Progression of climate models

Global climate models help us understand how our planet's climate is changing, by simulating the Earth’s climate system. While they have improved dramatically over time, it is still hard to explain what is happening at the local level.

Here’s how they work.

We know the climate system is complex, and impacted by many different processes and interactions, from what is happening on the land to changes in the atmosphere, to ocean currents to human influences.

Each of these processes is complex too — here’s what a simplified process looks like for solar radiation.

Progression of climate models

Global Climate Models, or GCMs, simulate the Earth's climate system using mathematical calculations to understand how these many processes interact. They do this by dividing the globe into grids and layers.

Traditional global climate models have a resolution of 150km to 250km. Because of their size, these cells are often too big to examine the impacts of climate change at local scales – such as for regions or towns. This makes them unsuitable for informing policy at the regional scale.

For example, a model with a resolution of 150km would mean that 22,500sq km of the South West region of WA could be covered by just a single cell — a resolution far too coarse to enable decision making at the local scale. 

Even fine resolution models with 50km resolution would see just 10 cells to cover a very diverse region.

To fix this, a team of Australian researchers are using Setonix to “zoom in”, in a project that is part of the Climate Science Initiative being led by the WA Government Department of Water and Environmental Regulation (DWER) in partnership with Murdoch University and the NSW Government Department of Climate Change, Energy, the Environment and Water (DCCEEW) via the NSW and Australian Regional Climate Modelling Project (NARCliM 2.0) and the Pawsey Supercomputing Research Centre.

Using regional climate models (RCMs) and a process called dynamic downscaling, they are able to better resolve processes that matter locally, such as storms and other types of extreme weather events.

Because they are built on physical principles, these RCMs allow for changes in the existing relationship between weather variables and climate drivers. RCMs create a new time series, rather than adjusting an existing one.

The researchers are checking their model’s accuracy by comparing the model outputs against highly detailed observations from 1979 to 2014.

Then, they will look ahead up to the year 2100 to examine the impact on our climate under different scenarios.

The result will be the most detailed climate projections ever created for South West and North West WA, helping us to understand the local impacts of climate change..

It's a game-changer for climate adaptation in Western Australia, providing the tools we need to prepare for our changing climate.

Each of these processes is complex too — here’s what a simplified process looks like for solar radiation.

Case study

Global climate models help us understand how our planet's climate is changing, by simulating the Earth’s climate system. While they have improved dramatically over time, it is still hard to explain what is happening at the local level.

Here’s how they work.

We know the climate system is complex, and impacted by many different processes and interactions, from what is happening on the land to changes in the atmosphere, to ocean currents to human influences.

Making connections

By bringing together HPC infrastructure, skills and expertise, we can accelerate the work of researchers whose studies of the natural world have real-life applications.

It means we can support larger and more diverse scientific inquiries, and improve productivity and performance by providing the powerful compute and visualisations needed to bring meaning to data.

Thanks to Pawsey’s technology, logistical support, and expert collaboration, we are continuing to enable high-impact science, wherever on the globe it may be.

USE SLIDER TO VIEW THE CHANGE IN MODELS

The proportion of Australian science requiring high-performance computing through the National Computing Merit Application Scheme that is delivered by Pawsey