Navigating complexity: agent-based simulations for climate-resilient economies

Navigating complexity: agent-based simulations for climate-resilient economies

Taberna, A. (TU Delft Policy Analysis)

Filatova, T. (promotor)
Nikolic, I. (promotor)

Delft University of Technology

2024-03-08

Amid the Anthropocene, the escalating threat of flooding, driven by extreme rainfall and sea-level rise, challenges societies worldwide. In the last two decades, floods have impacted billions and inflicted colossal economic losses. Concurrently, the global trend towards urbanization predicts that by 2050, about 70\% of the global population will inhabit urban areas. This demographic trend, heavily influenced by agglomeration forces, further underscores the vulnerability of these urban centers, many of which are precariously situated in flood-prone areas. Given the confluence of escalating climate risks and the surge in populations settling in vulnerable zones, a pressing question emerges: How will rapidly urbanizing coastal societies adapt to intensifying flood risks in the face of escalating climate-induced shocks and changing regional economic landscapes?

To address this multifaceted issue, this dissertation delves into the complex nexus between climate shocks, regional economic dynamics, and societal responses. Central to this exploration is the creation of innovative simulation tools tailored to incorporate the autonomous adaptation strategies of various actors within a regional economic framework. This thesis stands at the forefront of a new wave of computational models that encompass risk and embed resilience into complex adaptive systems.

I commence by examining the current advancements and gaps in employing Agent-Based Models to unravel the dynamics of flood risk and adaptation assessments. In this exploration, I underscore the pivotal role of human actions in shaping risks and resilience within flood-prone urban settings.

Building on this foundation, I introduce the Climate-Economy Regional Agent-Based (CRAB) model. The CRAB model employs an evolutionary perspective to provide a comprehensive view of the balances struck between the driving forces of economic agglomeration and the counteracting pressures of climate hazards. It focuses on the decision-making of heterogeneous agents, representing households and firms, as they navigate the choice of relocation between safer inland regions and hazard-exposed coastal zones.

Venturing further, I enhance the CRAB model to embody autonomous household adaptation behaviors, drawing from empirical data. Here, I challenge the traditional reliance on rational agents in sustainability models, unveiling a notable adaptation deficit when juxtaposed against boundedly-rational choices gleaned from real-world surveys. This nuanced exploration uncovers how varied adaptive capacities can potentially accentuate inequality and impede resilience.

Subsequently, I include in the CRAB model a layered risk strategy that encompasses an array of climate change adaptation measures. This refined model, enriched by extensive behavioral and flood data, bridges existing gaps in the current understanding of feedback loops and cascading effects triggered by flood shocks within a socio-economic system of boundedly-rational agents.

In conclusion, this dissertation pioneers a unique trajectory in understanding societal responses to the specter of flooding, offering invaluable insights and frameworks for devising future climate-resilient strategies.

Agent-based models,
resilience
flood risk
agglomeration forces
survey
climate change adaptation
distributional impacts
path dependency

https://doi.org/10.4233/uuid:ca535619-5459-4345-adf4-8c768c6934f6

978-90-361-0736-5

Institutional Repository

doctoral thesis

© 2024 A. Taberna