From Classics to Carbon: Brian’s Path to Impact
Brian Rogers’ journey into data science was anything but conventional. He started as a classics major before exploring multiple disciplines and eventually landing in geological engineering at the Missouri University of Science and Technology. What drew him in was the environmental aspect, particularly groundwater remediation and the challenges of contaminant monitoring.
It was during his work with groundwater models that Brian had his “aha” moment—these models were not only powerful but also highly generalizable. He realized that data-driven modeling could help answer a wide range of scientific questions. By the time he reached his junior and senior years, he knew that this was the field he wanted to pursue.
Diving into Data Science at Berkeley Lab
At the tail end of his undergraduate studies, Brian worked at Lawrence Berkeley National Lab, where he was introduced to reactive transport modeling—a method that combines hydrological and geochemical modeling to study soil systems. This experience shaped his research focus and led him to Stanford, where he found an advisor connected to his work at Berkeley.
When he applied for his PhD, Brian’s statement of purpose reflected a clear vision: he wanted to bring data science into the environmental and Earth sciences. Even now, as a PhD student, he sees a huge gap in the field where data science methodologies can make a significant impact.
Discovering Stanford Data Science
Brian first heard about Stanford Data Science through Lijing Wang and Kuai Fang, both of whom were involved in research groups he was familiar with. However, it was a course by Professor Jef Caers that truly solidified his interest. The data science methods he learned in that class are the very same techniques he applies in his research today.
A Shift in Research: Carbon Dioxide Removal
Midway through his third year, Brian made a bold decision—he switched his PhD topic to focus on carbon dioxide removal (CDR) technologies. He recognized a critical gap where data science could enhance the accuracy and efficiency of CDR models.
While some models existed in the field, Brian saw that they were often misused or not applied robustly. His goal became clear: integrating data science and modeling to improve carbon dioxide removal verification. Doing this effectively could reduce the costs of CDR technologies, making them more scalable and impactful.
Bridging the Gap Between Research & Industry
Although Brian’s project hasn’t been fully implemented yet, he has been actively working with standards development organizations—the groups responsible for creating verification standards for carbon credits. Through these collaborations, he realized that statistical modeling and data science gaps were bottlenecks preventing efficient carbon credit quantification.
To ensure his work is practical and useful, Brian has been in close contact with carbon dioxide removal project developers. These industry conversations have helped him refine his approach, ensuring that his tools and models answer real-world questions.
Enhanced Weathering: A Promising CDR Strategy
Brian’s current research focuses on a specific CDR technique called enhanced weathering. This method involves spreading crushed silicate rock, like basalt, over large agricultural fields. As the rock dissolves, it creates alkalinity, allowing the soil to absorb more carbon dioxide, effectively removing it from the atmosphere.
Enhanced weathering is estimated to have the potential to remove 1-5 gigatons of CO₂ per year, but this relies on optimistic modeling assumptions. However, there is concern that current models may overestimate CO₂ removal potential without proper validation.
Measuring CO₂ removal through enhanced weathering is incredibly complex due to soil variability, low signal detection, and uncertainty in existing models. While solid-phase methods may not be sufficient, new probabilistic and Bayesian frameworks could improve verification. Future work should focus on better sampling strategies, remote sensing, and integrating reactive transport models to improve accuracy.
Community-Powered Research and Impact
When you ask Brian how the Stanford Data Science Scholar Program is helping his research, his reaction is immediate. “Oh, it’s totally helping already.”
For Brian, joining Stanford Data Science (SDS) was an instant game-changer. He found himself surrounded by a consistent community of kind, smart, and genuine people—something he had hoped for but didn’t expect to find so quickly.
It's not hard to find kind and smart people at Stanford, but to find a consistent community where you're always around those people was the first thing that I knew I wanted out of Stanford Data Science, and I got it immediately.”
Beyond the social benefits, SDS has already provided technical support in ways that are accelerating his research. He’s connected with SDS scholar experts in optimization and statistical modeling, allowing him to refine sampling strategies crucial for his work.
“I don’t have time to take a whole optimization course. This is going to be so much more efficient and enjoyable.”
Long-Term Goals: Impacting the Carbon Industry
Looking ahead, Brian’s short-term goal is clear: by the time he finishes his PhD in a year and a half, he wants to integrate data science and modeling techniques into the carbon dioxide removal (CDR) industry. The biggest challenge he sees is quantification bottlenecks—ensuring that carbon removal processes are accurately measured and verified.
But beyond his PhD, his path is still open-ended. Whether he stays in carbon removal or applies his expertise elsewhere depends on where he’s needed most.
“There are cool opportunities to pursue philanthropic research, like developing tools for enhanced weathering or other open-system CDR methods.”
His skills also extend to broader geochemical and environmental challenges, including nuclear energy. In fact, the modeling codes he uses today were originally developed by the Department of Energy in the 1990s for nuclear waste management. With nuclear energy likely playing a growing role in clean energy transitions, Brian sees potential for his expertise to make an impact in this space as well.
“I’m just excited to see what’s needed the most and where this type of quantitative expertise can be the most impactful.”
Role Models: Inspiration from Jef Caers
One of Brian’s biggest inspirations is Jef Caers, whose class he took during his first year at Stanford. Beyond learning a variety of data science techniques, Brian became particularly fascinated with distance-based generalized sensitivity analysis—a method he now applies to everything he does.
But what truly makes Jef an inspiration is his personal journey.
“As a gay man from an extremely conservative background, I really relate to Jef’s story of moving to San Francisco and finally having the opportunity to be himself.”
Beyond their shared experiences, Brian admires the massive impact Jef’s work is having in critical mineral exploration—helping advance global clean energy solutions.
“Jef’s group is making an insanely massive impact on clean energy. That’s amazing.”
Another mentor Brian is excited to learn from is Chris Mentzel, SDS Executive Director, who has extensive experience in science and data science philanthropy—an area Brian hopes to explore further.
Advice for Future SDS Scholars
Brian’s biggest piece of advice for aspiring Stanford Data Science scholars? “Present your research questions with curiosity. Be upfront about pitfalls and challenges, and you’ll be pleasantly surprised by the ideas and support you receive.”
For Brian, this collaborative atmosphere is what makes SDS special. It’s not about pretending to have all the answers—it’s about engaging with the community and being open to new ideas.
Lessons for His Younger Self
If Brian could go back in time, he’d tell himself one thing: “Be careful with tunnel vision.”
“When I get excited about something, I get super focused. But sometimes, I need to stop and ask myself: Do I really enjoy this? Would it make sense to pivot?”
This realization came from personal experience. During his third year, Brian burnt out completely. He was forcing himself through a research path that didn’t truly excite him—until he finally crashed hard.
That’s what led to his switch to carbon dioxide removal. In hindsight, he wishes he had recognized the signs sooner and made the shift earlier.
“But I did it. I persevered, and it wasn’t too late.”
Beyond Research: A Passion for Art
Outside of his work, Brian finds creative expression through drawing.
“I draw all the time. It’s my favorite thing.”
While he’s modest about his artistic skills, he has pieces he’s proud of—and he’s open to sharing them.
For Brian, art is more than a hobby; it’s a way to relax and reflect.
And when he needs a moment of peace, he finds it in the Cantor Arts Center at Stanford—a place full of inspiration, much like his journey through data science.
Looking Ahead
Brian’s journey—from classics major to geological engineer to data-driven climate researcher—highlights the interdisciplinary nature of data science. By leveraging modeling techniques, he hopes to optimize and validate carbon removal technologies, ensuring they can play a meaningful role in the fight against climate change.
As he continues his work, one thing is clear: data science is reshaping environmental research, and Brian is at the forefront of this transformation.
Honors & Awards
- Stanford Data Science Scholar, Stanford University (2024-2026)
- Stanford Graduate Fellowship in Science and Engineering, Stanford University (2020-2025)
- Department of Energy Computational Science Graduate Fellowship, Department of Energy Krell Institute (2020-2024)
About CO₂ Removal
Carbon dioxide removal (CDR) refers to technologies and natural processes that extract CO₂ from the atmosphere to mitigate climate change. Key methods include:
- Direct Air Capture (DAC) – Uses chemical processes to absorb CO₂, which is then stored underground or used in industrial applications.
- Afforestation & Reforestation – Planting trees to absorb CO₂ through photosynthesis.
- Soil Carbon Sequestration – Enhancing soil's ability to store carbon through sustainable farming practices.
- Biochar – Converting biomass into a stable carbon-rich material that can be added to soil.
- Ocean-Based Approaches – Includes ocean fertilization to boost phytoplankton growth and enhance carbon uptake.
- Enhanced Weathering – Spreading minerals like basalt to chemically react with CO₂ and store it as solid carbonate.
While CDR is essential for reaching net-zero emissions, it must complement emission reductions rather than replace them. Cost, scalability, and potential ecological impacts remain key challenges.
