In Progress: Adding Built Environments to Climate Models

Three maps, each showing a quality (e.g. vegetation, building coverage) over latitude and longitude

Maps of Sydney according to various descriptors | Image supplied.

With data made available through AURIN, researchers at the University of New South Wales (UNSW) are working to incorporate more detailed features of the built environment into existing climate models. This will provide a basis for a better understanding of how urban areas both impact and are impacted by climate change.

Having more accurate models of urban areas will assist in developing appropriate mitigation and adaptation measures to support better and healthier cities and regions.


Climate change is increasingly impacting cities and urban areas, but these regions are often overlooked in modelling as it is difficult to ensure that they are accurately represented.

A key reason for this is the question of how cities should be described. For example, the physical and environmental features of a city like Tokyo are very different even to a city like Sydney, including factors such as green space and building density. This means that built environments require descriptions at a finer detail to ensure they are accurately understood.

For modelling purposes, simple descriptors are required, such as the proportion of hard surfaces to soil or plant surfaces, the average height of buildings, average street widths, and so on.

Clean, high-resolution data on these factors can be difficult to find, and therefore difficult to incorporate into modelling of climate systems and impacts. This means that our knowledge of urban areas in this context is currently limited, despite the importance of understanding how cities impact, and are impacted by, the climate.


Using Geoscape Buildings data accessed through AURIN’s Restricted Data program, Dr Mathew Lipson and his UNSW colleagues at the ARC Centre of Excellence for Climate Extremes, Dr Melissa Hart and Dr Negin Nazarian, are creating a new set of gridded data that can be fed into their models.

This is possible due to the scale of the data, which provides greater detail on urban characteristics than other data sets and allows areas within cities to be assigned more accurate descriptors.

“It’s such high-resolution data. From the data that you’ve provided us, we can calculate parameters describing the city and feed it straight into our climate models. This is rare data that isn’t available in a lot of places – often we’re trying to estimate parameters using satellite data or people going out and doing surveys. So it’s really, really valuable in that sense.”

Importantly, this work is quantifiable, replicable and scalable. For example, the dataset can be used to describe local characteristics around an individual school (for the Schools Weather and Air Quality project), or produce maps of neighbourhood characteristics across the whole city and beyond for regional climate modelling.

“It’s the scale that’s making the difference. Two things make this data set very valuable: it’s very high resolution, and it’s standard across many areas. People can do this work piecemeal, neighbourhood to neighbourhood, but to have it standardised across Australia means we can have more confidence in how we use the data.”

To date, Dr Lipson’s project has created a full set of gridded data for greater Sydney. The next step is to feed this data into their climate models.

Map showing mean building height over latitude and longitude
Map of Sydney according to building height | Image supplied.


By better incorporating cities’ characteristics into climate models, it will be possible to understand and simulate climate impacts and possible mitigation measures against problems such as heat island effect.

Dr Lipson notes that cities are in a really good position to mitigate some of the impacts of climate change. As they are built environments, we are able to adapt them relatively easily compared to other areas. For example, we can increase the number of trees or canopy cover, or take measures such as installing white or cool roofs.

In quantifying the effectiveness of these measures in different environments, decisions over resource allocation for mitigation can be made more effectively. It will also be possible to model the impact of new developments on other areas, for example, how new structures in one part of a city will affect the local area and have flow on effects to other regions.

Ultimately this project will assist our understanding of the impacts of climate change, as well as provide an evidence base to make decisions in pursuit of better, healthier cities.