Submarine canyons

Background | Key results | Canyons image library


Deep-sea submarine canyons present one of the most formidable challenges to marine scientists today. Canyons are deep incisions of the continental shelf and slope, and they dissect much of the European ocean margin. Were these canyons on land, they would present some of the most dramatic mountain scenery in the world. Hidden by the ocean, and covered in a drape of sediment, they have been largely ignored until now because of the difficulties in exploring their complex terrain. Advances in technology, notably Remotely Operated Vehicles (ROVs), swath mapping systems and deep-tow sidescan sonar, are opening up these frontiers, giving new insights into canyon life and processes.

Right: Complex submarine canyon systems in the Gulf of Lions, western Mediterranean.
Image courtesy IFREMER

Some canyons are closely connected to major river outflow systems while others funnel large quantities of sediment from the continental shelf into deep water. Canyons act as temporary depots for sediment and carbon storage. However, rapid, episodic flushing of canyons may mobilise large amounts of sediment carrying it to the abyss and overwhelming benthic ecosystems over a wide area. The frequency of these potentially catastrophic events and the fluxes of particles produced are largely unknown, as are the rates of recolonisation and restoration of the canyon ecosystems.

Canyons are complex systems in terms of their hydrography, sedimentology, biogeochemistry and biology. As more is learnt about canyons, it becomes increasingly obvious that there is great variability both within individual canyon systems and between different canyons. Individual canyons have very different environmental characteristics that determine the diversity and the ecology of their fauna. This makes it difficult to reach generalisations that will be useful in creating policies for whole ecosystem management, without a concerted effort to compare canyons from different biogeochemical provinces and different topographic settings, and coordinated, multidisciplinary projects relating the fauna to the environmental variables that regulate their distributions.

To see what HERMES scientists have discovered in canyons, see our canyons photo atlas.

Left: Bathymetric image of the Cap de Creus Canyon, W. Mediterranean.
Image courtesy of AOA Geophysics, Fugro and University of Barcelona


Key results

HERMES has examined canyons in the Eastern Mediterranean, off southern Crete, in the NW Mediterranean from the Var Canyon southwest to the Cap de Creus Canyon, along the western Portuguese margin in the Eastern Atlantic and lastly in the NE Atlantic off southern Ireland. At each of the main geographic regions we have attempted to determine the detailed morphology of the canyon systems, their tectonic history, the physical oceanography, particularly in relation to atmospheric forcing, and the sedimentary regime within the canyon. This physical background was used as a template for understanding the distribution of different-sized organisms within the canyon. Lastly, we examined what are the actual and potential anthropogenic impacts on selected canyons.

Right: This 3D bathymetric view of the Nazare canyon on the Portuguese margin shows its steep sides and narrow profile, which make exploration of this canyon a particular challenge. Image courtesy Instituto Hidrografico.

The morphology of a canyon is unique and is a function of its tectonic history and subsequent sedimentation. With the availability of modern mapping tools such as ship-borne swath bathymetry we have been able to produce high-resolution images of the submarine canyons in the four HERMES areas. These bathymetric maps show that in the NE Atlantic and Portuguese canyons, upper sections are deep incisions in the continental shelf, with steep rocky sides giving way to scarps, terraces and gullied walls in the middle canyon and broadening out to wider terraces in the lower parts of canyons, the lower terraces incised by a central channel (or channels) called the thalweg. In the NW Mediterranean the original angular morphology of the canyon has been smoothed over by subsequent sedimentation. The canyon morphology, particularly at smaller scales, is much more complex than really appreciated; this heterogeneity was emphasised by the use of submersibles capable looking at metre and sub-metre scales.

Physical processes as a function of atmospheric forcing are dominant factors in the canyons, best exemplified by a comparison of the Cap de Creus Canyon in the NW Mediterranean and the Nazaré Canyon along the Portuguese margin. In the winter months (although not in all years) an exceptionally cold Mistral wind blows down the Rhone Valley, causing Dense Shelf Water Cascading (as described on the open slopes pages). This cold wind cools the surface waters in the Gulf of Lions and in certain years under intense cooling conditions results in the cold water sinking down the Cap de Creus canyons by a process called cascading. This cascading resuspends sediment and carries it into the deep NW Mediterranean. This cascading has an important effect on the deep-water prawn fisheries of the NW Mediterranean by rejuvenating the population on a quasi-regular basis.

Left: Schematic showing the passage of DSWC in the Cap de Creus Canyon.
Image courtesy UB/CEFREM.

Along the Portuguese margin the intensity of flow of water down the Nazaré Canyon is determined by local storm conditions: under calm conditions, suspended sediment load in the waters of the canyon is low but following a period of storms there is a temperature change and an marked increase in the suspended sediment load in the water column. This flow transports both inorganic and organic particles into deeper water, bypassing the upper part of the canyon and depositing sediment as the energy level drops in the middle part of the canyons. Estimates of 1cm y-1 deposition were calculated for the middle Nazaré Canyon, with 0.1mm y-1 being deposited in the lower canyon. This suggests that the Nazaré Canyon acts as a sediment sink rather than a transport conduit. The middle part of the canyon is also a significant sink for organic matter, although the source is variable being phytodetritus in surface sediments and terruginous in the underlying sediments. Climate change, in terms of global warming, may well have negative impacts on both cascading, by increasing surface temperature and stopping cascading, and increased storms in the Atlantic increasing down-canyon supply of sediment to the deeper parts of the Nazaré Canyon. Evidence from cores in the Nazaré Canyon suggest sedimentation patterns over the last 1000 years are related to climate change - particularly changes in rainfall and thus fluvial input.

The physical condition of the canyon determines the distribution of individual species within the canyon and thus canyon biodiversity. After depth, substratum type is an important determinant of species distribution. Dominating our understanding of biodiversity in canyons are those species that inhabit sediments. In the eastern Mediterranean, off Crete biodiversity in prokaryotes is very high. The smallest sedimentary fauna are the meiofauna, which includes the nematodes and harpacticoid copepods. This size class has been examined in detail in all four HERMES areas. In the Samaria Canyon off southern Crete, nematode diversity was unexpectedly high but in the other southern Crete canyons lower than the adjacent slope. High meiofaunal diversity was also found in the canyons of the Portuguese margin and in the Whittard Canyon in the NE Atlantic.

Right: Crabs at play in the Var Canyon, western Mediterranean.
Image courtesy Ifremer/MEDECO 2007

Detailed macrofaunal analysis is available for the Nazaré, Cascais and Setubal canyons in the Atlantic and the Var Canyon in the NW Mediterranean. In the Nazaré Canyon there was a wide variety of taxa particularly of opportunistic surface and sub-surface feeders in organically- enriched areas. A detailed analysis of the polychaete fauna showed it was very diverse with many polychaete families represented, but decreasing in diversity with depth. In the Var Canyon biodiversity was relatively high amongst the macrofauna. The most significant variations in the Var Canyon were found between the upper and lower sites in relation to the contrasted hydrodynamic and particulate input regimes in both areas.

At the level of megafauna (those species visible in photographs) the type of seabed has an important influence on the species present. In sedimentary areas in the canyon the megafauna is dominated by deposit feeders such as holothurians, including infaunal species, but there are also enigmatic species such as the giant single-celled protozoans, the xenophyophores and gromids. However, if the substratum is rock, particularly in the upper part of the canyon, organisms that need to attach to rock are usually present including corals, gorgonians, anemones and brachiopods. In addition, the actual aspect of the rock face has an impact. In places where there are small overhangs a distinct filter-feeding fauna such as crinoids, bivalves and brachiopods dominate over this often very small spatial area. Preliminary observations suggest a strong zonation and some high densities of megafauna in the Whittard Canyon, the local diversity being mainly driven by sediment type. In the canyons of the NW Mediterranean the megafauna is dominated by scleractinian hard corals, particularly Madrepora oculata, with their distribution affected by the local hydrodynamics. Detailed analyses of the diverse fauna associated with these coral species have been completed. The fish fauna of the Cap de Creus and Nazaré canyons has been described, with evidence that fish diversity decreases with depth in the Nazaré Canyon.

Left, top: Multi-armed basket star in the Whittard Canyon, offshore Ireland. Bottom: Brisingid seastar in the Lisbon Canyon, offshore Portugal. Both images courtesy NOCS/JC10.

Anthopogenic contamination was also found in the canyons studied during HERMES. High concentrations of Pb and Zn were found in the Lisbon and Cascais canyons, particularly at shallower depths. Conversely, copper shows the highest enrichment values at in the lower canyon, most likely related to phytodetrital sedimentation. Persistent organic pollutants such as PCBs and DDEs were found in the canyons off Lisbon and in the Cap de Creus canyon of the NW Mediterranean. In the latter case the DSWC events appear to ‘cleanse’ the shelf of pollutants and deposit them in the deep sea. Litter of various types is found in a number of submarine canyons including plastic bags, plastic bottles and abandoned fishing gear from both trawling and long line. However, those canyons relatively remote from industrial activity such as Nazaré, Whittard and the canyons of southern Crete appear to be little affected by anthropogenic impact.

Above: Litter in the Lisbon Canyon, offshore Portugal. Image courtesy NOCS/JC10.

HERMES canyons research shows that biodiversity of canyons can vary on a variety of scales from the sub-metre up to 100s of kilometres as among the different sites studied during HERMES. The significant feature is that at one point biodiversity may not appear very high, but in canyons the heterogeneity of the habitat causes variation in the fauna present so that the biodiversity of an individual canyon as a whole is high and when all the canyons along the European margin are combined biodiversity within canyons is seen to be significant.

Above, from left: Waiting for recovery of the amphipod trap aboard RRS Discovery, Nazare/Setubal canyon, summer 2005 (image courtesy J. Ingels, University of Ghent); recovery of box corer RRS Discovery, Nazare/Setubal canyon, summer 2005; Galatheid crab and hexactinellid sponges in the Setubal Canyon; scientists getting their hands dirty after sampling in the Nazare canyon (images courtesy NOCS); the deep-water coral Lophelia pertusa, accidentally sampled during a video survey of the upper canyon wall in the Setubal canyon (image courtesy NIOZ); hungry abyssal grenadiers and an arrowtooth eel swarm around bait on the ROBIO lander in the Nazare Canyon (Image courtesy Oceanlab University of Aberdeen).

Links to HERMES canyons expeditions:

© HERMES 2009