IPCC AR6: Powerful climate modelling provides a glimpse into a bleak future... but could it also help us avoid it?

The latest IPCC report (AR6: The Physical Science Basis) makes for a sobering read, with global climate change shown to be intensifying and the risk of reaching crucial “tipping points” in the Earth’s climate system increasing. Perhaps on the brighter side, this report demonstrates improved climate modelling compared to the Fifth Assessment Report (AR5), allowing us a clearer view of the future.

While AR5 used the CMIP5 model ensemble to predict future warming, AR6 uses the CMIP6 model ensemble. This new suite of climate models demonstrates improved representation of physical processes, via the inclusion of new features, such as dynamic ice sheets, and through the refinement of existing features, such as sea ice representation and cloud representation. The CMIP6 ensemble also displays improved representation of ocean physics and marine biogeochemical processes, both of which are fundamental to the Earth’s climate system. It is important to note that an ensemble of models is used, rather than an individual model, as each model represents these complex Earth system processes slightly differently: considering outputs from a suite of models gives us greater confidence in the results. How the results from individual models are considered together has also changed from AR5. While previously a multi-model mean was produced from individual projections, the AR6 report weights the individual CMIP6 models according to how well their outputs align with observations (acting to reduce the impact of high sensitivity models on overall outputs). These constrained outputs are then combined with similar outputs from a simple energy balance model (or “emulator”) to create the “assessed” projections used throughout the report.

A key refinement resulting from this improved model ensemble is a decrease in the uncertainty surrounding climate sensitivity, or the amount of warming that can be expected from a doubling of pre-industrial CO₂ levels. While AR5 deemed a climate sensitivity range of 1^oC to 6^oC to be very likely (i.e. 90-100% chance) AR6 reduces this “very likely” range to 2-5^oC, despite individual CMIP6 models tending to exhibit higher sensitivity and warming values than CMIP5 models. The good news is we now have greater confidence that a climate sensitivity of above 5^oC is rather unlikely (despite some individual models of the CMIP6 ensemble predicting sensitivities far exceeding this). The bad news, though, is that the 1.5^oC warming goal of the Paris Agreement is slipping ever further out of reach.

Sea level rise (SLR) projections have also been revised in AR6, with predictions for 2100 “modestly higher” than their AR5 counterparts because of increased model predictions for the contribution of Antarctic ice sheet melt on SLR. In AR6, global SLR is projected to be between 0.38m (0.28-0.55m; likely range for SSP1-1.9) and 0.77m (0.63-1.01m; likely range for SSP5-8.5) by 2100. These projections of increased ice sheet melt are in line with recently observed contributions to SLR – increased ice sheet melt in recent years means the loss of land ice (glaciers plus ice sheets) is now the largest contributor to global mean SLR. This ice sheet melt is expected to worsen, with AR6 also highlighting the risk of abrupt and irreversible changes, including the onset of Marine Ice Cliff Instability (MICI) in Antarctica. Low confidence in this process means it is not included in most model projections but, when it is, SLR of up to 16m by 2300 is predicted.

In case there are still any sceptics out there hoping these changes are all part of the Earth’s natural cycles, improved climate modelling has also allowed AR6 to present higher confidence in the anthropogenic source of warming. Climate models were used to simulate global surface temperatures under both human (i.e. anthropogenic greenhouse gas emissions) and natural (e.g. solar and volcanic) forcings, and under natural forcings only. The simulations that consider both human and natural forcings align remarkably well with observations, while projections with only natural forcings fall far below (even distinctly below the envelope of uncertainty). As such, human activities are deemed directly responsible for approximately 1.1°C of warming since the late 1800s.

The ever-improving performance of climate models might just be the only positive take-home message from the first instalment of the IPCC Sixth Assessment Report, with the general findings showing an increasingly bleak future for planet Earth and humanity. What this highlights to me is the need to harness the power of these models and use their projections to prepare ourselves for the worst. While the mitigation of greenhouse gas emissions is clearly fundamental if we are to avoid a full blown catastrophe (with warming of up to 4.4^oC predicted by the end of the century), it is also clear that we need to prepare for the rapidly intensifying impacts of this warming.

If AR6 doesn’t hammer it home quite enough for you, just take a look at the extreme events being experienced around the world – in the past month alone we have seen wildfires in Greece, Italy and Siberia; floods in Germany, China and the UK; and even climate-change induced famine in Madagascar. Our countries, cities and societies are under threat, and well-informed adaptation planning is going to become ever-more fundamental to our survival.