Getting hot, hot, hot – climate modelling explained

By now, you might be familiar with the major insights from the IPCC’s Sixth Assessment Report, Climate Change 2021: The Physical Science Basis, collating developments in climate science over the last 8 years. 

It paints a clearer, albeit more terrifying, picture of how the earth’s atmosphere, oceans and land have changed since pre-industrial times. 

The world’s leading climate scientists, representing every region on the planet, have then simulated five different scenarios up to the end of this century. 

The 2015 Paris agreement set the target for avoiding catastrophic climate change and ecological disaster at below 2 degrees, striving for 1.5 degrees. Six years later and the prospects have significantly diminished. 

“We are likely going to cross that 1.5 degree threshold even in the most optimistic of these scenarios,” the Deep South Challenge’s Earth System Modelling and Prediction programme lead, Dr Olaf Morgenstern says.

The IPCC report is clear that “rapid, deep and sustained cuts to emissions” are needed to keep the global temperature anywhere near these targets. It’s only in “the most optimistic scenario” Morgenstern says, where there is cooling after 2050 by removing more CO2 from the atmosphere than we add, that global warming could reach less than 1.5 degrees by the end of this century. 

Whether we’ll have the technological capability to remove CO2 from the atmosphere, or whether polluters are prepared to change their ways, will be the focus of Mitigation of Climate Change, set to be published in 2022. It’s clear the current targets put forward by different nations won’t be enough to meet the 2 degree target. The upcoming “Conference of the Parties”, scheduled for November in Glasgow, will reveal whether countries are prepared to rise to the challenge.

This report focuses on the physical science of climate change – a complex task using some of the largest supercomputers in the world to generate simulations of our future climate. The amount of data produced for the report is much larger than what was generated for the previous IPCC report (5th Assessment), Morgenstern says, who was one of 13 lead authors of Chapter 3, Human Influence on the Climate System.

Questions have been raised about whether the latest climate models are “running hot”. A standard experiment in climate modelling involves doubling the amount of CO2 in the atmosphere and waiting until the Earth system has settled into a new balance, or “equilibrium”. Calculating the average warming since the pre-industrial age is then the “equilibrium climate sensitivity” of the model. This single number is a simple way to characterise climate models.

It takes hundreds to thousands of years for all temperature-influencing processes to come to balance, E.G for atmospheric warming to permeate into the depths of the ocean, so the climate sensitivity is calculated only once the system has settled into a new balance or “equilibrium”.

It is true that several of the most recent climate models are “running hotter” than others before them. But the IPCC assessment doesn’t rely solely on these models for its assessment of climate sensitivity. Rather, other insights are used, such as understanding the average global surface temperature during past periods of the Earth’s history. 

By combining these insights, the assessed climate sensitivity uncertainty has in fact halved. The best estimate of climate sensitivity, calculated in the report, is 3 degrees; this is unchanged from previous IPCC reports. However, scientists have reduced the error estimates around this number. For the first time, this improved understanding of climate sensitivity is used to assess end-of-century warming. 

Scientists use an amalgamation of climate models to measure the changes in the atmosphere, land and ocean. For large-scale climate features – temperature, precipitation, radiation – the report finds climate models today perform better than those of previous generations. But while there have been “incremental improvements” in climate modelling, there are still areas where more research is needed. 

Climate models don’t accurately capture what is happening here in Aotearoa. Our maritime climate is largely influenced by what’s happening in the surrounding ocean. The clean air over the Southern Ocean is creating an “error” in simulations as scientists seek to understand how aerosols interact with the clouds.  

Then there’s the small observed increase in Antarctic sea ice extent over several decades which in recent years has given way to an abrupt decrease. This complex behaviour isn’t generally captured in climate models. Understanding these unique features of our Pacific climate is at the heart of New Zealand Earth System Model (NZESM), a model being developed by Dr Olaf Morgenstern and his team. This is being developed in collaboration with the UK MetOffice. Without their support, the Deep South Challenge wouldn’t be able to undertake such a mammoth task.