Peter Klisiwecz ’25 was a physics major at Amherst who graduated last month. For his senior thesis with Professor Nick Holschuh in the geology department, he developed a method to infer glacial crystal orientation fabric from decades of existing radio data taken at different times and directions, instead of having to conduct another survey. Peter kindly sat down with me to talk about his thesis and his journey at Amherst College.
Ryogo
Would you like to start with introducing yourself?
Peter
My name is Peter Klisiwecz. I’m a senior physics major. I just completed my thesis with Professor Nick Holschuh from the geology department, though it was a physics thesis, and that was on inferring ice sheet crystal orientation fabrics using airborne radar data.
Ryogo
What was your goal for your senior thesis?
Peter
For my thesis, what we were trying to do was to use airborne profiling radar to infer what’s known as glacial crystal orientation fabric. Ice crystals are anisotropic, meaning their mechanical, thermodynamic, and electromagnetic properties depend on their orientation. Glaciers and ice sheets are polycrystalline, made up of many crystals oriented in various directions. The distribution of these orientations is referred to as the crystal orientation fabric. When this distribution is non-random—when the crystals are aligned—the glacier ice inherits anisotropic properties.
This is really important for understanding ice flow and dynamics, which directly relate to sea level rise. Crystal orientation fabric can alter the flow and viscosity of ice by an order of magnitude or more, resulting in major uncertainties into ice flow models. Currently, measurements of glacial crystal fabric are extremely sparse because methods used historically for collecting this data are limited, meaning that these modelers don’t really have the breadth of data necessary to analytically account for the crystal fabrics.
So my thesis is looking for how we can use the vast amount of existing radar data that’s perhaps not super well suited towards this purpose, but nonetheless could potentially be used to measure crystal fabric. Instead of having to conduct new surveys and things of that sort, which are very expensive and logistically intensive and take time, we use the data that we already have to try and infer it.
Ryogo
And you say it’s radio data. How do you have such a big survey?
Peter
That’s a great question. There are three primary methods we use to investigate glaciers. The first is ice coring, where you drill a kilometers-deep borehole and extract an ice core. This gives very precise measurements, but it’s extremely logistically intensive. You have to set up an entire field camp, and the process can take years—it’s very laborious.
The second method involves seismic techniques. You use seismometers and generate seismic waves—sometimes with plastic explosives or other equipment—to measure how those waves propagate through the ice. This allows you to infer information about the glacier’s interior.
The third method, which I’m most interested in, is radar. There’s a wide range of radar systems. Some are ground-based, but I focus on aerial radar profiling. These airborne systems can cover much larger areas because you don’t need to be on the ground. You mount the radar antenna—both transmitter and receiver—on an airplane, and just fly over the ice sheet to collect data.
Since the late 20th century—around the 1970s to the present—hundreds of thousands of line-kilometers of radar data have been collected essentially spanning over Greenland and Antarctica, in a way that you don’t have for any other type of data. You can just fly an airplane around and gather data efficiently over vast regions.
The radar works by transmitting a signal that propagates through the ice and reflects off various internal interfaces. These could be physical or chemical boundaries within the ice. We often talk about depositional layers, where atmospheric chemistry varies from year to year, affecting the chemical makeup of snowfall. Over time, this alters the chemistry of the glacier ice, and you can detect those differences in the radar signal.
For my thesis, I’m especially focused on comparing radar signals of different polarizations. Because ice is electromagnetically anisotropic, this anisotropy can cause birefringence—a splitting of the radar signal depending on polarization. By measuring this birefringence, you can infer the degree of anisotropy in the ice.
Ryogo
Could you elaborate on how the radar data comes as? Does it come as a spectra or any other forms?
Peter
That’s a great question. The radar data comes in the form of what we call a radargram. It’s essentially a two-dimensional profile of the ice, where the x-axis represents distance along the flight path, and the y-axis is depth in the ice. In its rawest form, the y-axis is actually two-way travel time—from when the radar signal is transmitted to when it’s received—but that can be easily converted into depth.
From the radargram, you get travel time information, intensity information, and—importantly for my purposes—phase information. You also get all the radar characteristics, like the polarization, from the properties of the radar system itself.
So you have these two-dimensional profiles, and because I need multiple polarizations, I look at places where radar flight paths intersect. At those intersections, I have several radargrams from different polarizations. What I do is take very small depth profiles—just tiny slices that go all the way down through the glacier—at those points of intersection.
Ryogo
I see. So, what were some of your results that you were able to take away from the project?
Peter
So, the first-order goal of my thesis was to ask: is this even feasible in the first place? One of the main challenges is that, in order to measure birefringence using radar data, you need to isolate the effects that are actually due to birefringence. That means you have to account for any differences between the radar observations that aren’t caused by birefringence. You’re trying to measure the birefringence, and so you need to measure the effect of birefringence upon the radar signal.
In the ideal case, many similar studies use specialized radar systems that measure several polarizations at essentially the same location and time. So those measurements are nearly identical except for their polarization, which makes it much easier to detect birefringence.
But in my case of these radar crossovers, the conditions vary greatly between the two observations. You know, they may not be taken on the same day. They may be taken days or weeks apart. Even though they use the same equipment—because I’m looking at a specific subset of data—the airplane’s elevation can vary by hundreds of meters. There are also other complications, like the focusing that’s applied to the radar data. So the big question was: is it possible to isolate the effects of birefringence from all these other sources of variation?
To investigate that, I analyzed crossover points that were close to known ice cores, since we already know the crystal fabric at those locations. I then compared the anisotropy I inferred from the radar data with the anisotropy measured in the ice cores. What I found was that they were relatively in agreement. Not a perfect match, but definitely similar enough to suggest that this method has real potential and can, in fact, accurately infer anisotropy. So I’d say that the first goal—demonstrating feasibility—was accomplished.
The next question I wanted to explore was: where does this method work well, and where does it break down? To answer that, I looked at observations from more complex flow fields. I focused on one glacier, which has extremely fast flow—hundreds to thousands of meters per year—and very complicated flow fields and internal layering.
Unsurprisingly, in that region, the method didn’t work as well. It didn’t yield accurate measurements of birefringence or inferred anisotropy. What I found involves a statistical measure called coherence—the similarity between two measurements. In high-coherence areas, typically simpler flow fields, the method produced accurate results. But in low-coherence areas, where you have all these extra factors causing differences between the measurements, the method was much less effective.
Ryogo
That’s really cool. Now, I mean, your thesis is done. But what is some work that can be done? I don’t know if you’ll pick it up, but maybe by future students or in the field. What are some things that can be done to this method specifically, and how can this method be used for further research?
Peter
Yeah, that’s a great question. My mind goes in a few directions with this. On a small scale, within the method I developed, there’s definitely room for improvement. For instance, I often found myself building calibration algorithms and thinking, “Wow, this would be so much better if I knew more machine learning.” A trained model could probably trace features in the ice more accurately than the basic algorithm I came up with. So there’s some refinement that could be done with better tools and more time.
Another area is the analysis of the results themselves. Because I was up against the thesis deadline, I had to wrap up without doing the more rigorous statistical analysis I would’ve liked. One example is the coherence of the radar signal—coherence is critical to the accuracy of the inferred anisotropy, and while I discussed that qualitatively, I didn’t have time to dive into it quantitatively. It would be really exciting to develop a more formal relationship between coherence magnitude and uncertainty in the anisotropy inference.
There’s also more technical tweaking that could be done to improve the signal-to-noise ratio. But what excites me most is the method’s potential for broader application. For my thesis, I only looked at one field season, 2014, and only a few selected locations. But there are years and years and decades of radar data out there. And so if we can synthesize data, apply this method more, and see if you can find regions where you know that exhibit high coherence, then this method has the potential to infer these crystal fabrics at a very rapid rate—much, much more rapid than if you had to go out and do a survey every time you wanted to learn about a new location, right? To me, that’s really where the exciting next steps in this method lie is in its application
Ryogo
Thank you so much. Is there anything that I missed or anything you want to talk about on your thesis?
Peter
I think that was pretty comprehensive. I mean, I have to think for a moment, but, yeah, in terms of the thesis itself, I think we covered it quite comprehensively.
Ryogo
Cool, let me know. If you remember something, we can always circle back to it.
Peter
I will let you know. But I think we really covered it all.
Ryogo
Then, I guess I’ll ask more personal questions.
Peter
Absolutely.
Ryogo
Starting from the beginning of your physics career, how did you become interested in physics?
Peter
That’s a good question. Well, I’ve always liked math. Actually, this is a great story. I had a great physics teacher in high school—definitely an eccentric guy, as high school physics teachers often are. But coming into Amherst, I knew I liked math and science, though I wasn’t totally sure what I wanted to do.
I remember I had a Zoom call with Professor Friedman, who happened to be my advisor, before my freshman fall. We were talking about course registration, and I told him, “I know I want to take math and some science, but I’m really going back and forth between physics and chemistry. They both seem so cool.” I forget exactly what he said— “Do you want to flip a coin?” or something? But then he was like, “Well, I’m a physics professor, so of course I’d encourage you to take physics.” And I was a young freshman, very easily influenced, so I was like, you know what, sounds good to me.
So I took Physics 123 in the fall with Professor Hunter, who I thought did an excellent job with the course—he had all these awesome demos. I remember the fire extinguisher bike and all that. I thought it was a lot of fun. My roommate was also taking physics, so it all seemed pretty good. Still, I was a little hesitant. I was still a little trepidatious. Physics has such a reputation—it seems so difficult.
But I knew I really wanted to take Modern Physics. I think that’s a common experience—it just seemed super cool. I wanted to learn about all the Einstein stuff, and I figured, okay, I’ll stick with physics at least through Modern, and then I’ll decide. And then I thought, well, if I’m taking Modern, I might as well take Waves too, just in case I want to major.
And then, halfway through sophomore fall, I kind of stopped and thought, “Okay, I’m taking Modern, I’m taking Waves, I already took 123 and 124… this stuff seems pretty cool, and I could see myself doing it long term.” So I thought, yeah, might as well just declare.
At every step I was like, “This next class seems interesting, I’m curious about it, so I’ll take it,” and eventually I found I’d done half the major. I’m really glad I did it.
Ryogo
Did you end up taking chemistry as well?
Peter
No, I didn’t. I had this grand plan—another classic college moment—during Spring Break of freshman year. I sat down and mapped out my entire college career, every single class I’d take each semester. I had this idea that I wanted to know everything about every science. So I’d major in physics, but also take some chemistry labs, some bio labs—just to stay well-rounded.
That, of course, did not happen. Or, not “of course,” but I decided that I was not interested in that when I realized how many cool physics classes there were that I could take instead. So no, I never ended up taking chemistry.
And, you know, who knows? Maybe there is a world where my first-year advisor had been a chemist, then I’d be sitting here talking about chemistry. But I’m glad I ended up with physics.
Ryogo
How did you end up taking geology classes?
Peter
That’s another great question. So, let’s see… another great anecdote. I was taking the same class with Will Dienstfrey, who was doing research with my current thesis advisor, Nick. I think we were in a linear algebra TA session, and I overheard him talking about his research, and I just thought, “Oh, wow, this guy is a physics major, but he’s using physics to study the Earth. That’s so cool.”
At that time, one of the things that made me hesitant to pursue physics was that I was definitely more interested in the macroscopic—studying the world, nature, things like that—rather than, you know, condensed matter or AMO. I’d say now I’m much more interested in those things, but back then, I was more focused on the real world, and that’s why hearing about Will’s research really caught my attention.
After the help session, I went up to him and basically said exactly that: “Wow, I think it’s so cool that you’re using physics to study the Earth.” And he was super generous and said, “Well, let’s sit down. I can tell you about my research.” So, we ended up sitting down at like 9 p.m., just the two of us, in the SMUD building, and he spent about half an hour telling me all about his work. And honestly, that’s one of those special Amherst moments—people here are just so generous with their time. They’re willing to help, share their passions, and give advice. I really credit Will for getting me on this track.
Then, that same semester, Nick was the speaker for the Physics Colloquium. I was really sick at the time—like, had strep or something—but I still thought, “I need to hear what this guy has to say.” So I bundled up in sweatpants and a hoodie, and if you know me, you know I never wore sweatpants out, even freshman year, and I went to the talk. I was in the back of Kirkpatrick with my mask on, but I had to go. And it was so worth it—it was one of the coolest talks I’d ever been to.
After that, I decided to enroll in Nick’s class, Geology 109, which is called Climate Change: Science and Rhetoric. It’s all about climate change, its science, and its societal impact. I thought, “Okay, this is a great way to get to know Nick better,” and I figured, maybe I’d go to his office hours and talk to him, and maybe he’d let me into his lab. And that’s exactly what happened.
Nick took me on as a research assistant during my sophomore year, over the January term, and I’ve just kept working with him as a research assistant ever since. Eventually, that turned into my thesis.
Ryogo
That’s really cool. That’s such a nice story. Okay, what was your favorite physics course, and also, what was your favorite course in general, at Amherst?
Peter
Oh, these are hard questions. These are hard, hard questions.
Ryogo
Some of your favorite courses?
Peter
Oh, that’s much easier.
I mean, Modern is just—yeah, there’s no course like Modern. It just blows your mind. It seems like magic. You know, it’s like time travel—it’s just so wondrous. You truly feel like you’re uncovering the secrets of the universe through, you know, the power of mathematics or whatever. So in many ways, I think Modern is hard to top.
I also had great TAs for Modern, which made it even better. I think Vietta and Julia Woodward were my TAs—so two great TAs. Great class, great TAs, taught by Professor Friedman. So it was just a great class.
Let’s see—I also really liked E&M II. Coincidentally, also taught by Professor Friedman—so maybe I just like Professor Friedman classes. But yeah, I just think electromagnetism is so cool. It’s just like magic. That’s why I like physics—it’s kind of just… it’s just so cool.
And that class also ended up being directly related to my thesis, because radar is all about electromagnetism and electrodynamics and whatnot. And again, it was a class full of great people. Once you’re a senior physics major, you’re in a class of like ten people, and you know every single one of them. So it’s just bound to be a great time. That was a great class as well.
In terms of favorite course at Amherst in general—that’s a big, big question. Let’s see. I took a course taught by Professor Gomes in the—I think it’s ASLC. The class was called Epic Tales, and it was all about the Ramayana, this ancient Indian epic.
I just found it was exactly what I looked for in an Amherst class. It was just the right amount of reading, the discussions were so interesting, and Professor Gomes was so engaging. Every class, I was never bored—the discussions were always so interesting. And the reading—you’re reading this awesome epic of huge proportions, you know, as epics are. But it was also interesting to analyze it through a sociological lens—like, what are the societal dynamics that are either played out or established by this epic? And at the end of the course, Professor Gomes did an excellent job tying it to the present—talking about fascist movements in India and how they relate to and use the story of the Ramayana, how they see themselves in it and use it to justify what they do. It was just so interesting.
I really enjoy these humanities classes where I get to learn about cultures I don’t know much about. I feel like it makes me a more worldly liberal arts person. Which is what we’re here to do, right? So that was an excellent class.
I also really enjoyed this past fall—I took Philosophy of Space and Time, which was co-taught by Professor Moore from Philosophy and Professor Jagu from Physics. That was really interesting—just talking about the philosophy of time. What is the past? What is the present? All these heady things that maybe we’ve thought about mathematically before, but philosophically it’s a whole other thing. That was a lot of fun.
Let’s see. I want to think for a moment—because I’ve taken a lot of courses at this point, and I don’t want to leave anything out.
Oh my gosh—how could I forget Climate Dynamics with Nick, of course! I mean, I love that class. The physics of the atmosphere and the oceans—it’s just such a dynamic, interesting system. You have this differential heating of the Earth’s surface, which results in all these circulation patterns. But then you’re in this non-inertial rotating reference frame—because, of course, the Earth is rotating—and that results in all sorts of crazy things.
So as a physicist, it’s such a fascinating system to think about. In many ways, it’s so simple, and it’s also so complex. The climate and the oceans—it’s like the perfect physics problem. The underlying concepts are rather simple, but the resulting phenomena are so complicated. I just found that absolutely fascinating.
It actually led me to pursue some atmospheric work—I worked at the NOAA lab last summer doing atmospheric modeling, and that was certainly a direct result of how much I enjoyed climate dynamics. That was a very impactful class for me that I really, really enjoyed.
I can’t believe I almost forgot about it—it just sounds so cool. It is so cool. The climate is so cool. The oceans are even crazier! Because, you know, the atmosphere is free to circulate as it wants, but the ocean is constrained by these continents, and that results in all sorts of wacky things happening.
So take Climate Dynamics! Especially if you want to learn Python—it has a lab component that’s like Python data science. So if you don’t know any Python, which I think is super beneficial to learn as a young physicist, I would highly recommend. That’s a great way to go about gaining those skills while also thinking about some very, very, very interesting and pertinent physics to our current life.
Ryogo
You are making me want to take it! Okay, three more questions. Do you think you’ve benefited a lot from doing senior thesis, and would you recommend people to do it if they can? What are the values and challenges of doing a senior thesis?
Peter
That’s a great question. I think I benefited a lot from doing a senior thesis, and I’m so glad that I did it. But I definitely wouldn’t recommend that every single person do a thesis.
I did a thesis because I already knew I was interested in this research—I wanted to do more glaciology research. Working as a research assistant for, you know, four to six hours a week is fun, and it’s educational and productive and all of those things. But if you really want to dive into something, you need time to dedicate to it. And that was my primary reason: I just wanted to do the research. And I think that really should be your first reason for doing a thesis.
I was also curious what doing such a large research project would be like. A lot of people say that the thesis is the best way to find out whether you’d be interested in grad school—or that it’s the best analog to what grad school would be like that you can find at Amherst. And I was interested in finding out whether I’d like that, so doing a thesis made sense in that regard, too.
Then I was also looking to challenge myself. Sophomore year, I was just taking a lot of physics courses, and that was its own challenge. Then junior year, I took some difficult electives—I went over to UMass and took fluid dynamics there. So I was really pushing myself in that way. And heading into senior year, I was thinking, “Okay, what’s my next challenge?”
I had basically wrapped up the physics major by the end of junior year—I mean, I still took E&M my senior year—but I was looking for something that would really push me, help me develop more skills, and give me room to grow. And the thesis seemed like the obvious answer.
So for me, that’s why I was interested in it. But ultimately, it was all because I really wanted to do the research, and I was deeply interested in it.
Ryogo
What advice would you give to current physics majors, or people who are interested in physics?
Peter
Yeah, that’s a really good question—and it’s a good one for me, because I really like giving advice.
So, this isn’t exactly advice, but more of a PSA I think is important to share: I strongly believe that the physics major is most difficult and most demanding during your freshman and sophomore years. Once you become an upperclassman, the major becomes a lot less intense. Assuming you’ve already worked through the intro series during your first two years, then by junior and senior year, you’re mostly taking upper-level electives. And those don’t have labs, they don’t have discussions—they meet twice a week instead of three times.
I was amazed during my junior year—I didn’t have class on Fridays. And as a first-year or sophomore, I thought that having no class on Fridays was something that only humanities majors got to enjoy. As a physics major, I thought that just wasn’t going to be in the cards for me. But it totally is, once you’re out of the intro sequence. Things really open up.
That’s not to say the upper-level electives aren’t hard or demanding—I spent hours and hours and hours on my E&M problem sets—but I do think it’s important to keep that long-term perspective in mind. When you’re an underclassman in the major, you’re working so hard, and it’s easy to assume that it’ll only get more intense from there. But in many ways, it’s actually the opposite.
That’s my first piece of advice.
My second piece is less practical and more philosophical. I think it is so important that you always follow your interests—when you’re picking classes, when you’re thinking about your future, when you’re planning anything. Always follow your interests.
That might sound obvious—it seems silly to even have to say, like, of course I’m following my interests. But it’s surprisingly easy not to. It’s easy to fall into certain tracks or compare yourself to other people and think, “Oh, they’re following their interests, and their interests are taking them to MIT. But what I’m interested in wouldn’t take me there. So… should I change what I’m interested in?” Or, “Should I just suck it up and pursue the thing that looks better?”
Everyone has different interests, and they lead us to different places. But it’s really easy to fall into the trap of comparison or to sideline your interests in pursuit of what looks more impressive or conventional. And I would strongly recommend against that.
Also, Amherst is a small school—it doesn’t have everything. So it’s easy to get discouraged and say, for example, “Well, I’m interested in meteorology, but Amherst doesn’t have any meteorology classes. Even the Five Colleges don’t really have anything. So… am I just out of luck?” But I think what you’ll find is that, if you look hard, you will find some way to either do work that’s adjacent to your interests, or work that prepares you for what you’re interested in, or something that brings you closer than you thought was possible.
But that only happens if you relentlessly pursue what you care about. So that’s my advice:
Pursue what you’re interested in. Be relentless. And don’t compare yourself to others—because their interests might take them one way, and yours might take you somewhere completely different. And that’s not a bad thing at all.
Ryogo
Thank you. Okay, last but not least, do you think you keep pursuing geophysics and climate dynamics?
Peter
Certainly something related to climate—definitely. Right now, I’m really interested in remote sensing and satellite or aerial detection of greenhouse gas emissions. That kind of technology can be used to identify methane leaks from natural gas pipelines, or to quantify emissions from landfills and other sources that are hard to measure accurately.
So at the moment, my main focus is on emissions reduction. That’s where my strongest interest lies. But I’m also still interested—though to a lesser extent—in atmospheric science and weather-related topics. I’ve done some work in that area before, and it still appeals to me. Things like catastrophe preparedness, climate resilience, or climate modeling are also areas I could see myself continuing to explore.
But again, right now, my primary interest is in greenhouse gas emissions mitigation.
Ryogo
Thank you so much. Anything else you want to say you want to add?
Peter
I think this has been quite comprehensive. I mean, I got to give advice. I got to talk about my thesis. I got to talk about, you know, my time at Amherst. So I think, I think we really covered it all.
Ryogo
I mean, it’s been an hour. Thank you so much for the long interview.
Peter
Thank you so much.