Skip to main content Skip to secondary navigation
Main content start

From Economic Models to Wildfire Mastery: Iván Higuera-Mendieta’s Data Science Journey

Beginnings in Economics

Iván’s journey starts in Bogotá, Colombia, where he trained as an economist pursuing a BA degree in Economics from Universidad de los Andes. “As soon as I graduated, I went to the place where all economists want to work, which is Banco de la República, Colombia’s central bank,” admits Iván. At the central bank, Iván focused on research, which he considers the most data-heavy side of economics. One of his key areas of interest was First Nations. Colombia has a significant portion of protected territory, including natural parks and indigenous reserves. At the time, there was a lot of debate about the effectiveness of these protected areas in terms of biodiversity conservation, which led his team to explore this question using big data. This was Iván’s first paper and his initiation into answering big and important questions using data. It was a big challenge for Iván but also an exciting opportunity to learn new programming languages and remote sensing techniques.

From Colombia’s Central Bank, Iván then becomes a Data Scientist at the Data Science and Public Policy Lab, now at Carnegie Mellon, and then a pre-doctoral fellow at the Energy Policy Institute at the University of Chicago, working at the Climate Impact Lab.

Fast-forwarding to Stanford

Iván is now a PhD student at the Doerr School of Sustainability and a scholar at Stanford Data Science. His research employs causal inference and computer vision to understand and measure the effects of environmental degradation, focusing on wildfires and climate change and how they impact vegetation and people. California has a great system for monitoring wildfires, and Iván’s work involves building faster models to predict the spread and impact of fires. Iván’s advisor, Marshall Burke, was particularly interested in wildfires, which led Iván to share his passion for understanding these phenomena. “We still have a lot of work to do,” says Iván, “including working with practitioners to refine our models.”

Iván’s Favorite Project: Addressing Wildfires in California

One of Iván’s most compelling projects tackles a pressing policy question: the role of prescribed burning in reducing massive wildfires in California and the Western United States. Prescribed burns, or fuel treatments, involve intentionally setting controlled fires to eliminate underbrush and reduce the risk of catastrophic wildfires. However, these practices are controversial due to their immediate negative impacts, such as emissions that harm human health.

Iván’s research delves into this trade-off. By using simulations and causal inference methods, his team evaluates the short-term costs of prescribed burning against the long-term benefits of reducing large-scale wildfires. This work is particularly timely given California’s ambitious goal of treating one million acres annually to prevent wildfires. Initially skeptical of this target, Iván’s data-driven approach revealed its necessity and feasibility.

Through this project, Iván’s team examines the emissions saved and the overall utility gained from these practices. The findings not only inform state policies but also provide a clearer understanding of how to balance immediate costs with future benefits. This work is vital as wildfires grow more intense and frequent due to climate change, highlighting the need for proactive measures like fuel treatments.

Transforming Sustainability with Data Science & Causal Inference

“Data science has revolutionized how we understand and tackle sustainability challenges. With advanced tools and techniques, researchers and policymakers can now better comprehend natural phenomena, model complex systems, and predict future environmental changes.” 

Iván believes that this transformation is especially impactful in assessing the consequences of environmental damage on human systems and the broader ecosystem. By leveraging data science, we gain insights into the future trajectory of our environment, including changes in atmospheric conditions and global temperatures. These insights enable us to adapt and address pressing issues effectively. “Without data science, many of these advancements would be unattainable,” concludes Iván.

A cornerstone of data science in sustainability is causal inference, which helps us identify the causes and effects of changes in the ecosystem. Understanding causality allows us to develop adaptation policies that address these changes effectively. These methodologies form the backbone of efforts to mitigate and adapt to climate-related challenges.

Stanford Data Science’s Impact

Iván attributes much of his academic growth to his exposure to Stanford’s data science community. Engaging with diverse fields has enriched his perspective, enabling high-level discussions on shared challenges, such as measurement errors in data. A key realization was the non-classical biases often present in datasets—a departure from the idealized assumptions taught in theoretical statistics classes.

This shift in thinking was influenced by collaborations and exposure to cutting-edge research at Stanford, including work on understanding and correcting biases in data models. Iván highlights the profound impact of a specific researcher’s work on his research trajectory, underscoring the value of interdisciplinary collaboration and access to advanced tools. Without Stanford’s data science ecosystem, Iván believes he wouldn’t have developed such a deep understanding of these critical issues.

Life Lessons: Advice to a Younger Self

Reflecting on his journey, Iván emphasizes the importance of staying relaxed and embracing uncertainty. As an undergraduate, he often felt overwhelmed by questions about the relevance of his studies and his future career prospects. Looking back, he realizes many of these worries were unnecessary.

Iván’s path to academia wasn’t linear; he worked in various jobs before returning to pursue a PhD. Initially, this non-linear journey felt like a disadvantage, especially when comparing himself to peers who transitioned directly from undergraduate studies to doctoral programs. However, he now sees the value of those experiences, which shaped his approach to research and deepened his understanding of real-world problems.

His advice to his younger self—and to others—is to fixate less on the past and to embrace the unique trajectory of one’s journey. For those in data science, curiosity often leads to unexpected paths, making it nearly impossible to follow a strictly linear course. Ultimately, Iván’s experiences outside academia enriched his work, proving that diverse experiences can be an asset rather than a hindrance.

Stay tuned for a soon-to-be-published paper!

Recent Publications & Useful Links

Higuera-Mendieta, I, Wen, J., & Burke, M. (2023). A table is worth a thousand pictures: Multi-modal contrastive learning in house burning classification in wildfire events. NeurIPS 2023 Computational Sustainability: Promises and Pitfalls from Theory to Deployment.

[video] Data Science Across Domains: An Example with Stanford Data Science Scholars—John Chambers, Emily Gordon, Thomas Teisberg, Iván Higuera-Mendieta, Stanford Data Science Conference (May 7, 2024)

Iván's Website

[save the date] Meet Iván at the upcoming Sustainability Data Science Conference on April 10, 2025