This image shows a number of curious factors. On the right hand side of the head of the main deposition apron is a subsidiary one, which ends in a regular set of barchan dunes, showing flow running in the same direction as the line of flow of material upstream, and of about the same width.

Water-lubricated material flow will not produce such a feature, but airflow will. As an airflow lost velocity and carrying capacity where the crater starts to level out, fines could very easily have remained in the train and been rapidly deposited as a dune set.

An airflow clearly had the carrying capacity to move the material in the dunes and sculpt them, on the margin of whatever event produced the apron. It is a short step to assume that in the centre of the event it may have had the energy to move the rest of the material present. It is a rather longer step to suggest that material was moved by a chemical not known to be present or stable on the planet in the phase required.

Next, at the end of the apron, the material is deposited in narrow elongated lobes, not as the broad lobes characteristic of deltaic water deposition on Earth. Lobe narrowing does occur in water streams here, but generally only when banks of deposited material are either topographically constrained, or trimmed by faster flowing water moving past. There is no evidence for flow going any distance past these terminal lobes in the East Gorgonum crater.

Lastly, though the resolution is not adequate to be sure, the entire surface of the apron appears to be irregular and pitted. This is not what is generally seen in alluvial fans or deltas. Whether water drains away rapidly or slowly, it usually leaves a flat upper surface to the deposit, sometimes ripple-marked on a very small scale.

If the flow that produced this apron was gas-driven, rapid and ill-sorted, once momentum was lost, de-gassing would perhaps produce the pitting that appears to be present. If the flow included substantial quantities of permafrost dry ice, pitted would be accentuated when this eventually sublimated.

The barchan dune sets seen running along the line of lowest topography in channel after channel are definitely a major hurdle for the flowing-water hypothesis. They are definitely very young. If they were old and the wind carried no sediment load now, they would have blown away ages ago.

That they are there proves beyond any doubt that recent sediment-laden airflows of considerable strength and carrying capacity have moved down those channels.

As these flows were clearly dry, and capable of moving that much material, they would equally have been able to move material similarly both closer to the erosional source and further down channel as well.

Similarly, any airflow capable of generating those barchans would be capable of and could not avoid eroding and depositing in the horizontal plane. In other words they would produce lovely meanders, as seen in valley after valley, if the banks were unconsolidated dust piles. By the bye, perhaps the height, curvature and wavelength of the barchans would give some indication of airflow velocities and volumes.

To give another specific example:

Jpeg gorgonum 2 c 100, of MGS MOC Release No. MOC2-236, 22 June 2000, (A, gullies close up, enlarged, above), shows a beautiful set of bakery-fresh barchan dunes all the way down the centre of the gully, one after the other, blocking the whole bottom of the channel and clearly created by the last flow down it.

No canoeist has ever ridden standing waves created by such, no diver has ever seen one underwater. As such I contend that this series was formed by a strong airflow with a regular wavelength. No liquid water stream on this planet ever created one of those. Famous last words in geology, but I'll bet my shirt on it anyway.

There is another barchan set further down the same channel, and the first moves apparently up and definitely around the older one to get past, leaving the original intact.

No liquid water flow ever did that either - flowing water would simply have ripped the old barchan line to pieces. Relatively dense air with its S.G. much raised by entrained dust and dry ice particles would do precisely that, though.

The location shot presented with this image by NASA suggests this terrain is a tableland controlled by tension cracks. There is zero sign of any developed integrated drainage pattern. If the canyon was formed over a long period of time by running water, there should be one.