Blown away by the eons of time

By: Chief Scientist, Bramley Murton.

The week-end has been dogged by lumpy weather. A nasty storm cyclone has hit us and we have had to run away, 200 miles to the North. The storm is unusual for this time of year, and is exceptionally deep. It has tracked from the west right across the Rio Grande Rise, lashing the ship and stopping science. Luckily we recovered all our gear before it hit.

A nasty wee storm finds us and sends us packing for the weekend.

The Autosub6000 being recovered after its second mission over the Rio Grande Rise.

Just after recovery, within a couple of hours, the sea picked up and we decided to postpone the next HyBIS dive. This was a pity as we have made some interesting discoveries on the last dive. However, without the heave compensation working on our main winch, we can’t use HyBIS in heavy seas as it is too hard to control. The break will give the HyBIS team time to make some minor adjustments to the motors and lights to optimize the next dive after the storm passes.

The sea get a wee bit lumpy shortly after recovering the Autosub6000.

The last HyBIS dive was made over the southwestern side of the Great Rift. Here we started at the top edge of the scarp and made our way slowly down a few terraces. The sight was amazing: ferromanganse crusts were there, as we predicted, but they have a polygonal pattern on the surface, resembling a large honeycomb. Closer inspection shows the crusts are being eroded by patches of sand in the centre of the polygons leaving them to resemble cracked dried mud.

Polygonal pattern to ferromanganese crusts covering lavas on the edge of the Great Rift.

Further down the slope we found lavas exposed. These were also polytonally jointed. The jointing on the lava surface turns into vertical joints (cracks) underneath, forming columns. This type of structure is well known in lava flows from land – a classic example is the Giant’s Causeway in County Antrim, Ireland. The difference here is that the lavas we are looking at are 900m below sea-level, yet they look like subaerial eruptions. In contrast, submarine lavas form distinctive pillow shapes. This is because the water quenches the lava and forms a skin, like an amniotic sack, that stretches as the lava fills it until it forms a bulbous shape. Where we get sheet flows on the seafloor, they rarely, if ever, form columnar jointing. Is this evidence that we are looking at eruptions on the Rio Grande Rise when it was an island?

Columnar-jointed lavas on the edge of the Great Rift, Rio Grande Rise.

The vertical face of the scarp exposing the lava flow. The scarps is 30m high and forms the edge of the Great Rift.

Red clay and mud underneath the columnar jointed lava, and on top of another lava flow, 900 metres below sea level.

Our suspicions were confirmed at the bottom of the terrace where the columnar-jointed lava gave way to an underlying bed of red clay. The clay deposit merged into the top of another lava flow beneath the first. Geologists recognize this mud as a bole, the top of a lava flow that has been weathered by sub-tropical and humid conditions, oxidizing the lava and turning its iron content red, and turning the rock into clay minerals. This is in effect an ancient soil, or as we call it, a palaeosoil. Millions of years ago, volcanoes erupted lavas on the Rio Grande Rise when it was an island. Sun and rain eroded the top and transformed it into soil. Plants probably grew in the sunlight and who knows what animals grazed on them. Then, after thousands of years, the volcanoes erupted again and buried the landscape in another flow of white-hot lava and everything was incinerated. As it cooled, the lava contracted forming the columnar jointing we see today. This was probably followed by further cycles of weathering, soil formation, plants growing, animals grazing, lava eruptions and incineration. We are just seeing a snap shot in the eons of time that have formed the Rio Grande Rise when it was the equivalent of a sub-tropical Iceland, and long before it was drowned and submerged beneath the South Atlantic.

Back up on the top of the scarp we see the lavas give way to a sediment-covered flat seafloor strewn with large rounded boulders. These are made of the black lava, but are smooth and round. Only in areas where there is very high energy can such boulders form. Areas such as the beds of fast flowing rivers, or on the sea shore where waves can crash and tear at the cliffs, dislodge chunks of rock and roll them around until they form smooth and round boulders. Certainly not at the bottom of the ocean, 700m below sea level, where such waves can’t affect the seafloor. Here the energy is low, and currents are slow, moving only the smallest of sand grains. Here, then, we are seeing evidence for the drowning of the Rio Grande Rise, when the sub aerial lavas became sea cliffs and the waves churned the rocks into a boulder strewn shore.

Boulders strewn over the seafloor close to the edge of the Great Rift are evidence that this was once a sea shore where waves crashed and tore at the cliffs, long ago as the Rio Grande Rise was drowned.

Further back from the edge of the Great Rift, the boulders give way to a hard seafloor covered in calcareous sands. These sands are derived from the skeletal remains of plankton in the overlying ocean. The sediments are thin, only a few tens of centimeters, and the seafloor underneath is hard when HyBIS lands to take a scrape. The underlying seabed is also calcareous, made of crushed carbonate shells forming a sandstone. Was this once a sandy lagoon behind the cliffs and boulder strewn seashore? Almost a kilometer back from the edge of the Great Rift we come across an increasing number of chunks of black ferromanganese crust lying on the white pelagic sediment.  Where have they come from? The answer lies another 100m west, where we come across a ledge that cuts across the seascape.

Terrace with ferromanganese crust on the top overlying a calcareous sandstone.

Only 2 metres tall, the ledge is capped by 20cm thick layer of black crust. Underneath the crusts is the hard calcareous sandstone seafloor, here eroded and exposing the layers of crushed shelly sand. The edge of the ledge is undercut and pieces, some a metres across, of the crust have fallen off or are overhanging the ledge. HyBIS tries to take a sample, but it is too big. Eventually we get a couple of pieces that fill the basket. Another beach terrace? Perhaps, but this one has been cemented and covered by the ferromanganese crust that we are searching for. We know that the crusts grow at about 1 to 3 mm per million years, so again, we are confronted by a scene that has been frozen in time for tens of millions of years; a fossil shore line from the distant past, now drowned and lost beneath the stormy South Atlantic ocean.