Postdoc Spotlight: Daniel Portner's Tomographic Technique for Exploring Volcanic Systems

Postdoc Spotlight - Daniel Portner
Photo of Daniel Portner taken on the coast of Iceland in February, 2020. Photo by Courtney Shepard.
Tuesday, November 17, 2020 

Understanding the inner structure of volcanic systems is essential to assessing their hazard risk, yet traditional techniques have left large chunks of the lower crust unknown. A recent study by Postdoctoral Fellow Daniel Portner harnesses a technique called tomographic inversion, normally used by doctors for CT scans, to create 3D images of these previously hidden inner structures. 

In this Postdoc Spotlight, Portner explains this new technique; what he found when he used it at the Cleveland Volcano; and how this study may raise more questions about the volcano than it answered. Enjoy! 

1) First, please introduce yourself.

I am Daniel Portner, a Carnegie Postdoctoral Fellow in Seismology and I study the Earth’s interior. 

2) What is your area of research, and why would you say it's important?

My expertise is in seismic imaging or structural seismology. In practice, what that means to me is that my research covers a number of disciplines that rely on a detailed understanding of what the Earth’s interior looks like, including plate tectonics, geodynamics, and, most recently, volcanology. 

3) What are the broader implications of your work?

The bulk of my work until recently has focused on mantle-scale imaging in South America and Anatolia of the subducting slabs that drive each tectonic system. One of the more interesting results to come out of that work is a persistent observation that slabs may be more fragile than we previously thought. Beneath Anatolia, the slabs seem to readily fragment and deform during their descent through the upper mantle, rather than cleanly sinking to the core-mantle boundary. Even beneath South America, where the Nazca slab is one of the largest contiguous slabs on Earth, we see that the slab experiences sharp contortions and tearing during its descent.

Of course, the broad implications of these findings remain to be seen. One of the nice things about seismic imaging studies is that the results serve as a tool for the broader community. My models of the mantle beneath South America and Anatolia will inform the next generation of research into the geodynamics and tectonics of those regions. I’m hopeful that these findings will also help shape the way we think about slab evolution and their eventual fate of assimilation into the ambient mantle. 

Photo of Cleveland Volcano (right) and neighboring Carlisle (left) from a boat. Photo by Diana Roman.

4) You recently turned your attention to the inner workings of the Cleveland volcano in the Aleutians. Can you briefly summarize what you found in your recent study?

Essentially, I was able to rigorously confirm that there is a system of hot, potentially partially molten crust beneath the volcano that extends from the near-surface to the base of the crust. 

A prior study by former DTM Postdoc Helen Janiszewski identified an interesting pattern in the travel times of waves generated at the bottom of the crust that implied that a large body of low seismic velocities should exist beneath the volcano and should extend across most of the crust. 

With my study, I took a more quantitative approach to determining the best-fit shape and size of the low-velocity body. Not only did we confirm that the conduit is real and likely crosses the entire crust, but we found that it is probably much slower than we thought, implying hotter and potentially more molten rock. 

5) How does this change our understanding of the Cleveland volcano? 

To be honest, this raises a lot more questions than it answers. Cleveland is among the most active volcanoes in the Aleutian arc, but it produces relatively small-volume eruptions. With such a well-developed crustal magmatic system, why isn’t Cleveland more hazardous? Are the deeper portions of the system contributing to the persistent activity at the volcano? How much eruptible melt does the reservoir contain, and how is it distributed in the crust?  

One simple way to answer your question is that our results suddenly make Cleveland a lot more interesting (not that it wasn’t before). 

6) Can you explain what a “tomographic inversion” is and how you use this technique to look inside our planet? 

Tomographic inversions are a technique for turning observations of the travel time of waves into two- or three-dimensional models of the wave velocity for the media they traveled through.  

The most common use of tomographic inversions is in medical computerized tomography (CT) scans. With CT scans, X-rays are sent through the body and recorded on the other side. Based on the amount of time it takes for the X rays to pass through the body, we can infer the wave speed of the body. If we do this many times at different locations and angles, we can then put together a three-dimensional model of different wave speeds in the body that serves as a “picture” of our bones and tissues. The inversion part is just a way to make this process easy by performing up to millions of travel time calculations in one fell swoop. 

In seismology, we essentially do the same thing, except the waves we use are seismic waves generated by earthquakes, our recorders are seismometers, and we are imaging rock rather than soft tissue. Our models of rock wave speeds give us a picture of what the inside of the earth looks like, which we can use to infer rock properties such as temperature, composition, and fluid content.

Admittedly, our pictures are a lot more difficult to interpret than those of the medical community because we have much less control over where to put our seismometers and even less control over where and when earthquakes will occur.

This figure shows the cross-section through the crust beneath Cleveland using Daniel Portner's tomography model.

7) Does this technique change the way we study volcanoes?

In theory, we can image volcanoes the same way we image any other part of the crust - by installing a ton of seismometers on top of and surrounding the area of interest and using a suite of conventional tomographic techniques. 

Unfortunately, most volcanoes don’t make that very easy in practice. Most volcanoes are remote and host extremely rough terrain, reducing accessibility and increasing the cost of seismometer installation, so ideal seismic networks on volcanoes are rare. This difficulty means that the field is always looking for new and creative ways to produce robust images of volcanic systems with relatively little seismic infrastructure.  

The technique I developed at Cleveland Volcano adds to this toolkit by providing a way of imaging the deep crust with a relatively small seismic network of only 11 stations surrounding the volcano. The hope is that the new technique will allow us to image more volcanic systems with fewer seismometers and provide us with baseline images at a far greater number of volcanoes.

8) What first inspired you to look at this question? 

After I arrived at Carnegie, I started talking more with the seismologists here about their own interests and recent work, and my supervisor, Lara Wagner, brought up the ongoing research at Cleveland Volcano. 

This work really intrigued me because it was the perfect storm of a complex scientific topic in understanding the relationship between magma reservoirs and the overlying volcanic processes, an interesting problem in figuring out how to use the available data in a tomographic inversion, and a great way for me to break out of my comfort zone into the field of volcano imaging. This dataset holds a lot of possibilities for pushing the field of imaging, and it was just too tempting to pass up. 

9) How has your background influenced your research or approach to science?

I find that in geophysics it is not uncommon for researchers to get lost in the physics while forgetting about the geo. In observational seismology, this happens too often when seismic models are produced that fit seismic data extremely well but cannot realistically represent Earth materials. 

I found geophysics through geology, which I think really helps me to ground my observations in Earth processes. Most importantly, this background restrains me from over-interpreting seismic models without supporting evidence from the geologic record. 

10) Why did you choose to work at the Carnegie Earth and Planet Laboratory? 

I think my answer about how I found my current research sums up well why I wanted to come to Carnegie - to be exposed to new and interesting research opportunities. 

The EPL community is so open and collaborative that I knew I would have ample opportunities to work with new networks and start projects I didn’t anticipate at the time. Those new ideas and collaborations ended up driving my current postdoctoral research agenda.

11) Do you have any advice for current grad students? 

I would say to go after any interesting research opportunities that come up. I value my primary Ph.D. research, but in the end a lot of the more valuable research I did happened because I said yes to trying something new. 

It might sound scary to take on a new project and the time that goes along with it, but when you look back on your Ph.D., that extra avenue of research will be more valuable to you than the months or longer you added to your Ph.D. to accomplish it.

12) Where can people find more information about your work? 

Sure, you can always feel free to visit my personal webpage at to find out more about me and what I do. On the website, you can find access to my published papers, seismic models, and analysis codes; find out the latest that I’ve been up to; and see pictures of some of the cool places seismology has taken me. 

I’d also say that if any of this interests or inspires you, I’m always looking for new avenues of research and new colleagues to work with, so don’t hesitate to get in touch.