The Paris agreement of 2015, among other developments, has generated a lot of interest in the effects of climate change at different levels of global warming since the pre-industrial era, including the 2 °C target. To understand climate change in Australia at these global targets, it is useful to first understand the historical change in Australia compared to the globe.

The reference period to use for ‘pre-industrial’ is a complex issue, and measuring global warming since pre-industrial times is also a tricky question. The majority of global studies use 1850-1900 as the ‘pre-industrial’ period, and global observational datasets (HadCRUT4, Cowtan and Way, NASA-GISS, NOAAGlobal, Berkeley) indicate that the globe has warmed by 1.0 to 1.1 °C between then and 2018.  However, these datasets have limited coverage prior to 1910, and mostly do not account explicitly for the introduction of Stevenson screens in Australia between 1890 and 1910, so have unreliable data and trends in the Australian region. For these reasons, Australia’s official temperature dataset ACORN-SATv2 is only provided from 1910.

Here we examine the various options for assessing warming in Australia from pre-industrial times, including the use of gridded observation-based datasets, climate models, and a quality-controlled early observation dataset for southeast Australia to derive Australia-wide temperature. We also examine different methods of assessing temperature trends, given the standard method of linear trends is becoming increasingly inappropriate as changes are non-linear. Preliminary results suggest Australia is approaching the threshold of 1.5 °C of warming since 1850.