The seabed on Earth is concealed from our view by the water, which prevents us seeing more than a few tens of metres in the best of conditions. We actually know less about the detailed textures and morphology of the ocean floor than we know about the surface of Mars. Some very strange and unfamiliar processes occur down there and we need to understand them before we can properly interpret processes on Earth and on other planets, like Mars.

Turbidity currents are produced on Earth when material collapses or is swept off the continental shelf and  pours down the continental slope towards the abyssal plains, several km deeper. As the material descends, it turbulently mixes with water and forms a thick "Turbid" cloud of water and mud, with sand and perhaps clods of sediment and even gravel and boulders mixed in with it. The density of this mixture is greater than surrounding seawater so the cloud hugs the bottom and descends the slope. As it accelerates downslope and gains speed, it is strongly erosive and slope channels result, remarkably similar to the outburst flood channels on Mars.

Turbidite flows are often triggered by Earthquakes, and they were first discovered after the 1929 Grand Banks earthquake, when density flows triggered by the 'quake flowed downslope at over 80 kilometres per hour and succesively broke a series of telephone cables laid on the seabed.

On the abyssal plains, the flows spread out and may form meandering or distributory channels, sometimes like those in subaerial rivers. The material from the flow is spread in a wide smooth fan, butying any prior topography. Often, adjacent fans coalesce to give the distinctive smooth, low relief abyssal plains that characterise the ocean floors.

Warning: This page is linked to many large images which may tax your browser or your internet connection. Do please try to view them, they will help you to understand the analogues to Martian density flows .

Schematic diagram of a turbidite channel and lobe system

Erosion patterns on the South-Eastern USA Continental slope. Despite the term "slope", gradients are typically <1 degree.Note the headwall ravines, anastomosing channels, and teardrop islands - just like outburst "flood" channels on Mars. The image is ~170 km N-S and 110 km E-W

Context image for the above. Note that the entire continental slope is channelised and that in some places extensive meandering channels are visible where major rivers feed sediment down canyons to the seabed.

Turbidite fan complex off the Oregon Coast.
The Fan complex, in the foreground of the image, emerges from a zone of seabed thrust folds (the ridges). Several criss-crossing channels are revealed, each with raised banks - levees, where deposition occurs when the density flows spill over the banks.

The following images are from Sandwell & Smith's predicted seabed bathymetry dataset. This is the best available image of the oceans. It is calculated from satellite observations of sea-surface deflections due to the gravity effect of the rocks at the seabed. The "dimpled" appearance is due to noise in the dataset.

Another view of submarine channels offshore Western USA: Note many different scales and geometries of channel, including one a little below centre of the image that bifurcates around a circular feature, like the channels on Mars going round a crater! This channel complex is over 1000 km in length, yet descends  only 2000 m along its length, a gradient of 1:500. (note that for a patch of the lower section of the image, real bathymetry is available, with less noise, hence the smooth appearance).

Submarine channel in the centre of Baffin Bay: The main channel is nearly 1000 km long and is joined by several tributaries. The main channels descends ~1000m in 500 km.

Submarine channels off New Zealand: A variety of channels, large and small. The largest is over 700 km long.

      Created: May 2002
      Last modified: May 2002
      Authorised by:  Head, Earth Sciences

      Maintained by: Nick Hoffman
      Email: nhoffman@unimelb.edu.au