The Mercator mud volcano, located in the Gulf of Cadiz off the coasts of Spain, Portugal, and Morocco (Figure 1), may pro-vide an accessible field laboratory for studying local active venting and its possi-ble internal and external controls. The recent discovery of the first active deep ocean ‘brown smoker’ chimney in this area can possibly be linked with the disintegra-tion of a gas hydrate layer between the sea-floor and a subsurface level that is depen-dent on pressure and temperature.

For more than a decade, the international marine scientific community has deployed considerable efforts in exploring the Gulf of Cadiz. Since the discovery of the Gulf’s first mud volcano in 1999, research cruises have steadily unveiled one mud volcano after another [Gardner, 2001; Pinheiro et al., 2003; Somoza et al., 2003]. These mud volcanoes are clustered in several fields on the Portu-guese, Spanish, and Moroccan margins (Fig-ure 1). Extensive geophysical evidence of shallow gas and subsurface fluid flow has been reported in the Gulf of Cadiz [Baraza and Ercilla, 1996; Pinheiro et al., 2003].

Rodrigues and Cunha [2005] described cold seep communities (specific deep-sea fauna around fluid hydrocarbon escape ways) in the area. The highest number of specimens in these communities has been collected in the Moroccan sector, especially on mud volcanoes that show evidence of relatively recent (between now and Upper Quaternary outflows) activity. Despite num-erous observations of carbonate crusts, fossil vent pipes, and chemosynthetic pogonopho-ran worms—which are signs of active vent-ing (the process of cold seepage: when fluid actively escapes from the sea floor)—no such venting had yet been reported.

The 2002 discovery of giant mud volca-noes and cold-water coral communities at

depths between 200 and 800 meters off Larache, Morocco [Van Rensbergen et al., 2005a] has moved this part of the Moroccan margin to the forefront of European and international projects. These include the UNESCO Intergovernmental Oceanographic Commission’s Training-Through-Research and Geosphere-Biosphere Coupling Pro-cesses projects, the European Science Foun-

dation’s (ESF) Euromargins program, and the European Union’s (EU) Sixth Framework Programme (FP6) integrated project HERMES (Hotspot Ecosystem Research on the Margins of European Seas).

Furthermore, in May 2005 the CADIPOR 2 cruise of the research vessel Belgica sur-veyed this exceptional area, in particular using a video frame (a steel frame in which deep-sea cameras are mounted).

During a survey of a proposed Integrated Ocean Drilling Program site on the south-ern flank of the Mercator mud volcano (Fig-ures 1 and 2), an active chimney was for the first time sighted in this region in a water depth of about 400 meters. The Mercator

Volume 86 number 49

6 december 2005

pages 509–520

Eos, Vol. 86, No. 49, 6 December 2005

EOS, TranSacTiOnS, amErican GEOphySical UniOn

PAGeS 509, 511

First Sighting of Active Fluid Venting in the Gulf of Cadiz

By D. Van Rooij, D. DepReiteR, i. BouimetaRhan, e. De BoeVeR, K. De RycKeR, a. FouBeRt, V. huVen-ne, j. RéVeillauD, p. StaelenS, j. VeRcRuySSe, W. VeRSteeg, anD j.-p. henRiet

Fig. 1. (a) Map of the El Arraiche mud volcano field, based on EM1002 multibeam bathymetry data. Contour spacing is 20 meters. (b) Map of the Gulf of Cadiz and the Moroccan margin. Triangles represent mud volcanoes. The solid triangle denotes the Mercator mud volcano. The dashed ellipse indicates the location of the El Arraiche mud volcano field. (c) Detailed map of the Mercator mud volcano, with indication of the seismic profiles shown in Figure 2. The location of the video line and location of the screen captures are shown in Figure 3.

Eos, Vol. 86, No. 49, 6 December 2005

mud volcano, with a diameter of about 2.5 kilometers, towers about 200 meters above the seafloor [Van Rensbergen et al., 2005a, 2005b], and it features an intriguing internal reflector that resembles a bottom simulating reflector (BSR) [Depreiter et al., 2005].

On video sequences collected during the cruise, the mud volcano flank generally showed a muddy, bioturbated seafloor with rocks that were conspicuous by their pres-ence and resemblance to chimneys. One of the cone-shaped rocks appearing below the camera featured a steady flux of white-brown suspended matter (Figure 3). The hor-izontality of the plume indicates that a ben-thic current may sweep the flank of this mud volcano. The observation of this new brown smoker-type chimney could indicate that many of the surrounding conical rock out-crops actually are the exposed heads of extinct chimneys, the product and proof of previous fluid venting.

Regional Context of Mud Volcanism and Fluid Venting

The geological setting of the Gulf of Cadiz is complex. The high tectonic deformation area is characterized by the presence of an accretionary wedge formed during the mid-dle of the Miocene Epoch by the westward motion of the front of the Gibraltar Arc. Mud volcanism in the Gulf of Cadiz is closely associated with this accretionary wedge-type setting, which is related to the Cenozoic Era convergence of the African and Eurasian plate boundaries.

The El Arraiche mud volcanoes are located at the southeastern extremity of the Spanish-Moroccan field (Figure 1). Van Rensbergen et al. [2005a] described eight

mud volcanoes of various sizes. Morpho-logically, these mud volcanoes feature a typical conical shape, with varying slope

profiles. Acoustically, they are character-ized by a generally reflection-free seismic facies (Figure 2), showing stacked outflow lenses (stacked products (mud outflows) of mud volcano ‘eruptions’) intruding within the stratified sedimentary series [Van Rensbergen et al., 2005b]. These fea-tures argue for the episodic expulsion of fluidized sediment.

In contrast to the general convergent set-ting of the Gulf of Cadiz, the El Arraiche area is characterized by superficial, detached extensional tectonics, expressed as large rotated blocks of sedimentary cover bound by large normal faults, controlling the Plio-Pleistocene sedimentary accommodation space (the available space for sediment deposition (i.e., a basin)). The El Arraiche mud volcanoes are positioned above these faults, which may serve as fluid migration pathways. Gas generation in the Mesozoic source rocks and focused fluid flow along the basal detachment of the accretionary wedge are both possible sources for sustained and repeated fluid injection into the upper sedi-ment [Van Rensbergen et al., 2005a].

Depreiter et al. [2005] inferred the pres-ence of a gas hydrate stability zone (GHSZ) inside the Mercator mud volcano from the occurrence of a BSR-like feature on seismic profiles. Gas hydrates occur within a restricted zone, the GHSZ, that is bound by well-described temperature and pressure

Fig. 2. Renard Centre of Marine Geology (RCMG) high-resolution single-channel reflection seismic profiles C020557 (West-East) and C020551 (South-North). Both profiles were acquired using an 80-electrode 500 joule sparker (high-voltage seismic source) over the Mercator mud volcano (Figure 1c). Both profiles clearly show the presence of a BSR-type reflector and the stacked out-flow lenses interfingering with the surrounding hemipelagic sediments (sediment with a mixed oceanic (shells) and terrigenic (clays) source).

Fig. 3. Interpreted imagery of the RCMG video frame over the fluid venting site on the Mercator mud volcano.

Eos, Vol. 86, No. 49, 6 December 2005

Many ongoing geophysical and geochemi-cal studies are attempting to understand how volcanoes work and to forecast volca-nic eruptions. While many geophysical meth-odologies for studying volcanoes have been completely standardized, continuous geo-chemical monitoring in active volcanic areas is still in its infancy.

One reason geochemical approaches lag behind geophysical ones is because current technologies are not able to cope with pro-hibitive environmental conditions. The devel-opment of techniques that allow monitoring of geochemical parameters in the harsh environments of active volcanoes should be a goal to achieve within a decade.

One method that may allow continuous monitoring of volcanic gases is the use of a diode laser spectrometer (DLS), which can provide an accurate determination of the ratio of carbon isotopes in carbon dioxide (13CO2/

12CO2) that is present in these gases.

The study of the chemistry and isotopic composition of volcanic gases offers the possibility to follow over time the ‘life’ of vol-canoes, which can involve variations in gas composition that occur on both long and short timescales, often inexplicable oscilla-tions, and, more important, changes in source-region feeding systems.

Recent tests of continuous measurements of chemical species in harsh volcanic envi-ronments [Zimmer and Erzinger, 2003; Gianfrani et al., 2000, 2003; Rocco et al., 2004] respond to the above challenge. These new studies also help to meet the need for the development of new surveillance strate-gies that can be applied in active volcanic areas. The continuous measurements of chemical species, and quick and reliable field analysis of carbon isotopes seem to be among the most critical needs for the con-tinuous monitoring of isotope geochemistry.

This article demonstrates the field opera-tion of a DLS. During the period from July to October 2004, the authors performed three field tests at the Solfatara crater in the

Campi Flegrei caldera (north of Naples, Italy). Fluids from Bocca Grande, the hottest vent (150°C), were collected for laboratory tests and field isotopic measurements. Chemical analysis showed that water (H2O) and CO2 are the main components (~80 and ~20%, respectively) of the gas, and that spe-cies such as hydrogen sulfide (H2S), meth-ane (CH4), hydrogen gas (H2), argon (Ar), helium (He), and neon (Ne) are present in small amounts.

During the last 30 years, carbon isotope data from CO2 emissions at Bocca Grande have shown remarkable stability with an average δ13C value of approximately -1.6‰ [Tedesco and Scarsi, 1999; Tedesco, 1994]. This indicates that a mixing between mag-matic and sedimentary fluids is occurring probably at mantle level [Tedesco, 1994].

The near-infrared DLS is a simplified and more compact version of that reported else-where [Gagliardi et al., 2003]. Designed to work in a volcanic environment, it is housed in a 60 by 60 centimeter square breadboard, thermally-insulated, and is fully-controlled by a laptop computer (Figure 1). CO2 is con-veyed to the DLS after the fluid is cooled in a Pyrex trap and the H2O is separated from the whole gas phase.

Two methods of analysis were done: (1) collecting the dry volcanic gas phase into flasks (off-line) and (2) allowing gas to flow directly from the fumarolic vent to the optical cell (online). Optical cells are transparent containers (for the two-micron wavelength)

conditions [Sloan, 1998]. Outside the GHSZ, gas hydrates become unstable and dissoci-ate because of pressure decrease or temper-ature increase. The base of this zone is often recognized on seismic data as a BSR, because it mimics the seafloor in areas of constant local heat flow (Figure 2).

However, locally elevated heat flow will thin the GHSZ. As a consequence, the base of the GHSZ can intersect the seafloor, and gas hydrates might dissociate, forming seeps, which often is the case on mud volcanoes [Van Rensbergen et al., 2002].

A model of the gas hydrate stability field, using thermogenic gas compositions, sup-ports the interpretation of Depreiter et al. [2005]. These calculations revealed a focused local heat flow through the crater.

Gas hydrates on this part of the margin have only been reported from a small number of deepwater mud volcanoes [Gardner, 2001; Pin-heiro et al., 2003]. It has been suggested [Somoza et al., 2003] that some of the wide-spread shallow fluid venting sites within the Gulf of Cadiz could be explained by the desta-bilization of gas hydrate-rich sediments in con-tact with the Mediterranean Outflow Water [Somoza et al., 2003]. In many aspects, Merca-tor mud volcano deserves further attention.

Acknowledgments

The authors acknowledge the shipboard time provided on board the research vessel

Belgica by the Management Unit of the North Sea Mathematical Model. This research was conducted under projects GeNesis (FWO, Fonds Voor Wetenschap-pelijk Onderzoek/Research Foundation Flanders), Moundforce and MVSeis (ESF), GBCP (Intergovernmental Oceanographic Commission), EURODOM (EU FP5: Fifth Framework Programme), and HERMES (EU FP6 contract GOCE-CT-2005-511234-1). DVR is a post-doctoral fellow of FWO-Flanders. VH is a Marie-Curie post-doctoral fellow and honorary fellow of FWO-Flanders. DD and AF are Ph.D. fellows of IWT and FWO-Flanders, respectively.

References

Baraza, J., and G. Ercilla (1996), Gas-charged sedi-ments and large pockmark-like features on the Gulf of Cadiz slope (SW Spain), Mar. Pet. Geol., 13(2), 253–261.

Depreiter, D., J. Poort, P. Van Rensbergen, and J.-P. Hen-riet (2005), Geophysical evidence of gas hydrates in shallow submarine mud volcanoes on the Moroccan margin, J. Geophys. Res., 110, B10103, doi:10.1029/2005JB003622.

Gardner, J. M. (2001), Mud volcanoes revealed and sampled on the western Moroccan continental margin, Geophys. Res. Lett., 28(2), 339–342.

Pinheiro, L. M., et al. (2003), Mud volcanism in the Gulf of Cadiz: Results from the TTR-10 cruise, Mar. Geol., 195, 131–151, doi:10.1016/S0025-3227(02)00685-0.

Rodrigues, C. F., and M. R. Cunha (2005), Common chemosynthetic species in the Gulf of Cadiz: Updated spatial distribution, in Geosphere-Bio-sphere Coupling Processes: The TTR Interdisciplin-

ary Approach Towards Studies of the European and North African Margins, IOC Workshop Report, vol. 197, edited by N. Hamoumi et al., pp. 26–28, U.N. Educ., Sci., and Cult. Organ., Paris.

Sloan, E. (1998), Physical/chemical properties of gas hydrates and application to world margin stability and climate change, in Gas Hydrates: Relevance to World Margin Stability and Climatic Change, edited by J.-P. Henriet and J. Mienert, Geol. Soc. Spec. Publ., 137, 31–50.

Somoza, L., et al. (2003), Seabed morphology and hydrocarbon seepage in the Gulf of Cádiz mud volcano area: Acoustic imagery, multibeam and ultra-high resolution seismic data, Mar. Geol., 195, 153–176, doi:10.1016/S0025-3227(02)00686-2.

Van Rensbergen, P., M. De Batist, J. Klerkx, R. Hus, J. Poort, M. Vanneste, N. Granin, O. Khlystov, and P. Krinitsky (2002), Sublacustrine mud volcanoes and methane seeps caused by dissociation of gas hydrates in Lake Baikal, Geology, 30(7), 631–634.

Van Rensbergen, P., et al. (2005a), The El Arraiche mud volcano field at the Moroccan Atlantic slope, Gulf of Cadiz, Mar. Geol., 219, 1–17, doi:10.1016/j.margeo.2005.04.007.

Van Rensbergen, P., D. Depreiter, B. Pannemans and J.-P. Henriet (2005b), Seafloor expression of sedi-ment extrusion and intrusion at the El Arraiche mud volcano field, Gulf of Cadiz, J. Geophys. Res., 110, F02010, doi:10.1029/2004JF000165.

Author Information

David Van Rooij, Davy Depreiter, Ilham Bouime-tarhan, Eva De Boever, Koen De Rycker, Anneleen Foubert, Veerle Huvenne, Julie Réveillaud, Peter Staelens, Jeroen Vercruysse, Willem Versteeg, and Jean-Pierre Henriet, Renard Centre of Marine Geo-logy, Ghent University, Gent, Belgium; E-mail: David.VanRooij@ugent.be

newsMethod Allows for Continuous Monitoring of Volcanic Gases

PAGe 510