Good morning.

Deformation mechanisms and rheology of evaporites play an important role in the evolution of the CEB. In this talk I will give you an overview of rheological properties of Halite in different tectonic settings, and show you how one can get information on this from natural Halite samples.



If one wants to quantify the rheology of halite deforming in nature, we have to know how to extrapolate the data from short-term, high-stress laboratory experiments to nature? The basis for such extrapolation is the correspondance between deformation mechanisms in the laboratory and in nature. But there is another question: are the rheology of the salt and the deformation mechanisms in every tectonic regime the same?

In this project we did a number of case studies of naturally deformed halite and compiled a database of the rheological data. In this talk I will try to summarize these, and try to define a consistent set of rheologies in Halite which can be used in future numerical models. As you will see there is a lot more variability in the rheology of salt in nature than generally believed and this variability has a major influence on the style of deformation.  At this point, none of the existing geomechanical models of salt tectonics (show) use the realistic rheologies. One of the aims of a project which has been proposed for the next phase of this SPP by Tanner and others, is to incorporate these results in Finite Element-based reconstructions of salt tectonics.



The deformation mechanisms of Halite, under natural and experimental conditions, can be cataclasis, dislocation-creep and solution-precipitation creep. A characteristic microstructure for dislocation-creep process – see on the right hand side – is the presence of subgrains. Dynamic recrystallization is commonly associated with this deformation mechanism. During pressure solution the highly stressed part of a grain goes into solution and precipitates on the less stressed part, forming very different microstructures. The flow is governed by the well known power-law, which has a form like this. In the equation of dislocation creep s1-s3 is the differential stress, A and Q are constant and the n is the power, and typically has a value between 3 and 7 for halite. In the case of solution-precipitation creep, the grain size comes into the equation, while the n – the power - here equals with 1. The total strain rate is the sum of the contributions of the two different mechanisms.



So how does this rheology look like? Lets have a look at the data from triaxial deformation experiments. Most experiments were done on natural coarse grained samples. In these experiments, at laboratory strain rates the main deformation mechanism is dislocation creep. The differential stress versus strain rate plot of these experiments is shown here. The different colours represent different temperature, and the slope of the trend in the data corresponds to the stress exponent n.

Only a few studies were made on fine grained, wet salts, where the main deformation mechanism is solution-precipitation creep. The blue squares here represent these experiments. You can see that at the same temperature this Halite is much weaker than the coarse grained salt. The slope of the data shows that n is about 1 in solution-precipitation creep.



Now let’s see how relevant are these experiments for natural conditions at low temperature. This brownish-yellow square represents the natural conditions both for differential stress and strain rate. You can see that no experimental data are available, and we need to extrapolate the laboratory measurements. The dislocation creep experiments have to be extrapolated to much lower stresses, while the pressure solution has to be extrapolated to larger grain sizes. An important implication of this slide is that in nature solution precipitation creep can contribute much more to the total strain rate than in laboratory experiments on natural samples. It is also clear that in the laboratory,  there is a relatively large – factor of 10 or sometimes 100 – variation in the strain rate  of natural samples. This means that in nature salt is much more heterogeneous than assumed in most models to date.



Now we will present three case studies from three different salt structures in order to see what deformation mechanisms are dominant and what kind of rheology is appropriate for these salts. We will look at: one weekly deformed bedded salt, some domal salts and also a glacier salt from Iran, representing the salt extrusions which were common in the CEB in Jurasssic times.



This is a micrograph of a gamma-irradiated thin-section. It contains many primary structures - these white areas – which are full of primary brine inclusions. These are overprinted by deformation features, as shown by these subgrains, and minor recrystallization, by grain boundary migration. Recrystallization erases the primary structures in the salt and by this process fluid inclusions, moved into the grain boundaries, so that, the solution-precipitation creep is accelerated by this process.



The subgrain size gives us stresses between 0.6 and 1 MPa for the bedded salt, and the depth of the salt layers imply maximum 50 C for deformation temperature. Taken these together, we see that both solution-precipitation creep and the dislocation creep contribute in the same order of magnitude to the total strain rate and the equation describing rheology should be composed of the contribution of both.




We studied domal salts from the asse, gorleben and klodawa salt diapirs. The microstructure shows clear evidence for dislocation creep – see the subgrains in this image - and fluid assisted grain boundary migration – see the beautiful growth banding in this grain




The subgrain size in the domal salts imply stresses of 1-2 MPa. The deformation temperature, even for deep salt diapirs, is not more than 150 C. The main deformation mechanism according to the microstructure images and to this graph is dislocation creep, while the solution precipitation creep contributes less to the total strain rate due to the relatively large grain size, and high temperature.




Our last example is the exotic, highly deformed glacier salt form Iran. Here the Cambrian Hormuz salt comes from a depth of 5 km and reaches the surface. From the topographic high, that is the outcrop of the salt dome, the salt mass moves slowly down under very low stresses. This is the salt glacier. The movement of the salt glacier is accelerated after rainy seasons.

Such salt glaciers have been shown to have been present in the CEB during the geological past – we saw in the previous talk from Mohr - so these samples are quite relevant to salt tectonics in Germany.



The microstructure is strikingly different from the previous samples. Lots of very small grains are present and only a few large grains. Those large grains are full of very small subgrains, while the small grains are free of subgrains. The presence of subgrains in the large grains means that those deformed by dislocation climb process, but the lack of subgrains in the small grains is indicative for pressure-solution.

How can we explain the structures in this salt? We have to explain somehow the presence of subgrains, and the deviatoric stress of 4-5 MPa that they imply. That high stress is clearly not present in the salt glacier. The most reasonable explanation is that they are introduced during the upward transport of the salt. In the glacier itself, the pressure solution is the main deformation mechanism but the old grains preserve the record of previous high stresses. The fluid which is necessary for this type of deformation comes from the seasonal rain.




In this case, we have to separate two parts of the process. One is with high stresses -3 to 5 MPa - recorded in the large grains and the other related to the low stress fine grained samples. If we plot the two different sets on the graph we see that while in the first case – the extrusion phase - the rheological behaviour is controlled by the dislocation creep, in the second case - in the salt glacier - the main deformation mechanism is pressure solution.




With the case studies we saw that in different salt structures different deformation mechanisms dominate, thus their rheological behaviour should be described with different flow laws. With other words, if one wants to build a proper geomechanical model, the model, as the structure matures, has to switch between flow laws!

Variations in temperature and switch between the flow laws yield variation in viscosity values between 1014 to 1019.




In this talk we demonstrated that in different tectonic regimes different deformation mechanisms are dominant.

These are associated with different levels of differential stress and different rheologies, and these different rheologies should be used for mechanical modelling of salt structure development.

Thank you for your kind attention.


7th oct. 2004 sofie