You are in site section: History & ideas

Stratigraphy Part 2 - Sediments

MIKE SMITH: Let’s begin with sediments, and sediments are the building blocks of our layers. Most layers are mixtures of different sorts of sediments. In a rock shelter, for instance, you may well have some material that is eroding from the wall of the rock shelter and other material that is blowing in from outside of the rock shelter to form a single layer.

Each type of sediment preserves clues to the source material and the mode of transport. The latter, the mode of transport, is particularly important because the velocity, the speed of wind and water, exerts a strong selection on the size grade of the material moved.

Energy governs the particle size basically. For example, in fluvial deposits, if you have high energy storm deposits, they can move quite large rocks, cobbles, boulders, et cetera. But the moment they hit a change in slope and the water begins to slow down, all of that material drops out and you are just left with the sand. Then as the water slows down further, the sand will drop out. And then as the water starts to pond, very slow-moving floodwaters carry the silts and the clays, and these are the last things that drop out. So the energy of floodwater, the entry of water flows, determines the size of the material it can carry.

Similarly with aeolian material, wind-blown material, for a given wind speed, coarse grain material - coarse sand and grit – is moved across the surface of the ground as a form of creep. It doesn’t move very far, it just moves along the surface of the ground. But fine sand is moved by a process called saltation and can be blown into dunes. Saltation is a process whereby a sand grain is bounced into the air, and as it lands it energises other grains and sets them out of bounds as well. Sand doesn’t stay in the air, it basically bounces across the landscape and accumulates in dunes. If you have vegetation, sand will be trapped by vegetation onto a soft sand movement.

Very fine material, the silts and clays, go into suspension. They go into wind-borne dust. Often it’s very hard to get this material out of the air unless you wash it out. Different grades of material are carried in different ways.

This diagram is from Mabbutt’s book on desert geomorphology. It’s a fairly classic diagram which shows the effect of transport, the inter-relationship between transport and particle size for aeolian sand. Fairly coarse sand is moved as creep; fine sand is moved by saltation as dunes; and silts and clays are moved in suspension as wind-borne dust.

Bearing in mind that transport is particularly important, what sorts of sediments might we find at an archaeological site? Wind-blown aeolian sand is the first. Usually it has a very narrow size range: 150 micron sand, very fine sand, often with an iron oxide skin around very rounded sand grains, red sands, orange sands. The redder they are, the thicker the iron oxide coat. Often, the older the sand is, as you move sand grains around, you knock off some of that iron oxide coating, so the sands that are very mobile are often quite light. There is very little clay or silt in aeolian sand. Often their deposits are fairly massive or a little bit cross-bedded.

You can get clay deposits, but wind-borne clay deposits such as loess are very rare in Australia. If you start to find clay in your sequence, I think you have to think of material that has either been borne by water or is being developed in situ through a cell-forming process.

One exception to this is that sometimes you can get sand-sized clay pellets around the edges of an old shore, an old lake, old Pleistocene lakes like Lake Mungo. The clay has cracked and dried and formed pellets. Sand-sized clay pellets will move just like sand. They were blown up into the clay dunes that we get at Lake Mungo like the Arumpo units.

Clay won’t blow like sand normally but, if it’s in these little pellets, it will. And at Puritjarra in the Pleistocene layer there was an unusual amount of silt and clay. This was coming in as coatings around the sand grains, coatings called cutans. These are typical of some soils. The coarse grains get coated in clay and then they’ll blow like a sand grain.

Fluvial sediments are usually without iron oxide coatings. They are often quite poorly sorted sands. They often have graded bedding. As water starts to slow down, different grades of particles will drop out. You will often have a grading from coarse to fine in the one layer. They will also grade along a layer laterally.

Imagine a depression where water ponds, the sand will drop out first, then the silt will drop out, and finally you get a little film of clay which might crack on the surface. It is quite characteristic of water laid segments.

Slope wash. We mustn’t forget slope wash. If you have a rock shelter on a slope and it’s open to input from material coming down the slope, you might get a lot of material just washing off the slope. This may not be very well sorted. It’ll be a promiscuous mixture of rocks, rubble, fine sediments and anything else that’s on the slope.

An extreme example of this, which we sometimes find in rock shelters in Papua New Guinea, is where an entire slope is mobilised and you get mass movement. So you get a whole plug of sediment that moves down the slope into your site. Sometimes this is the same with cave sediments. You get a plug of sediment moving along the cave system. Where you get a plug of sediments, they’re not going to be very well sorted, they’re going to be a promiscuous jumble of material, but it’ll also contain material that is reworked older material.

Fluvial sediments will sometimes excavate buried logs and old charcoal and move that. The mass movement of stuff on slopes will just bring down whatever’s on that slope into your site. You can get some real problems with your dating if you are dating reworked material.

Soils. You might have sediments that have been in place sufficiently long to form a soil where you’ve got a classic A horizon which is organic rich; you’ve got a B horizon that’s leached on the clays; and then you’ve got a C horizon where you’ve got enrichment of clays, carbonates and stuff like that.

Of course, some sediments are just going to be weathered bedrock. At the extreme end of this, you get structural failure in rock shelters, so you get a fall of boulders and large blocks of rock. Those failures are controlled by joint lines and weaknesses in the rock shelter.

In other cases, you’re going to find finer grade material in terms of spalls of rock that are popping off the rock shelter walls due to chemical weathering, wetting and drying, temperature changes. Often these will be little laminal pieces perhaps with a bit of patina or iron oxide staining on one surface. At the extreme end where you’re getting very active weathering of a very friable sandstone wall, you’ll just get clean sand weathering off the wall.

One of the things to really watch is that, often when you’re digging in a rock shelter and you get to the base of the deposit, you find yourself digging through what is effectively decayed bedrock. Maybe clean white sand with a bit of a crust. You dig through the crust and you keep going. Effectively you’re digging into the rock. When you draw your stratigraphic section, you shouldn’t be drawing that material as a layer of white sand. Sometimes when you’re digging towards the base of a site, you’ll find that you break through a little crust of sediment into, say, clean sand, maybe even clean white sand. In many of these cases, you’re actually digging into decayed bedrock. When you draw your stratigraphic section, you shouldn’t be drawing this as a layer of white sand so much as a layer of decayed bedrock. So the surface of that layer is actually the bedrock surface. That material hasn’t been moved and deposited in the site. It has weathered in situ.

Another form of sediment you’re going to find are anthropogenic sediments. They’re not so important on Australian sites as on historic European sites, et cetera where you have construction debris and stuff from digging trenches, pits, foundations, plow soils and stuff like that. But you are getting layers in Australian prehistoric sites where you’ve got a lot of ash and charcoal, animal bone, and stone debitage. So you’re getting an anthropogenic sediment effectively.

In most Australian sites we’re dealing with what the British would call ‘the natural’. We’re dealing with a natural sedimentary sequence into which our archaeological material is inserted in some way. Except for features, we have very few types of deposits that are primarily anthropogenic.

There is a few tips for how you recognise what sort of sediment you’ve got: its particle size; whether the grains are rounded or angular; how it’s laying in your site, whether it’s a jumble, whether it’s bedded or whether it’s graded in some sort of way.

What else can you do to decide what sort of sediment you’ve got? Well, look around, look at the context of the site, look at its geomorphic context.

What are the possibilities? You’re not going to get fluvial sediments in a site that’s on a totally flat plain and there is no possibility of run-off, wash or ponding. You’re not going to get sediments washing off a slope, you’re not going to get mass movement in some sites simply because of the location.

Have a look at your rock shelter. What is it open to? Where can the sediments come from? Is it a little rock shelter perched on a hillside or is it down at plain level? Can sediments wash in? Can they blow in? Or are they just going to be derived from the rock shelter itself and trapped in some sort of way?

One of the first things to do is to look around and try to assess what the possibilities are. Have a look at the wall of the rock shelter. How actively is it weathering? Can you see areas of fresh collapse? Can you see structural weaknesses?

Some rock shelters you go to that are at the base of an escarpment, and the whole edge of the escarpment is soft white sandstone. You brush it and it comes off. It’s a very friable area of active withering. You know that’s going to be a significant input into your site.

Have a look at the local soils. Your rock shelter might be at the base of an escarpment, but on top of the escarpment you might have highly mobile soils that are cascading over the cliff face into your site.

We’ve had examples in Australia like the classic site Devon Downs where, a long time after Norman Tindale dug that site in 1929, the farmer ploughed the land on the clifftop above, and a lot of those soils cascaded down and buried the old excavations. So if you want to get to the surface that Tindale looked at in 1929 it’s under a metre of sand just because of slope instability.

What else is happening in the landscape? Do you have active dunes out the front of the shelter? What’s the prevailing wind? Can stuff blow into your site? Look around, what processes can you see in other rock shelters in the area?

We often assume that sedimentation in rock shelters is switched on and off by human use of that rock shelter. Look at what’s happening in other rock shelters nearby where you have material either accumulating or not accumulating.

These are some of the things I’ve found useful to look for which give me a bit of a clue as to the source of sediments. I’ve already mentioned clay pellets or cutans. When you find clay in a site, you should ask yourself whether you have fluvial deposits but then have a look at the structure of it. Is it blowing in as sand-size particles, which are very distinctive of these lake shore environments or palaeosols?

You may find you have a fine, red sand. In the interior of Australia red sand is quartz but, on the West Australian coast, the dune sands, which look just like any quartz sand you’ll find in the Simpson Desert, are actually made of sand-sized carbonate. It’s all very fine shell material that has been eroded and blown off the continental shelf. But because it’s carbonate, it’s highly alkaline. If you put a bit of acid on it, it dissolves almost entirely. But it forms carbonate coatings on artefacts and on shovels. It case hardens very quickly to form a bit of a carbonate pan.

When you see a lot of carbonate in a site, this white lime concretion, ask yourself where is it coming from: Is it coming out of the rock? Is it because of carbonate sands blown off the continental shelf?

Similarly, desert dunes in the interior are quartz dunes and normally are very slightly acidic. They are either neutral in terms of pH or slightly acidic. But at Puritjarra I found the sand in the site was a pH of 3.5, which is very acidic. Why was that? Because the sediments that were feeding the site were not only coming from local dunes, they were blowing off the edges of saltlakes where you have gypsum, which is calcium sulphate. So you’re getting sulphates in the deposit which are increasing the acidity.

There are a lot of little tricks to looking at sites, looking at sediments and trying to work out what sort of sediment you’ve got and where has it come from. One of the keys is to look at the context of your site and how it sits in the landscape.

Sites are not stable monuments, and I can just give you a few examples. You might imagine a site where you’ve got some nice layers and you’ve got a few pits and fireplaces. If you’re getting a bit of invertebrate activity, even low levels of invertebrate activity, over time a lot of the structure, a lot of the detail in your site, is going to fade. It is just going to be bleached. Like an old photograph bleaches and fades over time, you lose the structure in your site.

Sites also represent some sort of position, a point in time which is a balance in the sediment budget. If you think about a rock shelter, it’s a void. It’s a void because not only has a sediment been eroded, it has been removed. When sediments accumulate, it is because something has happened. A rock has fallen at the entrance to the site and sediment can accumulate. Maybe that rock is then removed and the sediment can be evacuated out.

Sites are living things. They’re constantly in flux. We think they’re stable because they’re changing at rates that are well beyond our own lifetimes. But they can change extraordinarily rapidly if conditions change.

Sooner or later at the Puritjarra rock shelter, which is a magnificent, large rock overhang, sooner or later that rock overhang is going to collapse, and then the whole set of processes that are operating on that deposit are going to shift a bit.