Hydrobiologia

, Volume 690, Issue 1, pp 227–245

Jelly-falls historic and recent observations: a review to drive future research directions

Authors

    • GEOMARHelmholtz Centre for Ocean Research Kiel
  • Kylie A. Pitt
    • Australian Rivers Institute, Coast and EstuariesGriffith University
  • Andrew K. Sweetman
    • Norwegian Institute for Water Research
    • Centre for GeobiologyUniversity of Bergen
  • Daniel O. B. Jones
    • National Oceanography Centre
  • Joan E. Cartes
    • Institut de Ciències Del Mar de Barcelona, CSIC
  • Andreas Oschlies
    • GEOMARHelmholtz Centre for Ocean Research Kiel
  • Robert H. Condon
    • Dauphin Island Sea Lab
  • Juan Carlos Molinero
    • GEOMARHelmholtz Centre for Ocean Research Kiel
  • Laetitia Adler
    • Biocenter Grindel and Zoological Museum
    • School of Geological SciencesUniversity College Dublin
  • Christian Gaillard
    • Université de Lyon 1, UMR CNRS 5125
  • Domingo Lloris
    • Institut de Ciències Del Mar de Barcelona, CSIC
  • David S. M. Billett
    • National Oceanography Centre
JELLYFISH BLOOMS Review Paper

DOI: 10.1007/s10750-012-1046-8

Cite this article as:
Lebrato, M., Pitt, K.A., Sweetman, A.K. et al. Hydrobiologia (2012) 690: 227. doi:10.1007/s10750-012-1046-8

Abstract

The biological pump describes the transport of particulate matter from the sea surface to the ocean’s interior including the seabed. The contribution by gelatinous zooplankton bodies as particulate organic matter (POM) vectors (“jelly-falls”) has been neglected owing to technical and spatiotemporal sampling limitations. Here, we assess the existing evidence on jelly-falls from early ocean observations to present times. The seasonality of jelly-falls indicates that they mostly occur after periods of strong upwelling and/or spring blooms in temperate/subpolar zones and during late spring/early summer. A conceptual model helps to define a jelly-fall based on empirical and field observations of biogeochemical and ecological processes. We then compile and discuss existing strategic and observational oceanographic techniques that could be implemented to further jelly-falls research. Seabed video- and photography-based studies deliver the best results, and the correct use of fishing techniques, such as trawling, could provide comprehensive regional datasets. We conclude by considering the possibility of increased gelatinous biomasses in the future ocean induced by upper ocean processes favouring their populations, thus increasing jelly-POM downward transport. We suggest that this could provide a “natural compensation” for predicted losses in pelagic POM with respect to fuelling benthic ecosystems.

Keywords

Biological pumpGelatinous zooplanktonJelly-fallOrganic matter

Introduction: particulate organic matter (POM) and jelly-falls

The input of POM drives secondary production and most benthic ecosystem processes in the deep-sea (Ruhl et al., 2008; Smith et al., 2008). POM inputs mainly include autochthonous particles from the euphotic zone, ranging in increasing size from phytodetritus (organic-rich material derived from phytoplankton blooms) (Beaulieu, 2002; Smith et al., 2008), marine snow (Caron et al., 1986; Alldredge & Silver, 1988), mucilaginous aggregates (Cartes et al., 2007; Martin & Miquel, 2010), mucous sheets from zooplankton (Robison et al., 2005; Lombard & Kiorbe 2010), faecal pellets (reviewed by Turner, 2002), wood particles (Turner, 1973), and macrophyte detritus (Vetter & Dayton, 1998, 1999; Cartes et al., 2010) to fish and whale carcasses (Soltwedel et al., 2003; Smith & Baco, 2003; Gooday et al., 2010).

Jelly-falls can be defined as point source organic matter inputs (as corpses/carcasses) that sink through the water column (remineralizing as dissolved organic/inorganic components), eventually causing an accumulation of jelly-POM (J-POM) at the seabed (Fig. 1). Numerous gelatinous zooplankton groups have been shown to accumulate at the ocean floor including the Cnidaria (Scyphozoa) and Thaliacea (Pyrosomida, Doliolida, and Salpida) (Table 1). The significance and magnitude of sinking material in the biological pump are primarily assessed by a variety of indirect techniques (Buesseler et al., 1992; Jahnke, 1996; Marchant et al., 1999) that cannot target the J-POM associated with jelly-falls. They include remote sensing algorithms (Behrenfeld & Falkowski, 1997; Balch et al., 2007), surface-tethered and neutrally buoyant sediment traps (Lampitt et al., 2001; Buesseler et al., 2007), and acoustic backscatter profiling sensors (ABS and ADCP) to study particles and biomass in the water column (Merckelbach & Ridderinkhof, 2006; Jiang et al., 2007). Sediment traps are the most used device, but they often underestimate the contribution of large particles and detritus (e.g. Rowe & Staresinic, 1979; but see Conte et al., 2003; Buesseler et al., 2007). Jelly-falls can only be sampled directly using techniques such as video (Wiebe et al., 1979; Lebrato & Jones, 2009), towed/still photography (Roe et al., 1990; Billett et al., 2006; Sweetman & Chapman, 2011), or benthic trawling (Sartor et al., 2003) (see Table 2 for other techniques/strategies). Therefore, although many sources of organic material have been widely studied and POM/DOM remineralization dynamics considered in biogeochemical models as a result (e.g. Burd et al., 2010), jelly-falls are relatively unexplored sources of POM, despite a significant fraction of the pelagic biomass being sequestered in the bodies of gelatinous zooplankton.
https://static-content.springer.com/image/art%3A10.1007%2Fs10750-012-1046-8/MediaObjects/10750_2012_1046_Fig1_HTML.gif
Fig. 1

Conceptual model of common processes beginning with sinking and the start of remineralization in the euphotic and twilight zones to deposition at the seabed followed by decomposition and scavenging. Under “material arriving at depth z”, we have identified five critical factors that determine the amount of material reaching the seabed. The links to “bacterioplankton” and “phytoplankton” only proceed in the euphotic/twilight zone. Jelly-falls are linked to the “jelly-pump” concept (Condon & Steinberg, 2008; Condon et al., 2010) through the production of J-DOM in the water column and at the seabed. J-DOM is organic matter that fuels other trophic levels, which can occur while the organisms are still alive (e.g. Condon et al., 2011) or when dead (Hansson & Norrman, 1995)

Table 1

A compilation of naturally occurring jelly-falls

Location

Origin

Species

Material statea

Latitude (range)b

Longitude (range)b

Depth (m)c

Survey device

Durationd

Reference

Norwegian Sea (Atlantic Ocean)

Likely Scyphozoa

Pending DNA analysise

Det.

66.14°N

3.94°E

1,380 (8.3/−1)

ROV (video)

7 days (S)

Jones et al. (2010)

Norwegian Sea (Atlantic Ocean)

Scyphozoa

P. periphylla

F/Dec.

60.40°N–60.41°N

5.09°E–5.10°E

396–443 (−/7)

Yo–Yo (towed camera)

1 day (Spr.)

Sweetman & Chapman (2011)

Japan Sea (Pacific Ocean)

Scyphozoa

A. limbata

F

42.58°N

143.96°E

320 (17/2.2)

ROV (video)

Unknown (S)

Miyake et al. (2002)

Chesapeake Bay (Atlantic Ocean)

Scyphozoa

C. quinquecirrha

F

38.59°N

76.12°W

1.5–3 (15/14)

Visual (observers)

90 days (S)

Sexton et al. (2010)f

Japan Sea (Pacific Ocean)

Scyphozoa

N. nomurai

F

35.8°N–36.3°N

136°E–135.5°E

146–354 (22/10)

VTR system (towed camera)

35 days (S–A)

Yamamoto et al. (2008)

Japan Sea (Pacific Ocean)

Scyphozoa

P. polylobata

F

34.91°N

138.65°E

453 (16/8)

ROV (video)

Unknown (S)

Miyake et al. (2005)

Santa Catalina Basin (Pacific Ocean)

Scyphozoa

Pelagia sp.

Unk.

32.46°N

117.49°W

>1,000 (17/4)

Photographs

Unknown (N)

Jumars (1976)

Bermuda (Atlantic Ocean)

Scyphozoa

Cassiopeia xamachana

F/Dec./Det.

32.34°N

64.70°W

3 (25–25)

Photographs (quadrats)

Unknown (S)

M. Lebrato (unpublished)

Gulf of Aqaba (Red Sea)

Scyphozoa

A. aurita

F

29.50°N

34.91°E

20 (25–23)

Photographs (scuba diver)

Unknown (Sp. #)

Alamaru et al. (2009)

Arabian Sea (Indian Ocean)

Scyphozoa

Probably C. orsini

F

22.95°N

66.61°E

900 (25/9.5)

WASP (towed camera)

1 day (S #)

Murty et al. (2009)

Arabian Sea (Indian Ocean)

Scyphozoa

C. orsini

F/Dec./Det.

22.58°N–23.50°N

60.65°E–59.04°E

304 –3,299 (25/2)

SHRIMP (towed camera)

17 days (W #)

Billett et al. (2006)

Japan Sea (Pacific Ocean)

Thaliacean (Doliolidae)

Not identified

F

34.40°N

150°E

150 (16/12)

Sediment trap

5 days (Sp.)

Takahashi et al. (2010)g

Tyrrhenian Sea (Mediterranean Sea)

Thaliacean (Pyrosomatidae)

P. atlanticum

F/Dec.

42.30°N

10.60°E

300–650 (25/12)

Bottom trawling

1995–1999 (Sp. S)

Sartor et al. (2003)h

Alboran Sea-Gulf of Lions (Mediterranean Sea)

Thaliacean (Pyrosomatidae)

P. atlanticum

F/Dec.

36.24°N–42.39°N

5.20°W–3.63°W

43–791 (18/13)

Bottom trawling (GOC 73)

1994–2005 (Sp. #)

Bertrand et al. (2002), MEDITS-ESi

Madeira Abyssal Plain (Atlantic Ocean)

Thaliacean (Pyrosomatidae)

P. atlanticum

F/Dec.

31.28°N

25.40°W

5,433 (20/2.2)

BATHYSNAP (fixed camera)

16 days (S)

Roe et al. (1990)

Cape Verde (Atlantic Ocean)

Thaliacean (Pyrosomatidae)

P. atlanticum

F

15.80°N

23.50°W

Unknown (26/−)

Unknown

Unknown (N)

Monniot & Monniot (1966)

Ivory Coast (Atlantic Ocean)

Thaliacean (Pyrosomatidae)

P. atlanticum

F/Dec.

5.15°N−4.94°N

4.51°W−4.49°W

26–1,275 (25/4)

ROV (video)

60 days (W #)

Lebrato & Jones (2009)

Cook Strait (Pacific Ocean)

Thaliacean (Pyrosomatidae)

P. atlanticum

Dec.

41.73°S

174.3°E

100 (15/9)

Bottom trawling

Unknown (Sp.)

Hurley & McKnight (1959)

Tasman Sea (Pacific Ocean)

Thaliacean (Pyrosomatidae)

P. atlanticum

Unk.

42°S

148°E

330–640 (14/7)

Stomach content

Unknown (W)

Cowper (1960)

Gulf of Alaska (Pacific Ocean)

Thaliacean (Salpidae)

S. fusiformis

F

58.33°N

136.83°W

1–10 (7/4)

Visual (scuba diver)

120 days (Sp.)

Duggins (1981)

Sargasso Sea (Atlantic Ocean)

Thaliacean (Salpidae)

S. aspera

F/Dec.

38.60°N–39°N

71.4°E–71.1°E

2,500–3,000 (19/3)

ROV (video)

4 days (S)

Cacchione et al. (1978)

Sargasso Sea (Atlantic Ocean)

Thaliacean (Salpidae)

S. aspera

F/Dec.

38°N–40°N

72.5°E–70°E

2,000–3,000 (19/3)

ROV (video)

30 days (S)

Wiebe et al. (1979)

Gulf of Aqaba (Red Sea)

Thaliacean (Salpidae)

Not identified

F

34.90°N

29.50°E

20 (25/23)

Photographs (scuba diver)

Unknown (Sp. #)

Alamaru et al. (unpublished)j

aMaterial state refers to the condition in which the material was found: Dec. decomposing, Det. detritus, F fresh, unk. unknown (if not stated)

bRange for latitude, longitude and depth indicates that in some cases the material was retrieved along a gradient of depths and not in isolation (see reference paper for additional information)

cIn situ surface and BT (°C) are included in parentheses when available in the original study or otherwise compiled from the GLODAP database (Key et al., 2004) and the World Ocean Atlas (http://odv.awi.de/en/data/ocean) in the nearest place available at the same depth

dDuration only indicates the time that the material was observed or surveyed at the seabed and does not indicate annual events; otherwise the time-series is given for annual depositions. The season is indicated as: N not available; Sp. spring, S summer, A autumn, W winter. # indicated when the event happened after seasonal upwelling and/or monsoon winds (e.g. tropical latitudes or specific cases like the Mediterranean Sea)

eJelly material was unidentifiable to species level. Bar-coding with mtDNA and 18S rDNA ITS regions in progress to determine the affiliation

fThe authors do not show seabed evidence. The potential POC flux to the sediments was relatively small (12.5–72.5 mg C m−2 year−1) in comparison with the total annual flux to the sediments in the area (61.2 g C m−2 year−1 Kemp et al., 1997)

gCarcasses recorded in sediment traps (export flux = 1.05 mg C m−2 day−1, sinking speed = 4,000 m day−1, small degradation observed)

hThe data used for P. atlanticum correspond to the trawling catch from the seabed. The carcasses were dead at the seabed and decomposing

iThe data were in the MEDITS-ES project (International bottom trawl survey in the Mediterranean) (http://www.sibm.it/SITO%20MEDITS/). The data for P. atlanticum correspond to the trawling catch from the seabed

jA. Alamaru also reports on the presence of salps at the seabed in the same area as A. aurita

Table 2

Sampling techniques and initiatives that may be available to monitor jelly-falls

Study method

Description

Advantages

Disadvantages

Large scale (regional)

(1) Deep ocean observatories network (e.g. EUR-OCEANS, OceanSITES, ESONET) and offshore scientific platforms (e.g. PLOCAN)

Long-term reference network stations could be used to monitor seabed processes associated with jelly-falls. They could be used to study the organisms and carcasses in the water column and their arrival at the seabed by use of camera arrays. Global distribution

The moorings and cruises are in place so it is a matter of adapting the strategy. Could have camera systems throughout the year with periodic recoveries. In situ and real time monitoring

Challenging to study the jellies in the water column with camera devices. Possibility that jelly-falls do not occur near the stations and/or the seabed may too deep for jelly-falls to be observed)

(2) Collaborations with industry (e.g. SERPENT project and similar)

The offshore oil and gas industry has regular access to expensive equipment (e.g. ROVs, camera systems) used in monitoring their own infrastructures. This equipment is not routinely used in scientific studies, but through collaboration it could be used to study jelly-falls at specific times of the year

Access to state-of-the-art expensive equipment to study the seabed in deep waters. ROVs used in industry operations follow paths, allowing transect study. They cover a larger seabed area than normal in a scientific study

Obtaining agreements with key industry personnel with access to the facilities. Confidentially and data release may take time to arrange. Surveys are confined to where the infrastructure exists. Cannot deviate greatly from established survey lines

Medium scale (local)

(3) Scientific ROV surveys

Used at known sites of jelly-falls to monitor the depositions in transects

Real-time monitoring and quantitative or qualitative data available at the seabed

Time available to conduct the survey. Total area covered. Exclude water column processes. Expensive

(4) Towed and drop cameras from a vessel

Either used at known sites of jelly-falls or use to search for depositions in transects/specific locations

Real-time monitoring and quantitative or qualitative data available at the seabed. Can cover a relatively large area

Time available to conduct the survey. Camera angle of view much less than ROV camera. Exclude water column processes

(5) Time-lapse cameras (e.g. BATHYSNAP and benthic landers)

Used at known sites of jelly-falls to study the evolution of the material over time. A network of time-lapse cameras also feasible at specific locations

Real-time monitoring and qualitative data available at the seabed. Time component of the jelly-falls

Limited area covered. No biomass data. Jelly-fall may not occur where cameras are installed. Camera angle of view restricted. Exclude water column processes

(6) Trawling (from fisheries)

The fishery industry and other commercial species surveys (e.g. MEDITS) have records of bycatch organisms trawled at the seabed, including jelly-falls (carcasses)

Quantitative or qualitative data available at the seabed. Large areas covered over bathymetric gradients. Time component often available

Obtaining agreements with industry personnel that have access to the facility. Surveys confined to the industry study of commercial species. Excludes water column processes. Environmentally destructive

(7) Acoustic/electronic tagging studies

Living individuals in large blooms could be acoustically/electronically tagged to follow their fate

Real-time study of individuals in a jelly-fall. Time component of sinking and deposition

Difficulty of tag attachment to gelatinous body. Premature release of the tag. Limited information

Small scale (local)

(8) Large sediment traps (+5 m)

If a neutrally buoyant sediment trap is developed to follow blooms it may deliver data on the associated sinking material

Quantitative data in the water column. Possible to combine with a method at the seabed

Probably unable to catch much of the sinking jelly-fall. Problems with organisms that vertically migrate and are mistakenly trapped alive. Cost-effective problems

(9) Moored and free-drifting profilers

Some in development to measure water column properties over time (McLane labs, SeaCycler). If installed with a camera, study could cover the entire water column to the seabed

Possible to monitor entire water problem over time. Quantitative data. Possible to relate camera data and water column properties

Camera installation problems. Jelly-fall may not occur where the profilers are installed. Sinking speed of carcasses not tracked by profiler

(10) Genetic tools in sediments

Sediment proxy on freshly deposited gelatinous material can be tested using mtDNA and nuclear DNA

Possible to obtain a identification (general or specific) from jelly-falls in the sediment depending on the decomposition time. Possible to combine with camera studies if material is visible. Time component may be available

Limited to very fresh depositions. DNA contamination problems. Limited area covered

The study of jelly-falls represents a major challenge in the understanding of the biological pump mainly due to technical/sampling hurdles, and although there is no consensus that the oceans will turn into a “jelly-slime” ecosystem (e.g. Jackson, 2008), gelatinous zooplankton biomass appears to be increasing in certain areas of the world’s oceans (Mills, 2001; Richardson et al., 2009; Purcell, 2012). As such, increased gelatinous biomass may translate into increased transfer of this material to the ocean floor and thus enhancing the magnitude and importance of the biogeochemical and ecological processes associated with jelly-falls. Thus, there is a pressing need for research on gelatinous zooplankton post-bloom processes.

Our primary objective is to provide a qualitative overview of historical and present records of jelly-falls, as well as the environmental context in which they were studied. Secondly, we define and conceptually model a general jelly-fall within the biological pump, including a synthesis of the factors triggering these events. We also assess the seasonality of jelly-falls from the available data and the benthic organisms that were observed feeding on the material. Our third objective is to discuss the possible consequences of increased gelatinous biomasses in the future ocean and provide a summary of the observational techniques and platforms that are, or could be used to study jelly-falls and their biogeochemical feedbacks.

Jelly-fall observations in the field

Thaliaceans

During the 1872–1876 H. M. S. Challenger expeditions, Moseley (1880) realized the potential importance of jellyfish in the biological pump by experimentally assessing the time it took a dead salp to sink 20 cm in a cylinder (~20 s). He then left the carcass in the cylinder for 1 month and noticed that it did not decompose completely. He subsequently wrote: “the deep-sea has to derive food for its inhabitants entirely from debris of animals and plants falling to the bottom from the water above them. The dead pelagic animals must fall as a constant rain of food. It might be supposed that the animal carcasses would consume so long a time in dropping to the seabed that their soft tissues would be decomposed” (Mosely 1892). This is, to the best of our knowledge, the first mention of jelly-falls in the literature. A number of both quantitative and qualitative studies have followed since then (Table 1), but they still remain scarce when compared with studies that have assessed the importance of other POM vectors (Turner, 2002).

Hurley & McKnight (1959) were the first to report on a natural jelly-fall when they found the thaliacean Pyrosoma atlanticum Peron 1804 on the seabed off New Zealand. The organisms were sampled with a bottom trawl between 160- and 170-m depth (bottom temperature (BT) = 9°C) during spring and were described as “resting” on the seabed. Their observations are further supported by reports from the same area of seabed being covered in P. atlanticum carcasses in 1952 (H. B. Fell pers. obs. reported to Hurley & McKnight, 1959) and reports that local fishermen frequently trapped large quantities of moribund carcasses at certain times of the year. Similar fishermen’s reports occur in the Mediterranean Sea (e.g. Sartor et al., 2003). Later, in the Tasman Sea, Cowper (1960) found that the stomachs of freshly caught carangid fish were full of P. atlanticum carcasses during winter. All fish were caught close to the bottom (BT = 7°C); therefore, the authors concluded that they were feeding either on recently settled carcasses or on moribund individuals on or near the seabed. A further analysis of stomach contents from the same fish species in the Tasman Sea from January to October revealed that the carcasses were most abundant in stomachs in January and March (Cowper, 1960). There are other observations in New South Wales, Australia of the giant pyrosomid Pyrosoma spinosum Herdman 1888 near to or deposited on rocky bottoms, and also portions of salps being recovered from stomachs of carangid fish feeding at the seabed (Griffin & Yaldwyn, 1970).

A considerable number of more recent studies document pyrosome falls. In the tropical Atlantic (off Cape Verde), Monniot & Monniot (1966) recorded moribund P. atlanticum at the seabed. In the deep Atlantic Madeira Abyssal Plain, high densities of pyrosomids were observed in the first 800 m of the water column (Roe et al., 1990). A survey using a fixed camera photographed a single carcass in an advanced state of decomposition at 5,433-m depth (BT = 2.2°C). A starfish and a crustacean scavenged the carcass, which took >16 days to decompose completely. Recently, Lebrato & Jones (2009) reported a vast jelly-fall of P. atlanticum off the Ivory Coast, West Africa during ROV (remotely operated vehicle) surveys. Decomposing carcasses formed large patches (~1–20 m2) and accumulated in troughs and channels (to a thickness of at least 0.5 m) from the shelf (<200 m) to the deep slope (>1,200 m) (BT = 4°C). The organic carbon contribution was estimated to be more than 20 g C m−1 in some areas, which is almost ten times the annual fluxes in the area, as measured by sediment traps (Wefer & Fischer, 1993). Carcasses were very abundant (707 individuals 100 m−2) at the maximum depth surveyed (1,275 m) and the maximum depth of the deposit could not be determined. Megafauna (including echinoderms and crustaceans) were observed 63 times directly feeding on the material (Lebrato & Jones, 2009). In the Mediterranean Sea (Alboran Sea to the Catalan Sea), jelly-falls of P. atlanticum were identified and sampled from 1994 to 2005 (spring and summer) during bottom trawling down to 800 m (average BT = 13°C) in the MEDITS-ES surveys (Bertrand et al., 2002) (Fig. 2B). The catch often exceeded 300 carcasses per haul. This dataset provided the first evidence of jelly-falls encompassing entire continental margins during a period of 12 years (Fig. 2B). Living P. atlanticum were recovered from benthic trawls in canyon heads and walls (Cartes et al., 2009) near the wind-driven upwelling region of the Gulf of Lions (Johns et al., 1992). Sartor et al. (2003) reported catches in benthic trawls from 1995 to 1999 in the Mediterranean Sea (Tyrrhenian Sea) with numerous P. atlanticum occurring at the seafloor (100–500 g h−1 during >500 h over several km2) at 300 and 650 m (BT = 12°C) (Fig. 2B). In the Mediterranean Sea, benthic deposits of P. atlanticum seem to be a common feature that are generally unnoticed.
https://static-content.springer.com/image/art%3A10.1007%2Fs10750-012-1046-8/MediaObjects/10750_2012_1046_Fig2_HTML.gif
Fig. 2

A Global distribution of reported jelly-falls. Also included are the species that were recorded in each individual event (see Table 1 for detailed information). B Observations of P. atlanticum jelly-falls at the seabed in the Mediterranean Sea (from the MEDITS-ES project) (Bertrand et al., 2002). Numerous jelly-falls occur along the whole western Iberian Margin. The legend shows the average number of carcasses observed at each station from 1994 to 2005. Also included are observations of P. atlanticum in the Tyrrhenian Sea (northwest Mediterranean Sea) (Sartor et al., 2003). The bathymetric line (200 m) are from the general bathymetric chart of the oceans (GEBCO) Digital Atlas (IOC et al., 2003). The dotted line indicates the zones trawled in the MEDITS project that can be used to study jelly-falls from trawling data, as proposed in “Operational oceanography and exploration techniques” section

Thaliaceans other than P. atlanticum have also been recorded at the seabed. Cacchione et al. (1978) described sinking living/moribound Salpa aspera Chamisso 1819 in the water column from a series of ROV observations below 2,500 m (BT = 3°C) in the Hudson Canyon, northwest Atlantic, over 30 days during summer. Salp bodies were observed rolling down the canyon. In the same area, Wiebe et al. (1979) observed a jelly-fall of S. aspera at >2,000 m (BT = 3°C). The carcasses accumulated in channels and furrows and formed string-like aggregations at the seabed (see Grassle & Morse-Porteus, 1987; Grassle & Grassle, 1994). In the Pacific Ocean, Duggins (1981) reported thousands of Salpa fusiformis Cuvier 1804 in the intertidal/subtidal environment (BT = 4°C) of the Alaska Gulf over several months. Echinoderms fed preferentially on the gelatinous resource as soon as it was available. Recently, in the Red Sea, salps have formed jelly-falls during spring and after upwelling (20 m, BT = 23°C) although these observations were not quantified (A. Alamaru pers. comm). In the Sea of Japan, a doliolid jelly-fall was studied in the water column by means of a sediment trap below 150 m (temperature = 12°C) (Takahashi et al., 2010).

Cnidarians

For Cnidaria, the first natural jelly-fall recorded was in a photographic survey (Jumars, 1976) below 1,000 m (BT = 4°C), where ophiuroids congregated around a Pelagia sp. carcass in the Santa Catalina basin (northeast Pacific). More recently, jelly-falls of Aurelia limbata Brandt 1835, Parumbrosa polylobata Kishinouye 1910, and Nemopilema nomurai Kishinouye 1922 were reported on the seafloor down to 400-m depth (BT = 2.2–10°C) in the Sea of Japan during summer and autumn (Miyake et al., 2002, 2005; Yamamoto et al., 2008, respectively). Thousands of Crambionella orsini Vanhöffen 1888 carcasses were photographed using a towed camera at the seabed during winter and after seasonal upwelling in the Arabian Sea (Billett et al., 2006). Carcasses were recorded as freshly deposited on the shelf, while ‘jelly-lakes’ of decomposing detritus were observed on the continental rise deeper than 3,000 m (BT = 2°C). White mats, assumed to be bacteria decomposing and remineralizing the organic material, covered the detritus. A scyphozoan jelly-fall (probably C. orsini) also was reported near the Pakistan Margin at 900 m (BT = 9.5°C) during summer and after seasonal upwelling (Murty et al., 2009). A large gelatinous mat covering the seabed, presumably scyphozoans in a very advanced state of decomposition, was surveyed for 7 days with a ROV in the Norwegian Sea at 1,380 m (BT = −1°C) during summer (Jones et al., 2010). A jelly-fall of Periphylla periphylla (Peron & Lesueur, 1810) was studied in spring 2011 in the Lurefjorden, Norway between 396 and 443 m (BT = 7°C) (Sweetman & Chapman, 2011). Carcasses were documented with a camera in two different areas in seven transects at very low densities (0.01 carcass m−2), estimated to contribute <1% to the annual organic matter flux in the area. Numerous jelly-falls of Aurelia aurita Linnaeus 1758 occurred during spring and after upwelling events at 20-m depth in the Red Sea (BT = 23°C) (Alamaru et al., 2009). Sexton et al. (2010) reported a jelly-fall of Chrysaora quinquecirrha Desor 1848 medusae in a shallow sub-estuary of Chesapeake Bay during autumn.

Jelly-falls conceptualization

Processes from the euphotic zone to the seabed

A jelly-fall (Fig. 1) starts when gelatinous organisms die and sink from the so-called death depth subject to the organisms’ vertical migration and displacement. Because gelatinous detritus is denser than the surrounding seawater, the corpses sink through the water column at a rate determined by the material’s size and excess density (Stokes’ Law) (e.g. Yamamoto et al., 2008). The organisms can settle at the seabed while still alive (Wiebe et al., 1979; Gili et al., 2006) and then die, thus remineralization can start on the seabed. As it sinks, the material can be consumed by scavengers and return to the faunal food web or be remineralized by bacteria (bacterioplankton) and enter the microbial loop (e.g. Hansson & Norrman, 1995). Dissolved organic matter (DOM) leaching from living or dead organisms provides a link to the “jelly-pump” concept (J-DOM) of microbial communities being fuelled by DOM excretion (Condon & Steinberg, 2008; Niggl et al., 2010; Condon et al., 2011) (Fig. 1). Microzooplankton and small zooplankton may also consume J-DOM (e.g. Iguchi et al., 2006; Titelman et al., 2006; West et al., 2009a). Laboratory incubations of scyphozoan material using deep (334 m) and shallow water (<10 m) differed in the remineralization time (Iguchi et al., 2006), which was attributed to the microbial community as well as temperature in situ. Differences in the lability of gelatinous tissues (C:N ratios; Larson, 1986), the various rates at which the different materials sink (Apstein, 1910; Mills, 1981), and rates of scavenging and bacterial mineralization (which may vary with temperature and depth) greatly influence the extent to which the jelly-fall is recycled within the water column versus at the seabed (Fig. 1). Jelly-falls that reach the seafloor may be transported elsewhere (e.g. along geomorphological features) (Billett et al., 2006), or retained in situ and consumed by the local faunal and microbial community (Lebrato & Jones, 2009). Leaching of dissolved compounds fuels production in higher trophic levels (West et al., 2009a) and biogeochemical processes such as oxygen consumption in the water and in the sediment proceed during the organic enrichment (West et al., 2009b; Sexton et al., 2010). Associated total alkalinity changes from excess DOM (Hansson & Norrman, 1995; Hoppe et al., 2010) and the non-Redfield stoichiometry of nitrogen and phosphorus leaching from the corpses (Pitt et al., 2009; Condon et al., 2010; Tinta et al., 2010) should also be considered. The decomposition dynamics has been the focus of several papers targeting a variety of species at different temperatures, thus decay rates (k) are available (e.g. Titelman et al., 2006). The turnover of J-POM is rapid during the first few days (Sempere et al., 2000) and then slows down, but it is highly dependent on temperature (Iguchi et al., 2006). These quantitative data on decomposition dynamics have enabled remineralization of sinking carcasses to be modelled in open ocean conditions (Lebrato et al., 2011). They provided a new metrics based on decay rate, temperature fields, ‘death depth’, and sinking speed that helps to understand why different gelatinous zooplankton groups transfer organic carbon to the seabed (e.g. scyphozoans and thaliaceans), while others may be completely remineralized in the water column.

The transport to the seafloor of J-POM is an important source of labile material to the whole size-spectrum of benthic communities in continental margins and the deep-sea (Table 1; Fig. 1). Evidence of organisms consuming J-POM at the seabed has accumulated slowly from photographs and videos (Table 1; Fig. 2A). Gelatinous material has a low energy content (0.5–6 gross energy kJ g dry mass−1) compared to other types of carrion such as fish (5–22 gross energy kJ g dry mass−1) or algae (>10 gross energy kJ g dry mass−1) (Doyle et al., 2007). Among gelatinous species, the energy content is highest in salps and pyrosomids (4–6 gross energy kJ g dry mass−1) (Davenport & Balazs, 1991; Clarke et al., 1992), which are important parts of the diets of numerous benthic organisms (Table 3). Although high energy resources are readily available on continental margins, food is a limiting factor in the deep-sea (Gage & Tyler, 1991). Thus, jelly-falls may represent a valuable nutritional input at certain times of the year (Table 3). Unlike other large food falls, which are usually sparsely scattered over the sea floor, gelatinous corpses accumulate in large patches (Billett et al., 2006; Lebrato & Jones, 2009) making it easier for scavengers to locate; however, scavengers traditionally observed around fish falls (such as isopods or fish) have not been observed around jelly-falls (Sweetman & Chapman, 2011). The reduced energy spent searching for food, and the lability of the gelatinous carrion relative to other sources of detritus, may compensate for the reduced energy density of the jelly-falls at least for some scavenger species (Doyle et al., 2007). Additionally, jelly-falls may provide an environment for macrofauna/microbial communities to proliferate, which, in turn, may be preyed upon by other taxa (Sweetman & Chapman, 2011). Sessile organisms (anthozoans, including hexacorallians, octocorallians, and scleractians) also consume J-POM (Gili et al., 2006; Alamaru et al., 2009; Lebrato & Jones, 2009). Echinoderms dominate scavenging observations at any depth, followed by crustaceans and fish (Table 3). Remains of J-POM (e.g. Cymbulia peroni De Blainville 1810) are commonly found in guts of numerous benthic decapods, such as the Norway lobster Nephrops norvegicus Linnaeus 1758, the crab Geryon longipes Milne-Edwards 1882 (in Cartes, 1993a), and the squat lobster Munida tenuimana Sars 1872 (in Cartes, 1993b). J-POM (Iasis zonaria Pallas 1774, P. atlanticum, P. periphylla) is also found in the guts of deep shrimps, such as Plesionika martia Milne-Edwards 1883 (in Fanelli & Cartes, 2008) or fish (Carrasson & Cartes, 2002; Drazen et al., 2008; Goldman & Sedberry, 2010). Any remaining material that is not channelled through macro/megafaunal scavenging will eventually be respired by microbial communities (Fig. 1). The build-up of impermeable gelatinous material (as in Billett et al., 2006) on the seafloor leads to reductions in O2 flux into sediments (West et al., 2009b). This would favour microbial over metazoan biomass and remineralization processes, although low seawater O2 combined with no light slows microbial decomposition of settling organic matter (Gooday et al., 2010), and toxic remineralization products (e.g. ammonium and free sulphides) could accumulate and seriously impact sediment biota as well as pelagic ecosystems (Titelman et al., 2006; Pitt et al., 2009) (Fig. 1). Ultimately, jelly-falls could induce spatial heterogeneity in the biodiversity of benthic communities (Gooday et al., 2010) as a consequence of the mass accumulation of undegraded labile material.
Table 3

Occurrences of jelly-falls and the megafaunal taxa feeding on them

Year

Depth (m)

Taxon

Timinga

Duration

Units

Feeding taxa

Reference

2011

396–443

P. periphylla

Mar—Sp.

1

days

Crustaceansb

Sweetman & Chapman (2011)

2010

150

Doliolids

May—Sp.

5

None

Takahashi et al. (2010)

2009

20

Salps

Post-upwelling

Anthozoans

Alamaru et al. (unpublished)

2009

20

A. aurita

Post-upwelling

Anthozoans

Alamaru et al. (2009)

2009

900

C. orsini

Post-upwelling

None

Murty et al. (2009)

2009

1,380

Scyphozoans

Jun—S

7

days

None

Jones et al. (2010)

2007

3

C. xamachana

Sep—S

None

M. Lebrato (unpublished)

2006

146–354

N. nomurai

Sep/Oct—S/A

30

days

Crustaceans, echinoderms

Yamamoto et al. (2008)

2006

26–1,275

P. atlanticum

Post-upwelling

60

days

Severalc

Lebrato & Jones (2009)

2005

1.5–3

C. quinquecirrha

Jun/Sep—S

90

days

None

Sexton et al. (2010)

2003

304–3,299

C. orsini

Post-upwelling

17

Crustaceans, echinoderms

Billett et al. (2006)

2002

453

P. polylobata

Sep—S

Echinoderms

Miyake et al. (2005)

2001

320

A. limbata

Aug—S

Echinoderms

Miyake et al. (2002)

1999

300–650

P. atlanticum

Sp./S

3

months

None

Sartor et al. (2003)

1998

300–650

P. atlanticum

Sp./S

3

months

None

Sartor et al. (2003)

1997

300–650

P. atlanticum

Sp./S

3

months

None

Sartor et al. (2003)

1996

300–650

P. atlanticum

Sp./S

3

months

None

Sartor et al. (2003)

1995

300–650

P. atlanticum

Sp./Sum

3

months

None

Sartor et al. (2003)

1985

5,433

P. atlanticum

Jun/Jul—S

17

days

Crustaceans, echinoderms

Roe et al. (1990)

1978

1–10

S. fusiformis

Mar/Jun—Sp./S

3

months

Echinoderms

Duggins (1981)

1975

2,500–3,000

S. aspera

Aug—S

None

Cacchione et al. (1978)

1975

2,000–3,000

S. aspera

Aug—S

4

days

None

Wiebe et al. (1979)

1955

330–640

P. atlanticum

Jun/Jul—S

Fish

Cowper (1960)

1952

100

P. atlanticum

Oct–Sp.

Fish

Hurley & McKnight (1959)

aThe month is abbreviated (when available), and the season is indicated as: Sp. spring, S summer, A autumn, W Winter

bCaridean shrimps grazed on carcasses, and density was higher around jelly-falls compared to non jelly-falls settings. Galatheid crabs were observed near carcasses, but no grazing was observed

cAnthozoans, crustaceans, echinoderms, fish, arthropods, polychaetes

Causes and seasonality of jelly-falls

Factors driving the onset of jelly-falls are mostly linked to the ageing and end of a bloom (Purcell et al., 2001) and a long-term cumulative effect of negative factors, such as parasitism, starvation, infection, and predation (Mills, 1993), with subsequent deposition at the seabed if the material is not completely remineralized while sinking. In other cases, the material floats and it is washed ashore (e.g. Pakhomov et al., 2003; Houghton et al., 2007). The life history of individual species dictates their fate, although some generalities apply to all groups, such as seasonal disappearance from the waters (Mills, 1993). Life cycles are often completed within a year or a few months, with subsequent death (see Franqueville, 1971; Mills, 1993). For thaliaceans, there is evidence that high concentrations of particles and suspended organic matter [e.g. chlorophyll a >1 mg m−3; Perissinotto & Pakhomov (1998)] clog their feeding apparatus causing death (Acuña, 2001) despite food being abundant (Harbison et al., 1986; Zeldis et al., 1995). This explains the salp jelly-fall studied by Duggins (1981) in the subtidal zone in Alaska and the beaching of salps reported by Pakhomov et al. (2003) in the Southern Ocean. Thaliacean jelly-falls tend to appear at the seabed after strong periods of upwelling (Lebrato & Jones, 2009) or after the spring bloom months when chlorophyll a levels are high (Wiebe et al., 1979; Duggins, 1981; Roe et al., 1990) (Table 3). Re-assessment of the season (n = 24) when carcasses of all groups arrive at the seabed indicates that >75% of jelly-falls occur after the spring bloom in temperate/subpolar areas and >25% in post-upwelling periods in the tropics. This happens irrespectively of the depth at which they are deposited. It remains unclear for thaliaceans if the concentration or particles per se causes clogging and subsequent death, or if the biological composition of the particles and autotroph community play a role. Potential connections between climate and gelatinous zooplankton populations in the water column and the jelly-falls at the seabed have not yet been investigated. In tropical areas, monsoon patterns trigger upwelling events that alter water column properties (e.g. lower temperature, high nutrients and chlorophyll a, high DOM levels, higher POM export) (Coble et al., 1998; Honjo et al., 1999), thus forcing in these zones is different than in temperate/subpolar latitudes.

In the Cnidaria, several variables may trigger the onset of jelly-falls, including sudden or sustained changes in temperature exceeding physiological performance (Gatz et al., 1973) (relevant in upwelling systems where organisms can experience rapid changes in the water mass properties due to physical forcing), ageing of the bloom followed by food depletion (causing starvation and poor nutrition) (Mills, 1993; Purcell et al., 2001; Sexton et al., 2010). The latter cause may have relevance for the C. orsini carcasses studied by Billett et al. (2006) and the depositions of N. nomurai observed by Yamamoto et al. (2008). The food exhaustion hypothesis would explain why we often observe scyphozoan jelly-falls after the spring bloom but predominantly in the late spring/early summer months (Table 3). Other factors include grazing damage (Arai, 2005), parasitism/injury/viral infections (Mills, 1993), senescence (Sexton et al., 2010), extreme weather events triggering large changes in physical properties of water (Cargo, 1976), and sinking driven by low temperatures and inducing deposition and later death owing to temperature changes (Sexton et al., 2010).

Operational oceanography and exploration techniques

The jelly-fall concept originates from a handful of studies undertaken in the field that either described accidental encounters or, in few cases, targeted known gelatinous depositions. In >80% of the cases (n = 22), ROV video and/or towed/still cameras were used as the sampling technique (Table 1). Unless a large area was covered and transects used to count individual carcasses (Billett et al., 2006; Lebrato & Jones, 2009), these techniques remain qualitative (Roe et al., 1990; Miyake et al., 2002, 2005; Yamamoto et al., 2008). Other techniques, including scuba diving and sediment traps account for <5% of the observations. Trawling is the only other technique that allows large quantitative studies (MEDITS-ES dataset; Sartor et al., 2003). Field work has been accompanied by a series of laboratory or mesocosm studies that target associated biogeochemical processes (e.g. Sempere et al., 2000; Pitt et al., 2009; Tinta et al., 2010). Although we now have important information about the occurrence of jelly-falls and their potential influence on elemental cycling, we still lack combined effort and large-scale projects on this topic. Temporal monitoring can be addressed by ‘ocean observatories’ (Table 2; Claustre et al., 2010; Send et al., 2010). From these ocean observatory initiatives (e.g. EUR-OCEANS, OceanSITES, ESONET) and scientific projects that collaborate with offshore industries (e.g. SERPENT (Jones, 2009), DELOS (http://www.delos-project.org/), and HAUSGARTEN (Soltwedel et al., 2005), regular access to the deep-sea will increase our chances of making informative observations. We need to move beyond the present semi-empirical state of understanding to local or regional monitoring and quantification of jelly-falls. ROVs, AUVs (autonomous underwater vehicles), benthic landers, and towed, drop, and time-lapse cameras should be used (Table 2). In particular, the use of repeated AUV surveys or a network of time-lapse cameras strategically placed at the seabed in areas where jelly-falls have been observed could provide insights into seasonality and decomposition at the seabed. Benthic crawlers (Karpen et al., 2007) can survey inaccessible areas where jelly-falls have been observed via a optical cable from a shore-based station for long periods of time.

For large-scale quantification of jelly-falls, logbooks of bottom-trawling surveys from historical to present times are a unique tool that have not fully utilized. They mainly target commercial demersal fish and crustaceans species, but non-commercial or ‘discarded’ (bycatch) species, including gelatinous zooplankton, are sometimes consistently recorded (e.g. Sartor et al., 2003; Sanchez et al., 2003; Bastian et al., 2011). Data from jelly-falls have been collected in this way (Fig. 2B) and also data on living biomass (Bastian et al., 2011). Many benthic trawling programmes exist worldwide [e.g. MEDITS (International bottom trawl survey in the Mediterranean Sea) (Bertrand et al., 2002); Relini (2000) (Italian Seas); Sartor et al. (2003) (Tyrrhenian Sea); International Bottom Trawl Survey (ITBS); NOAA Gulf of Alaska bottom trawl survey; NEFSC bottom trawl survey (Gulf of Maine Area); Wilkins et al. (1998) (The 1995 Pacific West Coast bottom trawl survey); Bastian et al. (2011) (North Atlantic Ocean)]. Information could also be retrieved from fisheries information networks [e.g. PacFIN (http://pacfin.psmfc.org/index.php); AKFIN (http://www.akfin.org)]; and from state and wildlife agencies and fishery management councils (e.g. http://pacfin.psmfc.org/pacfin_pub/links.php)]. Trawling surveys normally cover specific depth ranges in the so-called trawlable areas in the shelves and slopes. The surveys do not normally work beyond the continental slope [(e.g. 0–800 m in the MEDITS-ES, 250–800 m in Sartor et al. (2003)] (Fig. 2B), but effectively sample the shelves consistently and repeatedly. The problem often is that to reduce cost and effort and increase efficiency, the size, weights, and numbers only of commercial species are recorded in logbooks, and the living and dead gelatinous component, if present, is overlooked. This issue was discovered in the MEDITS-ES project, where certain partners recorded the same data for commercial and for non-commercial species (jelly-fall data used in Fig. 2B), while the majority did not. Only through effective science-industry communication and collaboration can we make use of their potential to quantify jelly-falls and living biomass (Bastian et al., 2011).

At local scales, we suggest use of acoustic/electronic tagging (e.g. Seymour et al., 2004; Gordon & Seymour, 2008; Hays et al., 2008) on individuals found in blooms to discover their fate (Table 2). Tags can be mechanically secured in cnidarians in the bell area and peduncle, or using setting glue. Large neutrally buoyant sediment traps also could be used (Lampitt et al., 2008) that could drift under blooms, as well as free-drifting profilers with mounted cameras to investigate the water column. Genetic tools (e.g. Reusch et al., 2010) could also be used to characterize a jelly-fall signature in the sediment. Further research should quantify and study the diversity of the scavenging communities attracted to an ‘artificial’ jelly-fall (e.g. Yamamoto et al., 2008). This has traditionally been done with a bait in the field of view of a still camera (reviewed by Bailey et al., 2007). This can be combined with labelling studies to assess the fate of jelly-derived organic material, as for phytodetritus (Middelburg et al., 2000; Witte et al., 2003; Franco et al., 2008).

Can jelly-falls provide ecosystem services in the future?

The future ocean is expected to be a warmer, more-stratified, acidic, and oxygen-poor system characterized by reduced upwelling (Cox et al., 2000; Gregg et al., 2003). As a result, production exported to depth is expected to be reduced as phytoplankton communities shift from large diatom-based assemblages to picoplankton with lower export efficiency (Buesseler et al., 2007; Smith et al., 2008). Reduced export production and changes in community structure are expected to result in reduced delivery to, and an overall change in the composition of organic material reaching the abyssal ocean floor (Laws, 2004; Smith et al., 2008). This is expected to reduce food availability to the already food-limited deep-sea floor, causing a decline in deep-sea biomass and ecosystem changes (e.g. in faunal behaviour (Kaufmann & Smith, 1997; Wigham et al., 2003), bioturbation (Smith et al., 2008; Vardaro et al., 2009), faunal densities (Ruhl & Smith, 2004; Ruhl, 2007, 2008; Smith et al., 2008), reproductive traits (Tyler, 1988; Young, 2003; Ramirez-Llodra et al., 2005), faunal diversity (Levin et al., 2001), body size (McClain et al., 2005), taxonomic composition (Ruhl & Smith, 2004), sediment infaunal response (Sweetman & Witte, 2008a, b), and dominance (Cosson et al., 1997; Sweetman & Witte, 2008b). Reduced carbon export may also inhibit the ocean’s ability to sequester carbon (Smith et al., 2008). The potential consequences of the combination of altered food inputs to the benthos and increased CO2 content of seawater on ecosystem functioning and services (e.g. nutrient regeneration, energy transfer to higher trophic levels) could have large implications because recent studies suggest that an organism’s ability to cope with acidification and elevated water temperatures may be regulated by food supply (Wood et al., 2008; Gooding et al., 2009).

Gelatinous zooplankton populations, on the other hand, may benefit from anthropogenic impacts on the marine environment (Purcell et al., 2007; Purcell, 2012). There is evidence of some populations increasing during the last decades, such as thaliaceans in the Southern Ocean (Loeb et al., 1997; Atkinson et al., 2004) and jellyfish in the Mediterranean Sea (Molinero et al., 2008). It has been suggested that jelly-biomass will become an increasingly important component in the future ocean (Purcell et al., 2007; Jackson, 2008; Richardson et al., 2009; Purcell, 2012). Therefore, if classic POM vectors (e.g. phytodetritus) become less important in the future ocean, an increased amount of J-POM sinking to the seabed could mitigate some of the losses of carbon from phytoplanktonic carbon sources, although it is likely to be much more heterogeneous at the seafloor (Gooday et al., 2010). Because the majority of jelly-falls deposits are located in deep, cold (<10°C) marine environments (Table 1), we hypothesize that J-POM:phytodetrital-POM flux ratios are likely to be higher in deep-sea and polar settings (Lebrato et al., 2011). This may maintain certain ecosystem functions in some areas (dependent on the threshold POM flux) by ensuring a continued minimum POM flux from the surface to the seafloor. It is also likely to have drastic implications for benthic community composition just as changes in surface phytoplankton community composition can substantially modify abyssal community composition.

Acknowledgments

We are grateful to the scientific and environmental ROV partnership using existing industrial technology project (SERPENT) for enabling access to data off west Africa and in the deep Norwegian Sea. We thank the following contributions from individuals: L. Gil de Sola, C. García, J. Pérez Gil, and P. Abelló from the I.E.O (Fuengirola Oceanographic Center, Spain) for the facilities providing MEDITS-ES data, Brian J. Bett and R. S. Lampitt from the National Oceanography Center Southampton, UK provided unpublished data of C. orsini from Billett et al. (2006) and from Roe et al. (1990). This work was also supported by the “European Project on Ocean Acidification” (EPOCA), which is funded from the European Community’s Seventh Framework Programme (FP7/2007–2013) under grant agreement no 211384. EPOCA is endorsed by the International Programmes IMBER, LOICZ and SOLAS. This work was funded by the grant Becas mineras. exp. 210001 to M. Lebrato and by the Kiel Cluster of Excellence “The Future Ocean” (D1067/87).

Copyright information

© Springer Science+Business Media B.V. 2012