Intelligent Simulation for Extreme Climate Disasters: From Readiness to Response

Climate disasters are intensifying in scale and frequency, yet communities remain dangerously unprepared, and research on these events is fragmented along disciplinary lines. Atmospheric science can now reproduce the pyro-convective dynamics of megafires with event-scale coupled fire-atmosphere models. Machine learning offers transformer and convolutional models trained on simulator outputs through simulation-to-reality (Sim2Real) pipelines that can generate fire scenarios in near real time, although scarce high-resolution observational data still limits their validation. Immersive visualization research shows that high-fidelity simulation strengthens situation awareness and decision-making for frontline responders and at-risk residents. Aesthetic research reveals that climate risk suffers from a 'perception problem', and that visceral, experiential scenarios can bridge the gap between knowing and acting. Yet these fields remain largely disconnected. Dominant models prioritize abstract prediction over experiential modelling and falter under the uncertainty of the low-probability, high-impact events that define the climate emergency. No framework yet integrates immersive aesthetic visualization, machine learning and fire-atmosphere science across the full arc of preparedness: from anticipatory risk perception and scenario rehearsal to collaborative decision-making and post-event review.Physics-based and empirical simulators provide scientifically grounded 'anchor' scenarios, and deep learning models trained on them can rapidly generate diverse, unforeseen scenarios as users change wind, temperature, fuel or ignition. This Research Topic aims to establish immersive aesthetic visualization, coupled with machine learning and fire-atmosphere modelling, as a way of making uncertainty perceptible rather than resolving it: 'plausibility engines' in which researchers, responders and residents can safely inhabit extreme scenarios, rehearse and review decisions, and build situation awareness. It also considers how these tools can help communities and institutions adapt to changing fire regimes by strengthening preparedness, building adaptive capacity and informing long-term planning. Finally, it seeks rigorous evidence on the limits of these systems, including validation against observed fires, the boundary between training and operational use, and ethical, trauma-informed design.To gather further insights into intelligent simulation for climate disaster preparedness and response, we welcome articles addressing, but not limited to, the following themes:• Aesthetic and affective approaches that render climate risk and uncertainty perceptible• Plausibility, storyline and scenario modelling for low-probability, high-impact events• Situation awareness, decision-making training and post-event decision review• Risk communication with at-risk communities, including culturally linguistically diverse communities at the rural–urban interface• Event-scale coupled fire-atmosphere and pyro-convection modelling for early warning• Machine learning for 2D and 3D fire behavior (spread, crowning, flame height, smoke), including Sim2Real training and hybrid physics–AI pipelines• Human–AI interaction and trustworthy AI that keeps humans in the decision loop• Immersive and AI-supported tools for community preparedness and adaptation planning• Validation against observational data, uncertainty quantification and efficiency for real-time interaction• Learning from past extreme events to inform adaptation strategies and resilience• Ethical, trauma-informed designWe welcome Original Research, Review, Perspective, Methods and Case Report articles from AI, art, design and visualization researchers, atmospheric and fire scientists, emergency responders and social scientists. Interdisciplinary work and real-world deployments with first-responder or community stakeholders are particularly encouraged. Topic editors: Dennis Del Favero, Theodore Giannaros, Yang Song Source: https://www.frontiersin.org/research-topics/86634/intelligent-simulation-for-extreme-climate-disasters-from-readiness-to-response