What science questions do you investigate?
My work asks how terrestrial ecosystems and the carbon they hold respond to - and retain a memory of - a changing climate, and whether we can quantify and anticipate those responses well enough to act on them. The specific questions are concrete, and they span several systems.
In the Arctic: How fast is permafrost thawing, where, and what controls it? How much carbon dioxide and methane does that thaw release, and how confident can we be in those numbers? When, precisely, does the ground stop and start releasing carbon across the cold season?
In the global carbon cycle: Why do our two independent ways of accounting for carbon - measured from the ground up vs. derived from the atmosphere down - still disagree by billions of tons each year, and how do we reconcile them?
In drylands and rangelands: Which plant species are present, how is land condition changing, and how can that guide management on working land?
In climate intervention: If large-scale efforts to modify the climate become more likely, can we forecast their consequences in a system we observe imperfectly? In a recent study, we assessed mitigation and large-scale intervention scenarios against seven decades of observations, atmospheric reanalysis, and Earth system-model simulations - and found that the interventions consistently overshot, overestimating how much they would steady temperature and rainfall, with sharp regional disparities.
One question underlies all of these: how does the Earth system carry its past forward? Thaw, drought, or disturbances leave traces in the earth system that shapes what happens next, and we build methods to read and understand that trace. To answer them, we build artificial intelligence models that learn simultaneously from satellites, ground sensors, and physics-based simulations while remaining bound by physical laws, combined with information theory and causal inference - formal methods for separating cause from coincidence - and, increasingly, quantum computing.
Collecting thermal infrared (TIR) data while teaching Alaskan primary school teachers field sampling protocol (2023).
What inspired you to pursue a career in earth systems science?
Neither a single moment, nor a straight line. I trained as a field ecologist, spent several years in climate and carbon market policy in Washington, and worked as a full-time studio artist for five years before returning to research. What moved me toward Earth system science was a persistent dissatisfaction with working at the surface of these problems and a wish to work on their mechanisms - the feedbacks, the causes, the way one change propagates into another. In policy, I was describing those feedbacks without understanding them, and that gap is what eventually returned me to science. When I returned, the separate parts of my background did not read as detours so much as different routes to the same subject. I am here because Earth system science is where the questions I care about - how the system is coupled, and how it carries its history forward - can be investigated rather than only described.
Active layer depth (thaw) measurements of permafrost degradation were recorded along established transects between the Sagavanirktok River and the drill sites near Prudhoe Bay in the North Slope, Alaska (2023).
What is one space mission that you are particularly excited about, and why?
The new satellite, NISAR (NASA-ISRO Synthetic Aperture Radar), and its corresponding synthetic aperture radar (SAR) instrumentation. Synthetic aperture radar is an imaging radar flown on a satellite or aircraft. Because it supplies its own energy - transmitting microwave pulses while timing their echo off the surface - the instrument 'sees' at night and through clouds which optical and infrared sensors are not capable to do. "Synthetic aperture" more directly and explicitly means the instrument uses the satellite's own motion along its orbit to mimic an antenna larger than it can physically carry, i.e., larger antenna yields sharper image (meters). For our work, SAR's real value is measuring mm-to-cm changes in ground height between passes - the slow sinking and heaving of thawing permafrost.
NISAR can globally track ground movement and soil moisture availability in the presence of color cover and no solar illumination. These overpasses occur every 12 days, which is a consistency and reach that did not exist before, and one I have built my recent work around to utilize and ask challenging questions. Because permafrost change is relatively slow, spatially 'patchy', and occurs across the circumarctic - and no prior instrument quantifies all three at once. Earlier radar records were "stitched" together from ground campaigns and flight missions with different sensors, incomplete coverage, and irregular timing, yielding estimations at best; thus, it is difficult to separate real ground motion from measurement artifacts. NISAR images approximately the entire land surface of the planet on a fixed ~12-day repeat, at consistent resolution with the same instrument. So, why is this important? This is the first time we can observe the entire Arctic's subsurface deform and surface collapse/heave on a regular cadence as opposed to reconstructing this behavior from scattered observations and measurements. Such regularity is what converts observations into something digestible a model can train on and validate against.
I am particularly excited by NISAR’s circumpolar GeoCryoAI product; our methane-scaling work at the Jet Propulsion Laboratory supported several upcoming Earth-observing missions; and the rangeland work I am involved with utilizes imaging spectroscopy - more specifically, EMIT (NASA’s Earth Surface Mineral Dust Source Investigation) instrument currently operating aboard the International Space Station as well as EnMAP (the German Environmental Mapping and Analysis Program) - to identify plant species, delineate fractional cover, and quantify rangeland health metrics. I use both to identify plant species, estimate how much of each is present (fractional cover), and derive rangeland health metrics.
Surveying contaminants and hydrology along the Sagavanirktok River, AK (2023).
What is one research project that you are particularly excited about, and why?
The projects that stem from, utilize, or are supported by the GeoCryoAI framework, and the line of work it grew from. It began with a spatially-explicit individual-based forest gap model, SIBBORK-TTE, which simulates vegetation and thaw dynamics at the boundary where boreal forest gives way to tundra; we calibrated that model during an internship at NASA GSFC, and it gave the work its mechanistic foundation. GeoCryoAI grew out of it - an artificial intelligence framework, held to physical laws, that learns permafrost thaw and the resulting carbon release directly from data: roughly 62.7 million ground measurements harmonized with about 3.3 billion satellite observations. Using GeoCryoAI, we resolved the "zero curtain" - the window during autumnal freeze-up and spring frost-heave when the soil hovers near 0ºC and continues to release carbon - at ~ 96% accuracy and found that these events behave most often as short, discrete pulses of ~1-2 weeks rather than the single sustained seasonal plateau. We also found the spring window to be far longer than the autumn one - by roughly seventeen to one - and the more carbon-relevant of the two, which changes how the cold-season carbon flux should be represented. The direction we are most engaged in now is a quantum-enhanced version, which uses quantum methods to map the relationships among the governing variables more completely and to quantify the uncertainty in the result. That uncertainty determines whether the numbers can be trusted in a projection.
Multi-blizzard event in February; sledding in Maryland (2026).
If you were to expand your current research focus, what new topic(s) would you explore?
I am already expanding in three directions.
- The first is drylands: through a NASA partnership with a working cattle ranch, we use imaging spectroscopy - fine-resolution spectral measurements of reflected sunlight - to map plant species and rangeland health conditions in support of regenerative grazing, a deliberate move from the frozen ground of the Arctic to the semi-arid rangelands of the American West, on land that people actively manage.
- The second is the global carbon budget: through a reconciliation effort (RECCAP3+) and a quantum-enhanced model, CarbonQAI, we are working to close the gap between two independent ways of accounting for carbon – the ‘bottom-up’ account built from ground measurements and models, and the ‘top-down’ account inferred from carbon concentrations in the atmosphere
- The third is anticipation: using quantum computing and causal methods to detect when a system is losing resilience and approaching a threshold before it crosses it, with applications extending into agriculture and food security.
The constant across all three is carrying a method from one biome to another and asking whether the system's memory can be read early enough to be useful.
Pit stop on the commute from Fairbanks to Deadhorse. Finger Rock, Alaska (2023).
What is one of your favorite moments in your career so far?
Returning to NASA GSFC in 2025, to the laboratory where I had earlier interned. The draw was not so much the position; it was honestly the reunion - coming back to familiar faces and working with friends and colleagues again, in a group whose work I already knew well. The two years before that, at Jet Propulsion Laboratory, pushed my creative limits in an applied direction and shaped how I work now. I would not change the path that led here, including the parts that looked like departures from science; they converged, and they put me where the work I want to do actually is.
What keeps you inspired by your work?
Two things. The first is that every result exposes a relationship I did not know to look for; that steady uncovering is what holds my attention. The second is that the work is used - the permafrost results inform mission planning and international carbon assessments, and the rangeland work informs decisions made on actual land by the people who manage it. That the science leaves the page and changes what someone does matters to me as much as the result itself. The collaborations sustain the rest; the work is shared, start to finish.
Ventura sunset, required respite from AMPAC travel in Scandinavia (2024).
Tell us about a unique or interesting component of your work-life balance.
A more recent one: I have been learning to DJ – trance and deep house music, specifically. It is a developing skill rather than a long-standing one, and what draws me to it is structural in origin. Mixing is an exercise in discipline and repetition - matching tempo and key, holding a phrase, and timing the release of a drop so the energy lands where you intend. It asks for a narrow, sustained kind of attention, and an hour of it clears and focuses my mind in a way little else does: the same concentration I bring to a problem at work, but with an immediate, albeit sometimes frustrating, audible result. I find the constraints of the form clarifying, and I return to the science sharper with it.
Working in my studio. DC Arts Studios, Takoma Park, DC (2017).
Published Date: .
Hometown:
Omaha, Nebraska, USA
Undergraduate Degree:
B.S. Biology, University of Nebraska, Lincoln, NE
Post-graduate Degrees:
M.S. Environmental Sciences and Policy, Johns Hopkins University, Baltimore, MDPhD Earth Systems and Geoinformation Sciences, George Mason University, Fairfax, VA