Just Accepted

This article has been peer reviewed and accepted for publication. It is in production and has not been edited, so may differ from the final published form.

Disentangling the uncertainties in regional projections for Australia

Sugata Narsey , Michael Grose , François Delage, Gen Tolhurst, Christine Chung , Alicia Takbash, Ghyslaine Boschat, Malcolm King, Acacia Pepler , Marcus Thatcher, Benjamin Ng , Cong Hoang Son Truong , Chun-Hsu Su , Emma Howard , Christian Stassen , Mitchell Black , David Jones, Richard Matear, Sarah Chapman , Jozef Syktus, Ralph Trancoso, Giovanni Di Virgilio, Rishav Goyal, Jatin Kala, Vanessa Round, Jason Evans

Abstract

Understanding, quantifying, and visualising projected ranges of future regional climate change is important for informing robust climate change impact assessments. Here we examine projections of Australian sub-continental regionally-averaged surface air temperature and precipitation in the CMIP6 global and CORDEX-Australasia regional model ensembles and illustrate the relative sources of uncertainty from emissions scenarios, from models and internal climate variability. As expected, the uncertainty in temperature change for all regions by the end of the century is predominantly determined by the emissions scenario. Here we examined a low and high emissions scenario, bookending a range of plausible cases. In contrast, the uncertainty in precipitation changes towards the end of the 21st century is largely related to model-to-model differences. In particular due to the differences between global models, with regional models contributing a smaller, but still significant, source of uncertainty. Regional models can significantly alter precipitation projections, however we find few cases of consistency across the regional models. Decadal variability is an important contributing factor for precipitation uncertainty for the entire 21st century. Large changes in interannual precipitation variability are projected by some climate models by the end of the 21st century, and these changes tend to be well-correlated to mean precipitation changes. Robust responses to climate change must account for all of these dimensions in a structured way.

ES25015  Accepted 13 August 2025

© CSIRO 2025