Three Core Pillars

Our research is built upon three core pillars—Complexity Science, Physically-Based Models, and Exponential Technologies—which together form the foundation of our pursuit toward Environmental Intelligence. Each pillar represents a distinct yet interconnected way of exploring, understanding, and unfolding the coupled human–natural systems that define our changing planet.

Complexity Science

Complexity science serves as the lens through which our laboratory perceives and interprets environmental systems. Every system that becomes the subject of our modeling is inherently complex, and by understanding and interpreting this complexity, we aim to realize environmental intelligence.

The idea that “everything flows” (Panta Rhei)—first proposed by the ancient Greek philosopher Heraclitus—captures the essence of humanity’s understanding of nature and existence. It denies the permanence of being and emphasizes that all things are in a continuous process of change. This principle of universal flux forms the philosophical premise of complexity, and complexity science, in turn, represents its scientific embodiment.

In this context, our lab has employed Sociohydrology as an illustrative example of understanding and interpreting complexity in coupled human–natural systems. Dr. Jeong’s perspective on this approach is presented in his article The Present and Possible Future of Sociohydrology.

Representative publications

Physically-Based Models

Physically-based models are the most essential research tools in our laboratory. They represent the intellectual culmination of centuries of curiosity from brilliant scientists and persistence from countless ordinary ones, enabling us to mathematically describe natural processes and to explore the causes and dynamics of environmental phenomena under explicit scientific assumptions.

Effective environmental management requires a clear understanding of both the causes of environmental problems and the transformations they induce in natural systems. This entails viewing the system of interest as a white box—a system whose internal mechanisms are transparent and interpretable. Thus, continuous efforts are needed to improve our scientific representation of natural processes. At the same time, the assumptions employed in physically-based modeling play a decisive role: when they are flawed, they may lead to entirely different narratives of change and alternative futures.

Our research group has primarily focused on the hydrosphere as the central domain of inquiry. Through this work, we have come to recognize that achieving a truly white-box understanding of the environment requires the explicit inclusion of human processes as endogenous variables, which poses a profound scientific challenge. Dr. Jeong’s perspective on this issue is elaborated in his article The SOHO Hydrological Modeling Framework.

“Everything flows—and everything is interconnected. Physically-based hydrological modeling, grounded in static conditions, remains valid and reliable in most cases. However, in an era dominated by uncertainty and complexity, such as the climate crisis, the true mission of hydrological modeling is not to predict a single future but to explore multiple possibilities and pathways, revealing even the ‘unknown unknowns’ that lie beyond our current perception. The SOHO (Self-Organizing, Holarchic, Open systems) framework represents a philosophical shift in hydrological modeling by integrating the deterministic Newtonian view of physical processes with the Darwinian view that emphasizes the evolutionary and adaptive nature of social systems. This synthesis enables a more holistic understanding of the complex interactions between human and water systems.”

Representative publications

Exponential Technologies

Dr. Jeong was initially deeply skeptical about artificial intelligence. Informed decision-making—an ideal that his research has long pursued through the white-box transparency of physically-based models—seemed fundamentally incompatible with the black-box nature of AI, which often lacks interpretability and explainability.

However, as he came to recognize the necessity of modeling humans to better understand environmental systems as complex systems, he experienced a personal realization. It appeared both ironic and inspiring that modern AI itself originated from an attempt to model human cognition. This realization led him to see the potential of machine learning as a tool not merely for prediction, but for representing and interpreting human and societal behaviors within environmental systems.

Our laboratory has since explored the use of artificial intelligence as a means of human social sensing—that is, interpreting unstructured data that reflect human activities and societal responses to environmental change. We are particularly interested in how such approaches can be integrated with physically-based models, bridging the physical and social dimensions of environmental systems. Dr. Jeong’s thoughts on this integration can be found in his article Digilog Water Management.

“Digital water management, supported by structured information from diverse devices, assists and sometimes even automates decision-making. Yet, the emergence of human wisdom—which cannot be learned by machines—and the awareness of unknown unknowns must remain human responsibilities. … ‘Digilog’ water management can be defined as a reflective process of rethinking the human role within the paradigm of digital water management. It represents the fusion of digital water management, oriented toward the present and the future, with analog water management, which embodies the continuity that connects the past, the present, and the future.”

Representative publications

While our work to date has primarily focused on artificial intelligence, our future research aims to extend beyond AI toward robotics and AI deployment, advancing innovations in environmental monitoring and management systems.

Current Main Projects

This section will be updated soon.