Interview with geology researcher Zoltán Sylvester: understanding river meandering through time

Zoltán Sylvester is a research scientist in the field of geology and a world-known sedimentologist, stratigrapher, and modeler. He is a co-PI of the Quantitative Clastics Laboratory (QCL) at the University of Texas at Austin in the United States. His work requires lots of data collection and processing, as well as the use of mathematical models and computer simulations to build 2D and 3D models. Zoltán Sylvester works especially on river meandering over time and how to precisely track how a river – that can sometimes erase its own sedimentary deposits – has moved and changed its path over time.


Hi Zoltán Sylvester, thanks a lot for sharing your time and expertise through this interview. Could you tell us more about your field of research ?

Yes, thanks for your interest. My primary research interests are in clastic sedimentology and stratigraphy (1). Stratigraphy is essentially what evolving depositional landscapes leave behind; therefore, I think there is a lot of opportunity in using geomorphologic insight to better understand stratigraphy.
I work at the University of Texas in Austin and I am the co-PI of a research group that focuses on clastic (2) sedimentary systems. One of the things we do is to look at modern rivers and coastline changes, how they evolve and what this implies for the sediments left behind. My field of research has practical implications, historically for the oil and gas industry, nowadays for CO2 capture and storage, and geothermal energy as well. We document the structure of the subsurface and how fluids are moving through the pores in the sand.
During the last few years, I have spent significant time working on the morphology and stratigraphy of meandering / sinuous channels. The focus was initially on submarine channels and largely based on three-dimensional seismic data and simple kinematic modeling. However, I think that development of sinuosity in submarine channels shares similarities with meandering rivers, and a lot remains to be learned from trying to apply our knowledge of fluvial systems in the submarine realm. In addition, there are many simple aspects of fluvial meandering models that may be well known to geomorphologists but have not been fully explored yet in terms of impact on stratigraphy. 

I discovered your work through one 3D object you made from a river meandering computer model, and then I saw that you were a scientist. What is your perspective on artistic creation in relation to your academic research ? Do you consider one to be a personal hobby and the other to be your work, or is it blurrier than this ?

Primarily a scientist, I find patterns I see beautiful and very artistic. I make generative art from river models that are designed on the computer and sent to a drawing robot called  AxiDraw to make the drawings. I have made prints of  meandering river models with this technique. I would say it’s partly because of my research, partly because I am drawn to the interesting and beautiful patterns it generates. The file has to be vector graphics (just lines), SVG format, that is then fed to the robot that moves the pen along two axes (x,y) and draws every line with high precision. I have an Etsy shop for those. There is definitely no clear line between the research I am doing and the prints and objects I make. I would say they are feeding on each other.

One example of Zoltán Sylvester‘s artwork that was generated with the AxiDraw

What were your initial thoughts and research questions leading to the making of the meandering through time 3D object ? What inspired you ?

The surface of the earth is relatively stable at a human scale: you don’t notice the vertical changes in topography unless there is a massive event (like an earthquake that creates a two meter uplift). Usually things happen at a geological rate, very slowly, especially in coastal areas. Also, rivers are moving a lot faster laterally than they are vertically: this leads to the river erasing its own traces and makes it close to impossible to reconstruct its whole history. We dig through sediments to understand that history [the deeper, the more ancient], but it is likely that some layers will be missing or a lot thinner than they should be to accurately reflect the shape of the river at that moment in time. That is also why we use computer modeling to simulate how the flow of water would have moved and what sediments it would have left behind.

It is therefore challenging to understand properly how a river has evolved through time by looking at a static, 2D image (such as on Google Earth). It’s hard to comprehend how the patterns came to be. It remains possible to use tools like Google Timelapse and look at old satellite images to see what happened in the last decades. I have actually watched those visualizations over and over again, and analyzed them to measure the rates of change. We have also worked to develop ways to measure those changes, but 2D still does not convey well enough the idea that the shapes of rivers have completely changed over time and makes it hard to grasp it all at once. So my initial problem to solve was to find a way to visualize everything at once, chronologically.

The 3D models we have generated are interesting because time is on the vertical axis: this allows us to see everything in one object and to really measure how much the river’s path has evolved [usually from a simple, straight line to very complex meander patterns].

Top view of Zoltán Sylvester’s first model print. The yellow color represents the most recent meanders, whereas the purple color represents the oldest river path.
The river’s path was almost straight at the beginning and that it got curvier and curvier through time.
The first model printed by Zoltán Sylvester had no base, so it could not stand on its own. It is however very “readable” already like this, and easy to manipulate.

In fact, these models are chronostratigraphic models, a very specific format [for representing stratigraphic data]. Often these diagrams contain a lot of ‘blank’ space with no data. It’s a fundamental problem of stratigraphy: often what you see in deposits is extraordinary events (large floods, tsunami, landslides), and tens of thousands of years can be just a few centimeters of “not so interesting mud”. We have to be careful not to over-represent major events that happened very fast versus very long, slow evolutions that are less visible in the material we work with but contain information about the changing shape of a river.

Does the model embeds data from an actual river ?

The 3D prints I have made so far (with time on the z axis) are from a numerical model, not a specific river. I am interested in trying this approach on a specific river, for instance South American fast moving rivers, especially the Mamoré River in Bolivia that we have studied.

What are the time scales involved in the study of river meandering, and how would you reflect those in further 3d objects ?

Most river banks move very slowly by human standards, but some rivers move extremely fast (even for human scale). It is the case especially in South America where the river bank can move 30 meters in one year, which is extremely fast from a geological point of view. Anything above 10 cm per year is already a lot. In general these models would capture a minimum of a few decades to hundreds of thousands of years.

Are the 3D models being used as lab-objects or do they remain at home for your personal enjoyment ? Have you showcased them in another context, like an artistic venue ?

I do show the 3D models to colleagues and students. They help understand the river’s evolution through time, which is difficult to fully understand by looking at 2D images on a screen, so they became lab-objects indeed. People are usually not so much interested in the 3d printing process, but in the data it represents and how it allows them to understand the history of the river better.

I also see the models as art. I would love to see something like this in a museum. I think people are interested in this kind of representation. I have been following for a long time the creators of the design studio “Nervous System”, who create 3D art inspired by biological processes that are producing similar patterns as river meandering: it’s not the same, but the way it goes from simple and linear to complex and curving patterns, it’s striking to see those similarities.

A creation of the Nervous System design studio : those pieces have been made by a piece of software that simulates the biomechanics of nature to yield objects that resemble leaves growing and flowers blooming. Nervous System calls this technique Floraform

I have also made more traditional topographic 3D printable models, so I have a broad interest when it comes to representing the scientific material I work with.

What has been your specific process to create this 3d model ? Do you own a 3d printer ? Did you have to learn or acquire some new tools/skills to transition from 2D graphic designs to 3D models ?

It’s actually one of my colleagues who looked at the 3d model on the screen and said “you should print it!”. But there were some technical challenges because it requires more thinking than just drawing in 2D. Something that looks great in 3D on the computer will not necessarily print well either, so you really have to create a very clean surface, geometrically closed and with not too many nodes or edges. The 3D printing process actually improved our model by forcing us to make it smoother. 

I say that in part it’s art to me, but on the other hand you need a very precise way of tracking where the river goes in order to make such an object. The river bank is a line in 2D. To map it like this requires to take an arbitrary point on the line and to pick another point on the next line to pair with : that is not easy to do mathematically, and it is a coding challenge as well. We had to figure out a way to not leave much error, it has to be very precise in order to make a nice looking object; the surface has to be smooth and water tight. We spent time figuring out how to do that. The initial goal was not to print a 3d object, it was science !

I don’t have a 3D printer in the lab, but I am thinking of getting one. I am outsourcing the printing to a company called Shapeways. The reason I am postponing the decision to acquire a printer for the lab is because I quite enjoy the fact that the models have color (the color is embedded in the printing material). In this case, the color represents time and I think that the model looks more interesting this way. Shapeways has a wide choice of printing materials; the one I am currently using is called “Sandstone”, which is a kind of plastic with a rough texture that feels like sand. I like it because of its matte, non-shiny surface.

Do you have other ideas for making more 3D physical objects inspired by the first one you designed and/or by another research question or case study ?

If you have a set of lines (curves) that describes something through time, you could use the same process to make a similar model. I would like to explore what other types of lines, and systems, would look like. It could be data, a bit like the project I saw on your blog, to represent a dataset through time [n.d. : the Critical Shapes CO2 vase project]. You could for instance represent temperature per year on a circle and stack the circles. I could use the algorithm to make a 3D time model of that.

It would be interesting to generate a 3D object from the evolution of rivers with multiple channels, e.g., braided rivers and deltas : stacking those images through time would make a super interesting object. It would be a more difficult project : it involves more lines with sand banks, and then it becomes a data accounting problem.

Are you aware of any other colleague making meandering 3D artifacts or any other objects representing sediments changes through time ?

I have not seen anybody creating 3D-printed objects placing the time on the vertical axis. In stratigraphy, there is a classic paper by Harry E. Wheeler that was published in 1964 in Geological Society of America Bulletin, which describes this idea that geological cross sections can be better understood by placing time on the vertical axis. For a long time I thought this was just an interesting academic exercise, but now I am completely convinced that understanding those diagrams are essential to understanding geology.

Links

Zoltan’s 3D models : https://www.shapeways.com/shops/riverplots-3d

Zoltan’s academic articles and blog posts : https://zsylvester.github.io/

Zoltán’s Etsy shop: Riverplots

To follow Zoltan Sylvester on Twitter : https://twitter.com/zzsylvester

WHEELER, HARRY E. « Baselevel, Lithosphere Surface, and Time-Stratigraphy », GSA Bulletin. 1 juillet 1964, vol.75 no 7. p. 599‑610. [Online] https://pubs.geoscienceworld.org/gsa/gsabulletin/article-abstract/75/7/599/5772/Baselevel-Lithosphere-Surface-and-Time

Definitions 

(1) Stratigraphy is a branch of geology concerned with the study of rock layers (strata) and layering (stratification). It is primarily used in the study of sedimentary and layered volcanic rocks. Stratigraphy has three related subfields: lithostratigraphy (lithologic stratigraphy), biostratigraphy (biologic stratigraphy), and chronostratigraphy (stratigraphy by age).
Source : https://en.wikipedia.org/wiki/Stratigraphy

(2) Clastic rocks are composed of fragments, or clasts, of pre-existing minerals and rock. A clast is a fragment of geological detritus, chunks and smaller grains of rock broken off other rocks by physical weathering. Geologists use the term clastic with reference to sedimentary rocks as well as to particles in sediment transport whether in suspension or as bed load, and in sediment deposits.
Source : https://en.wikipedia.org/wiki/Clastic_rock


OpenEdition suggests that you cite this post as follows:
Anne-Laure Fréant (September 7, 2022). Interview with geology researcher Zoltán Sylvester: understanding river meandering through time. Datartefacts. Retrieved July 19, 2024 from https://doi.org/10.58079/nh56


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Search OpenEdition Search

You will be redirected to OpenEdition Search