Global climate models represent our coastal cities as oceans, while regional models fail to represent their true complexity, posing a barrier to future planning and resilience, writes UNSW researcher A/Prof. Negin Nazarian.
In Australia, 87% of our population lives along the coast, most in cities that drive much of Australia's economic activity.
Understanding the impact of a rapidly warming climate and changing weather on cities is of national importance – and yet, the national datasets and models available to researchers are not fit for purpose.
Without a robust understanding of climate risks tailored to our urban settings, we risk compromising the safety, health, and well-being of the millions of Australians living in towns and cities.
The central problem is this: well-established and highly effective global climate models were not built to capture the specific conditions within urban environments. These models were designed and tested for global circulation patterns. When zooming in on cities, they lose accuracy and precision.
For example, the global grids that should represent Australian cities often categorise them entirely as oceans. Even inland cities (and megacities) are ignored, showing that this is not just a resolution or coastal issue.
Regional models with higher resolutions offer some improvements, but they still rely on oversimplified representations, commonly modelling cities as concrete blocks or rocks.
This is currently what weather forecasts do – entire cities like Sydney, Melbourne, or Brisbane are treated as uniform slabs of concrete, failing to capture the complex physics and diverse landscapes of urban spaces.
Moreover, the physics used may be too simplified to capture the interactions between complex urban surfaces, heat, water, and atmospheric flows. Key urban factors, such as city layout, construction materials and their embodied carbon, and human activities like transport, are still largely simplified in regional climate models.
This gap in our modelling capabilities has real-world consequences. How can we develop effective strategies to ensure our developments are resilient to future climates, and assess the true impact of major infrastructure projects (such as greenfield developments or large-scale precincts to tackle housing shortages), if we don't have models that accurately represent our evolving urban areas?
Our national models, even those capable of achieving a relatively accurate kilometre-scale resolutions, are unable to inform or assess the potential effect of long-term urban greening policies, like those which aim to reduce urban heat by 2050.
If we are to advance, we need national climate models and datasets that understand Australian cities as they truly are, not as abstractions.
Urban climate & weather
To understand why this is a problem, we first need to appreciate just how urban areas significantly impact their environment, along with the ways in which the climate affects urban areas.
One well-known example is the urban heat island effect, where cities become notably warmer than their rural surroundings due to factors like concrete and asphalt absorbing heat.
It's important to acknowledge that cities interact with weather and climate in ways that extend beyond heat.
Cities can also influence local weather patterns. Buildings in urban areas alter wind flows, creating updrafts that can affect cloud formation and sometimes alter rainfall patterns, especially downwind of urban areas. In some cases, cities can even experience higher frequencies of hail.
Urbanisation also increases the intensity of surface runoff, as impermeable surfaces like roads and buildings prevent rainwater from seeping into the ground, often leading to a higher risk of flooding.
Without accounting for this it's impossible to assess policies aimed at countering the effects of climate change.
Recognising the complexity of these interactions is crucial if we are to develop strategies that truly make our cities resilient to the challenges of the 21st century.
We can't prepare for the challenges cities face unless we understand the problems. Accurate, consistent information is essential for crafting effective solutions and to reduce negative impacts.
Mitigation and adaptation strategies need good data, too
Cities, as major contributors to greenhouse gas emissions, can adopt various net-zero initiatives to help mitigate the production of greenhouse gas.
But beyond slowing the curve of climate change through mitigation, cities also need adaptation strategies. These are measures that ensure cities and their residents can withstand and bounce back from damaging climate impacts.
In practice, these challenges translate into significant responsibilities entrusted to local councils, private businesses, and property owners and developers, who are increasingly expected to make decisions on their investment and livelihood or disclose climate risks for their business.
Yet, these risk assessments often rely on a patchwork of methods and models that are not nationally consistent, many of which lack the resolution needed to accurately represent urban environments. This makes it difficult for stakeholders to truly understand their risks.
How can businesses, for example, accurately disclose their climate vulnerabilities when the available tools fail to capture the specific threats faced by cities as different as Sydney and Wagga Wagga?
Where to now?
To effectively address these challenges, Australia needs to develop the capability to produce nationally consistent projections of weather and climate risks that consider the unique characteristics of our cities.
Importantly, we should not place the burden of figuring out the best approach on individual councils, property developers, or business owners. Instead, this responsibility should be shared across national and state governments and supported by established entities for our National Research Infrastructure (NRI) through the National Collaborative Research Infrastructure Strategy (NCRIS), working collaboratively with experts to create standardised methodologies and resources. The NRI 2026 roadmap, currently in development, presents a critical opportunity to create nationally consistent insight, and bridge the gap between urban data and climate modelling, two areas that remain largely disconnected.
The Insurance Council of Australia, in its Accelerating Climate Action report (2024), calls for building a national picture of climate risk as "unchecked climate change could shrink Australia's GDP by 6 per cent by 2070". A national problem demands national leadership and a shared solution.
This involves enhancing our models to accurately reflect urban environments, ensuring they account for the intricate interactions between urban structures and weather patterns.
We want to avoid making decisions based on incomplete or inaccurate information, as this can lead to misguided policies and planning that ultimately diminish the liveability of our cities.
Associate Professor Negin Nazarian is an urban climatologist at UNSW's School of Built Environment, Chief Investigator at the ARC Centre of Excellence for 21st Century Weather, and Fellow at the City Futures Research Centre. She leads the Climate-Resilient Cities (CRC) research lab, a multidisciplinary group focused on pressing challenges of urban climate.
Key Facts:
