Extreme storms are becoming stronger, more frequent and more disruptive, placing increasing pressure on the places people live, work and travel every day. Across transport networks, housing, energy systems and coastal communities, the impacts of severe weather are no longer distant or abstract; they are being felt on the ground [1]. For those responsible for planning, preparedness and long-term resilience, the challenge is more than understanding these risks but ensuring that the wealth of climate and hazard data available is able to meaningfully inform real decisions that protect communities and infrastructure.
Why all this data still not reach real planning decisions
Over the past decade, the amount of climate and hazard data available to planners and policymakers has grown rapidly. Advances in satellite monitoring, forecasting models and historical datasets mean that storms are now tracked in a very good detail. Yet, despite this abundance of information, much of it struggles to influence the everyday decisions that shape how places are planned, built and managed.
One reason is that climate data is often produced at scales or in formats that do not align easily with planning systems. National risk assessments, regional emergency planning and local development control frequently operate with different datasets, different time horizons and different priorities. Then, valuable hazard information can remain siloed, technically sophisticated, but disconnected from the points where decisions about land use, infrastructure and investment are actually made.
There is also a gap between how data is generated and how it is used. Many planning decisions rely on clear categories, thresholds and standards that can be embedded into policy and guidance. By contrast, storm data is often presented as complex, continuous or rapidly changing information, making it difficult to translate into rules, regulations or long-term strategies.
All together, these mean that climate intelligence does not always travel effectively across planning scales. So, bridging this gap requires approaches that not just analyse storm behaviour but reframe hazard information in ways that are usable, consistent and meaningful for real-world planning decisions.
Turning cyclone data into planning insight
This project was carried out during my PhD research in Urban Planning at the University of Dundee, with the aim of exploring how storm data could be made more useful for real planning and resilience decisions. Rather than starting from forecasting or prediction, I focused on classification, asking whether storms could be grouped in consistent, meaningful ways that align with how planners and emergency services already think about risk.
I worked with a large historical cyclone dataset compiled by the National Oceanic and Atmospheric Administration (NOAA), which contains detailed records of storm behaviour over more than a century [2]. Using this data, I tested different machine-learning approaches to see how reliably cyclone events could be classified based on key characteristics such as wind speed, pressure and storm structure.
Throughout the project, the emphasis was on understanding what kinds of outputs are interpretable, repeatable and usable across planning systems. By comparing different modelling approaches and examining where they performed well, and where they struggled, the project focused on how data-driven analysis can support clearer, more consistent decision-making across national, regional and local planning contexts.
This line of thinking led to the development of SmartScale, a planning-focused approach designed to help storm information travel more effectively across national, regional and local decision-making. Rather than producing more data, SmartScale focuses on how hazard information is structured, translated and shared, so that it can support clearer, more consistent planning and resilience decisions.
What storm data can (and can’t) tell us
Working with historical cyclone data showed both what data-driven tools do well and where their limits lie. For the most common types of storms, clear and consistent patterns emerged. Tropical Depressions, Tropical Storms and Hurricanes could be distinguished reliably based on characteristics such as wind speed, pressure and overall structure. This is important because these categories are already the language used by planners, emergency responders and infrastructure managers. When storm information is organised in ways that are familiar and stable, it becomes much easier to build into planning guidance, emergency procedures and regulatory standards.
The analysis also highlighted some challenges. Less frequent or transitional storm types were harder to classify consistently, reflecting both their rarity in historical records and the way storms change as they develop. Rather than undermining the analysis, this reinforced the importance of using data-driven tools carefully and transparently in planning contexts.
So, the real value of climate intelligence lies not just in technical accuracy, but in clarity. Tools that balance scientific robustness with interpretability are far more likely to support joined-up, real-world decision-making across planning systems.
Satellite image of a major cyclone system over the ocean. Source: NASA, public domain
Why resilience is built in everyday places
Resilience is not built in policy documents alone. It is built in the everyday places where people live, work and travel, in homes, streets, transport networks, coastal communities and essential services. Planning decisions shape how exposed these places are to extreme weather, and how quickly they can recover when disruption occurs. Evidence from Scotland’s Climate Change Plan highlights that climate impacts do not affect all communities equally, with socio-economically disadvantaged groups often facing greater exposure and fewer resources to recover [3].
When hazard information flows smoothly through planning systems, it supports safer buildings, better-located infrastructure and more effective emergency response. It can influence where development takes place, how critical services are protected, and how communities are prepared for future risks. So, climate data becomes more than information; it becomes a form of protection.
For communities, this can mean reduced disruption, clearer communication during emergencies and greater confidence that long-term planning decisions reflect real environmental risks. Strengthening the link between climate intelligence and planning practice is central to creating places that are safer, more resilient and better able to adapt to an uncertain climate future.
Firth of Clyde: the estuary of the River Clyde on Scotland’s west coast, where researchers, communities, and stakeholders are working together to co-develop local governance frameworks for coastal resilience and sustainability. Source: TRACC (TRansformative Actions for resilient Coastal Communities) project, 2024.
What surprised me along the way
One of the most surprising aspects of this project was how often the challenge was not a lack of data, but a lack of connection. In many cases, the information needed to understand storm risk already exists, yet it remains fragmented across institutions, scales and systems.
Developing SmartScale reinforced for me that resilience is as much a question of governance and coordination as it is of science or technology. The challenge was rarely about producing better data, but about ensuring that information could move between institutions, scales and decision-making contexts in ways that planners could use. Even the most robust analysis can struggle to make an impact if it cannot be aligned with the ways planning decisions are made.
This realisation has shaped how I now think about the role of artificial intelligence (AI) in planning. Rather than offering solutions on its own, AI has the greatest value when it helps bridge gaps between datasets, decision-makers and scales, and supports more coherent, joined-up approaches to resilience.
Looking ahead: planning resilience with Digital Twins
Digital Twins, virtual representations of cities, regions and infrastructure systems, are increasingly being explored as tools to support planning and resilience currently. Their value lies in their ability to bring together data, decisions and responsibilities across different planning scales.
When hazard classification and climate intelligence are embedded within these systems, Digital Twins can help planners and policymakers explore “what if” scenarios before extreme events occur. This includes stress-testing infrastructure, understanding how risks cascade across systems, and assessing how local decisions align with regional and national resilience objectives.
This perspective also connects with my wider research on scalability in smart spatial planning, where I have examined how digital tools can support coherence between local, regional and national decision-making [4].
Used carefully, Digital Twins offer an opportunity to shift resilience thinking upstream, from reacting to disruption toward anticipating it. In doing so, they can support more transparent, coordinated and forward-looking approaches to climate resilience.
Bringing storm data and planning decisions together
SmartScale brings together the insights from this research into a single, planning-focused approach that connects storm data with real decision-making across scales. Rather than treating hazard analysis as a technical exercise in isolation, it reframes classification as a shared language that can travel between national policy, regional coordination and local planning practice.
By focusing on clarity, consistency and usability, SmartScale highlights how climate intelligence can be structured in ways that align with existing planning systems. It shows how historical storm data, when carefully analysed and translated, can support more coherent discussions about risk, preparedness and long-term adaptation.
Ultimately, SmartScale is less about introducing a new tool and more about strengthening connections between data and policy, between institutions, and between the different levels at which resilience is built. In a context of growing climate uncertainty, approaches that support this kind of joined-up thinking will be essential for creating places that are safer, more resilient and better equipped to adapt over time.
Paper (open access): https://doi.org/10.20933/100001469
Code repository: GitHub
Maryam Forghaniallahabadi
PhD Candidate in Town and Regional Planning
College of Art and Design (DJCAD), University of Dundee
Scotland, United Kingdom
Discovery, University of Dundee Research Portal
References:
[1] Scottish Government (2025). Scotland’s Draft Climate Change Plan: 2026-2040 – Strategic Environmental Assessment Environmental Report. Available at: https://www.gov.scot/publications/scotlands-draft-climate-change-plan-2026-2040-strategic-environmental-assessment-environmental-report/.
[2] National Oceanic and Atmospheric Administration (NOAA).
HURDAT2 Atlantic Hurricane Database. Available at:
https://www.nhc.noaa.gov/data/
[3] Scottish Government (2025). Scotland’s Climate Change Plan 2026–2040: Equality Impact Assessment. Available at:
https://www.gov.scot/publications/scotlands-climate-change-plan-2026-2040-eqia/pages/4
[4] Forghaniallahabadi, M. (2024). A Framework for Scalability and Intelligent Integration in Multi-Level Smart Spatial Planning. Poster session presented at Urban Transitions 2024, Barcelona, Spain. Available at:https://doi.org/10.20933/100001474
