Stratification Experiments and Flood Geology

by Wayne Spencer

The following is a summary regarding experimental research done by creationist Guy Berthault from France and a colleague, Pierre Julien, an Associate Professor of Civil Engineering at Colorado State University in Fort Collins, CO. The original research related to this work was done in 1989 and 1990. Technical papers have been published related to this work by Guy Berthault, Pierre Julien, and Yongqiang Lan in Geological publications in France and in Civil Engineering Report by Colorado State University. Guy Berthault has presented this information and its significance to the Noahic Flood at the International Conferences on Creationism, in 1994 and 1998. Berthault has also produced popular level video programs explaining the concepts, first in a program called "Evolution: Fact or Belief". Recently in 1999, a new short video has been produced called "Experiments in Stratification". This program is a revision of an earlier program with the same name. This short program is a well produced presentation of the essence of the experiments and Berthault's views on the significance of these experiments for Geology and specifically the discipline of Geology known as stratigraphy. While the validity of the experiments has not been questioned seriously, the question of their significance has been somewhat controversial among leading creationists. The following is my attempt to comment on what Berthault's experiments show and what their significance may be for a creationist view of geology.

Guy Berthault and Colorado State University have been working on some hydraulic sedimentation experiments that may have important implications for stratigraphy. The experiments show how sediment deposits in water when there is a current. This work shows one mechanism for how a series of rock strata consisting of multiple horizontal layers can be laid down by water in a short time. Because of the unique way that particles are deposited by this mechanism, multiple layers can form simultaneously, being added to horizontally rather than forming in a vertical sequence as is sometimes taught and assumed in evolutionary geology. This is a mechanism that could apply in a number of locations in the real world, but it is not a universal mechanism. I will try to clarify the scope and limits of this work, to the best of my ability. Creationist geologists, often referred to as Flood Geologists, have identified a number of mechanisms whereby sediment can be rapidly deposited or eroded under catastrophic conditions that would exist during and after a true global Flood. Mechanisms such as this do not prove that the Biblical Flood of Noah occurred, but they show plausible known mechanisms that may have operated during or after the Flood that could have had geological effects we could see today. Catastrophic mechanisms can rapidly accomplish effects that have sometimes have been interpreted as requiring long periods of time.

In general, creationist geologists would not disagree with many methods of field geology. More at issue between creationists and evolutionists is the way geology is taught and the assumptions of historical geology, which addresses the origin of geological features on Earth. It is possible evolutionary bias could cause field geologists to neglect certain alternatives that would fit a catastrophic Flood interpretation, just because they have not been trained to look at the facts another way. The following research does call into question some assumptions related to the geologic column, at least the way it is often taught in introductory secondary and post-secondary geology courses. We must guard against being overly "married" to traditional accepted ideas, and against overly applying a new idea. The issue revolves around how valid the following two accepted principles or assumptions of stratigraphy are.

1) The Principle of Superposition - In a sequence of strata consisting of several layers, the lower layers were deposited first (in horizontal layers) and moving up, each layer is younger than the layer beneath it.

2) The Principle of Continuity - That within each layer that is of the same type of material and which makes a contiguous layer, all the material within each layer is of the same age. (This does not mean precisely the same age. The age spanned within the thickness of the layer is assumed to be much less than the age spanned across the layers of the whole sequence.) I will try to show this below with dots and "o" characters representing different sizes of particles in a sequence of strata. t1, t2, t3, and t4 represent a time sequence. The dots and o's represent particles of varying size in a sequence of sorted layers.

:::::::::::::::::::::::::::::::::::::::::::::::                        t5      youngest

OOOOOOOOOOOOOOOOOOO              t4 (largest particles)

ooooooooooooooooooooooooooo          t3 (larger particles)

::::::::::::::::::::::::::::::::::::::::::::::::                       t2 (small particles)

OOOOOOOOOOOOOOOOOOO              t1      oldest

When sediment is deposited in STILL water, one horizontal layer and then the next, it can give this kind of pattern in the rock strata, with the bottom layer being the first to form and the top being the last. This does apply in some locations. But there seems to be a problem in the standard approach of evolutionary geology that this model of sedimentation has been over applied to too many locations. In Flood geology, catastrophic processes would produce circumstances in which powerful water currents would exist. Indeed even the daily tides can produce significant currents along the margins of the continents, not to mention many flood-related processes associated with volcanism, earthquakes, impacts from space, etc. The Berthault experiments show that when there is one layer of one type of particles separated from a different layer under it, that does not necessarily mean they formed one after the other in time. They can actually all form at the same time, in a horizontal progression rather than a vertical progression.

This is in accordance with an important principle of stratigraphy known as Walther's Law (or the law of correlation of facies). This law describes how sedimentary rock units that formed beside each other horizontally end up superposed on top of each other. This occurs not because they are somehow moved or turned over, but because multiple horizontal layers may form simultaneously. This law, then, describes a relationship between the horizontal and vertical variations in sedimentary materials. Geologists routinely apply this law on small scale phenomena, such as stream deposits. It could apply in some larger scale phenomena where evolutionary assumptions tend to prevent its application. In encountering rock strata classified as different geological periods, such as Jurassic and Cretaceous for example, evolutionary geologists would not think of applying Walther's Law and considering the whole sequence being related and having formed in a short time. But, large scale high energy phenomena could deposit according to Walther's Law as well. A global Flood could provide conditions where this could take place. Thus, Berthault's work has important implications for sedimentary geology. Not because the deposition process he studied is unknown to geologists but because it can be applied on a larger scale than is usually assumed.

Berthault's research is not guesswork but is shown in flume experiments in which a current is setup and the particulate mixture is made up of various sizes of sand grains. The sediment is mixed and then put into the water flow in the flume, which is a special channel with machinery that can recirculate both the water and the sediment. The engineering laboratory at Colorado State University did the large scale flume experiments, other experiments have been done in France and most of the technical papers on it were presented to French geological societies, with the exception of some creationist conferences. Most of the experiments of Berthault and Julien seem to include sand particles ranging from 0.11 mm to 1.90 mm. The flow velocity of the water ranged from about 26 to 52 cm/sec. On the other hand, some of their experiments were at higher flow velocities (such as 3 m/sec), allowing the mechanism to be applied at the Grand Canyon. There were a number of different types of sand particles studied. Some were coal particles, some limestone, some angular, some rounded, some light, and some dark. Particle mixtures for the flume experiments were made by taking an equal weight of material from two of the various types of particles. Some experiments were done by pouring the mixture through air, some through standing water. Some mixtures segregated into separate layers and some did not.

Three possibilities in the segregation of particles:

1) No segregation - This means the particles are randomly mixed, not in separate layers.

2) Fine particles can deposit on top of course (larger) particles

In this case, you have particles of about the same size, but different density. The higher density particles will sink first.

3) Course particles can deposit on top of fine particles

This occurs in mixtures of particles of different sizes and densities. It also occurs in underwater turbidity deposits and debri flows, where particles are suspended in water currents and then deposited.

The Moving Water Experiments

There is an important relationship between the water flow velocity, the size of the particles, and whether the particles will be picked up by the current (erosion) or deposited by the current (deposition). For a certain particle size there is a flow velocity that will begin to erode or suspend the particles of that size in the water, rather than depositing them on the bottom. In reading about the Berthault experiments, it is easy to misunderstand the experiment and think that they are in the wrong velocity region for what they are trying to simulate, so their conclusions are not valid. However, it is important to understand how the experiment is done. The flume is several inches deep. The flow velocity is set just fast enough to suspend and carry the particles, so it is just barely in the velocity range for erosion. But then a flat object, called a gate, is introduced at a certain point in the flume to obstruct the flow. What this does is that the lower part of the water, nearer to the bottom, moves slower than the water near the surface. So, the flow velocity in the bottom half is slow enough to just be in the deposition range, so the particles sink to the bottom. Thus the gate allows them to trigger when and where deposition will take place. This is all a normal experimental procedure. To apply the mechanism for real rocks, one must determine from the site some information about the particle sizes, densities, and what the water velocity would have been when the sediment was laid down.

Here's the important thing. The experiments show that the particles are sorted by the moving water so that the smaller particles deposit first (less bouyancy) and the larger particles are essentially carried along by the current longer, so that the larger particles roll over the smaller ones and this makes the fine particles deposit below the larger particles. But, what looks like horizontal layers when its all over are not of the same age. All the "layers" form simultaneously, in the direction of the current. This is known to occur in floods and observed geological events, especially around coastlines or beaches apparently. It appears to have occurred at the Grand Canyon on a huge scale, and it is consistent with principles of hydrology in experiment. The process also occurs in deep water deposition processes such as turbidites and debri flows. This is very significant because many formations can be explained as underwater turbidity deposits. Below I will try to show a time sequence of how this works.

OOOOO                                    t1                    Current direction >>>

ooooooooo                               t1

::::::::::::::::::::::                             t1

OOOOOOOOO                       t2

ooooooooooooooooo            t2

::::::::::::::::::::::::::::::::::::::            t2

(older) > (younger)

This means that the portion of the sequence that is really of the SAME AGE would be something like this.




To form multiple layers, all that is necessary is changes in current velocity. A change in the current velocity was shown in the experiments to cause multiple sets to form simultaneously. But in this mechanism the horizontal layers are always arranged such that the material with the smaller sized particles is on the bottom.


One of the interesting implications of this is that fossils on the bottom are not necessarily older. In my simple illustrations above, fossils deposited farther left are older and fossils farther to the right were deposited more recently, regardless of whether they are on top or bottom.

Now, these results are really undisputable. You can see it all happen in Berthault's videos, which show the flume experiments. The question is how broadly applicable is this to real sedimentary rocks. Berthault makes the following argument. In the deposition of strata according to processes we can see today, which would be a more broadly applicable assumption? 1) That deposition takes place in still water or 2) that it takes place in moving water. Moving water would be more typical of real observed processes. Yet, the Principles of Superposition and of Continuity mentioned above seem to assume still water. Therefore these two principles are unrealistic and called into question.

Another implication of this work is that if a whole series of layers can form simultaneously, then the strata could form quickly if you have a catastrophic event of the right kind. These experiments also showed fine laminations in the deposited sediments. It seems that sedimentary layers form in a manner completely surprising, completely different than what you might naturally think, just looking at a sequence of horizontal layers on top of each other.

This kind of research needs to be considered in evaluating the significance of the geologic column. The geologic column was based on observations of similar strata in different locations, based on the type of rock. But considering only correlations of rock types, it was only viewed as a relative chronology, not an absolute one. And, at that time, many geologists were apparently creationists who believed in a global Flood, or at least some were. Then, fossils and old-age assumptions were factored in so that the geologic column could be made an "absolute chronology." It is primarily the second idea that it is an absolute chronology which is questioned by creationists. However, the above experiments and other issues may imply that the geologic column may not even be valid as a relative chronology, at least not in a way that applies the same all over the world. If sedimentary rock layers have formed by Berthault's mechanism, then they do not show evidence of macroevolution (large changes from molecules to man) over long periods of time, but are simply the result of what water can do in sorting and transporting particles. If fossils in the sediment are laid down in a horizontal way (left to right in the illustrations above) rather than from the bottom up over long times, then the placement of the fossils has nothing to do with changes in living things over time as macroevolution implies. But how common are fossils in sedimentary rocks that this mechanism could apply to? Berthault does not comment at great length on how deposition of fossils relate to this experimental research. The placement of fossils may tell us more about how living things were buried, than where or how they lived.

This tends to call into question some evolutionary assumptions in geology and confirm a young-age Flood view of geology, though there are unanswered questions to be pursued still in "Flood Geology." Sometimes geologists, in looking at a sequence of finely laminated sedimentary rock layers, count how many layers and assuming 1 or 2 layers (for instance) form per year, multiply to obtain the total age of the whole layer (facies). Sometimes this approach is not applicable and a mechanism like in Berthault's experiments may apply. This would mean the sequence could form quickly in a manner that indicates hydrologic processes, such as flooding or tides or turbidity flows under water, rather than slow build up over long periods of time.

Berthault uses his research on sedimentation to attack the very basis of the evolutionary Geologic Column. There are locations in the world where the entire sequence represented by the Geologic Column does exist. But a location that allowed you to drill straight down in the rock and obtain a core sample that would contain all the layers would be a very rare exception. Often various layers are missing from the sequence, sometimes enough to represent a long period of time, by an evolutionary viewpoint. The Geologic Column is based on some questionable assumptions about fossils and rock layers and how rocks can be correlated in age from place to place. It possible to correlate rock strata from one location to another to a certain degree. But, whether it is possible to say that a Triassic rock on one continent is of the same age as a Triassic rock on another continent thousands of miles away is another question. These issues are matters creationist geologists continue to research.

What are the conditions that would stop or prevent Berthault's sediment segregation mechanism?


Note that Berthault and Julien are not the first to do experiments of this nature and this mechanism was not totally unknown prior to Berthault's experiments. The research of Berthault and Julien does clarify the physical mechanism for this kind of segregation of particles. The implications of hydrologic processes like this are often not adequately incorporated in the way geology is taught. Real physical principles of hydrology do not always seem to bear out some ideas taught to beginning geology and earth science students. I seriously doubt that this type of mechanism is given adequate consideration even by geologists in the field, in some cases. Geologists would routinely apply Walther's Law for smaller scale phenomena such as a stream deposit or small valley. But on larger scales it is not considered, though there is no reason the physical mechanism in Berthault and Julien's experiments could not be upscaled to higher water velocities and sediment particle sizes. This type of mechanism might apply more often in real rocks than might be expected by an evolutionary geologist, if a global Flood actually happened. In such a Flood, there would be many locations where this mechanism could apply, though it would not apply for all sedimentary rocks. The Biblical Noahic Flood would have been a complex event and many different types of erosion and sedimentary processes would have been involved during and after the Flood proper.

Berthault and Julien's mechanism would not apply where there were highly turbulent environments. It may not apply for some mixtures of small silt or clay sized particles. Most importantly, some mixes of sediment particles could not sort by this mechanism because the mechanism depends on large particles being able to roll over small ones and allow the small ones to fall down in the spaces between the larger particles. In still water and air, somehow the particles seem to sort themselves in a similar manner just before they come to rest.

Still, the mechanism has been applied successfully at the Grand Canyon by Steve Austin, geologist at the Institute for Creation Research. It has also been applied in Colorado, at a location known as Bijou Creek, where there are extensive laminated sand deposits. Berthault's experiments are actually broader than what has been explained above. He also showed that some sediment mixtures laminate even falling through air and sometimes in still water, but it is not clear how much this applies to real rock layers. The mechanism described above for moving water Berthault refers to as the "mechanism of non-horizontal layers." This mechanism can apply in either transgressive environments (such as a rising tide) or regressive environments (a falling tide, for instance). Berthault also did experiments in which he started by grinding up a sedimentary rock into particles, mixed them, then redeposited them and reformed a laminated deposit similar to the original rock's layers. And, this only required a brief period of time. Some of Berthault's experiments on lamination also were done with finer silt and clay particles, which again showed lamination in both water and air. Papers on this aspect of Berthault's work were apparently published only in the French journals (see Austin below, page 54, ref. 68). Also, Berthault and Julien's experiments also show that layers of sediment can form on slopes of significant angles. This means that sediment layers do not have to form on a horizontal surface. Geology students are always taught in introductory Earth Science and geology that sedimentary rock layers always started horizontal and then were tilted after they formed, to form sloped strata as we see today. It seems this is not always the case.

It may be that Berthault somewhat overstates the applicability of this work to real rocks in the field in some respects. The main reason I say this is that in real sedimentary rocks, the type of mix of particles may not be conducive to this mechanism.  Also, very turbulent conditions would definitely take place in Noah's Flood, and this would prevent this mechanism in many cases. On the other hand, where there would be smoother water currents, but with changing current directions and flow velocities, it would be very conducive to the "mechanism of non-horizontal layers." I believe this work does have a lot of applicability in the real world and needs to be considered seriously by geologists. This mechanism is in fact applied by geologists, but on limited scales. The significance of Berthault's research, it seems to me, is not that it refutes the geologic column (it does shoot certain holes in it) or that it explains all sedimentary rock. Its significance is that the mechanism is better understood thanks to Berthault and Julien and the mechanism can be applied on larger scales than most evolutionary geologists consider. This research shows that principles of hydrology are very applicable to interpreting sedimentary rock in a young-earth Flood geology context. Laminated horizontal layers do not require long periods of time to form on small scales and they need not on large scales under the right conditions. Evolutionary geologists believe that present small scale slow processes, scaled over long periods of time, can explain the large scale formations of the Earth. But instead of scaling up the time to explain Earth's features, it is more physically realistic to upscale the energy and catastrophic nature of the events to explain Earth's features. Geologists do apply catastrophic approaches today on scales that are logical extrapolations of what we see today. But, many features on Earth's surface point to processes of the past that were of a much larger scale than is observed in the present. Evolutionist geologists propose many separate catastrophic events over long periods of time to explain Earth's geology. But one global tectonic and sedimentary event such as the Flood would spawn many other smaller catastrophes. These catastrophes still threaten us even today!

Many other topics in geology have been researched by creationist geologists and related to the possible processes of a global Flood. There is often more than one way to interpret a given set of facts, especially when dealing with scientific origins. There are also multiple possible approaches to some geological problems that creationists are researching. But, I believe qualified creationist geologists have done much valuable research that shows how many features of the Earth under our feet can be explained by a global Flood, just as is described in the book of Genesis.


Austin, Steve, Editor, 1994, Grand Canyon: Monument to Catastrophe, Institute for Creation Research.

Berthault, Guy, 1998, Genesis and Historical Geology: A Personal Perspective, Creation Ex Nihilo Technical Journal, Vol. 12, No. 2, p 213-217.

Berthault, Guy, 1997, Sedimentation Experiments: Is Extrapolation Appropriate? A Reply, Creation Ex Nihilo Technical Journal, Vol. 11, No. 1, pp 65-70.

Berthault, Guy, 1994, Experiments on Stratification, Walsh, R E. (editor), Proceedings of the Third International Conference on Creationism, Technical Symposium Sessions, Creation Science Fellowship, Pittsburgh, PA, pp 103-110.

Bogs, Sam Jr., Principles of Sedimentology and Stratigraphy, pub. by Macmillan, 1987.

Cooke, Jim, 1999, Personal correspondence. Jim Cooke is a geologist for Mobile Oil Corporation and specializes in stratigraphy.

Julien, Pierre Y.; Lan, Yongqiang; and Raslan, Y., 1998, Experimental Mechanics of Sand Stratification, Creation Ex Nihilo Technical Journal, Vol. 12, No. 2, pp 218-221.

Julien, Pierre Y.; Lan, Yongqiang; and Berthault, Guy, 1994, Experiments on Stratification of Heterogeneous Sand Mixtures, Creation Ex Nihilo Technical Journal, Vol. 8, No. 1, pp 37-50.

Hoskin, William, 1997, Sedimentation Experiments: Is Extrapolation Appropriate?, Creation Ex Nihilo Technical Journal, Vol. 11, No. 1, pp 61-64.