
1731–1741. ROE, H. S. J., D. S. M. BILLETT, AND R. S. LAMPITT. 1990. Benthic/
KORANTENG, K. A. 2001. Structure and dynamics of demersal midwater interactions on the Madeira Abyssal Plain; evidence
assemblages on the continental shelf and upper slope off for biological transport pathways. Prog. Oceanogr. 24:
Ghana, West Africa. Mar. Ecol. Prog. Ser. 220: 1–12. 127–140.
LALLIER-VERGES, E., P. BERTRAND, AND A. DESPRAIRIES. 1993. ROGER, C. 1982. Macroplankton and micronecton of the tropical
Organic matter composition and sulfate reduction intensity in Atlantic. I. Biomass and taxonomic composition. Oceanogr.
Oman margin sediments. Mar. Geol. 112: 57–69. Trop. 17: 85–96.
LE-BORGNE, R. 1983. Thaliacean proliferation in the Gulf of SMITH, C. R., AND A. R. BACO. 2003. Ecology of whale falls at the
Guinea. Oceanogr. Trop. 18: 49–54. deep-sea floor. Oceanogr. Mar. Biol. Annu. Rev. 41: 311–354.
MADIN, L. P., P. KREMER, P. H. WIEBE, J. E. PURCELL, E. H. TITELMAN, J., L. RIEMANN, T. A. SØRNES, T. NILSEN, P.
HORGAN, AND E. A. NEMAZIE. 2006. Periodic swarms of the GRIEKSPOOR, AND U. BÅMSTEDT. 2006. Turnover of dead
salp Salpa aspera in the slope water off the NE United States: jellyfish: Stimulation and retardation of microbial activity.
Biovolume, vertical migration, grazing, and vertical flux. Mar. Ecol. Prog. Ser. 325: 43–58.
Deep-Sea Res I 53: 804–819. VETTER, E. W. 1996. Enrichment experiments and infaunal
———, AND J. E. PURCELL. 1992. Feeding, metabolism and population cycles on a Southern California sand plain:
growth of Cyclosalpa bakeri in the subarctic Pacific. Limnol. Response of the leptostracan Nebaiia daytoni and other
Oceanogr. 37: 1236–1251. infauna. Mar. Ecol. Prog. Ser. 137: 83–93.
MAYZAUD, P., M. BOUTOUTE, R. PERISSINOTTO, AND P. NICHOLS. WALSH, J., G. ROWE, R. IVERSON, AND C. MCROY. 1981. Biological
2007. Polar and neutral lipid composition in the pelagic export of shelf carbon is a sink of the global CO2 cycle.
tunicate Pyrosoma atlanticum. Lipids 42: 647–657. Nature 292: 196–201.

Limnology limn-54-04-40.3d 24/4/09 14:42:36 1208 Cust # 08-493

 Pyrosoma atlanticum deposition event 1209

WEFER, G., AND G. FISCHER. 1993. Seasonal patterns of vertical YOON, W. D., J. C. MARTY, D. SYLVAIN, AND P. NIVAL. 1996.
particle flux in equatorial and coastal upwelling areas of the Degradation of faecal pellets in Pegea confoederata (Salpidae,
eastern Atlantic. Deep-Sea Res. I 40: 1613–1645. Thaliacea) and its implication in the vertical flux of organic
WEST, E. J., D. T. WELSH, AND K. A. PITT. 2009. Influence of matter. J. Exp. Mar. Biol. Ecol. 203: 147–177.
decomposing jellyfish on sediment oxygen demand and ZAR, J. H. 1999. Bioestatistical analysis. 4th ed. Prentice Hall.
nutrient dynamics. Hydrobiologia 616: 151–160.
WIEBE, P. H., L. P. MADIN, L. R. HAURY, G. R. HARBISON, AND L.
M. PHILBIN. 1979. Diel vertical migration by Salpa aspersa
and its potential for large-scale particulate organic matter
Associate editor: Peter G. Verity
transport to the deep sea. Mar. Biol. 53: 249–255.
YAMAMOTO, J., AND oTHERS. 2008. Transportation of organic
matter to the sea floor by carrion falls of the giant jellyfish Received: 18 November 2008
Nemopilema nomurai in the Sea of Japan. Mar. Biol. 153: Accepted: 12 March 2009
311–317. Amended: 15 March 2009

Limnology limn-54-04-40.3d 24/4/09 14:42:36 1209 Cust # 08-493