CORDEX-CMIP6 Australian cyclone tracks

This archives cyclone tracks for the historical and ssp370 scenarios generated from 40 regional climate models which downscaled CMIP6 data using the CORDEX-Australasia framework. The models are described in more detail in Pepler et al. (2015), and are the produce of four different Australian state and federal agencies:

• The Bureau of Meteorology (BOM) BARPA-R, which downscaled 7 GCMs and is available at https://dx.doi.org/10.25914/gjzx-kr91
• The CSIRO CCAM simulations, which downscaled 7 GCMs and is available at https://dx.doi.org/10.25914/rd73-4m38
• The Queensland Government (UQ-DES) simulations from QldFCP-2, which has 16 simulations using a range of GCMs and CCAM settings, available at https://dx.doi.org/10.25914/2c0z-8t40
• The NSW Government simulations from NARCliM2.0, which downscaled 5 GCMs using two different WRF setups and is available at https://dx.doi.org/10.25914/3r9s-pb86

For all RCMs, cyclones were tracked in 6-hourly MSLP and 500hPa geopotential height data using the Murray and Simmonds (1991) and Simmonds et al. (1999) cyclone tracking scheme. The source code and documentation is available at https://cyclonetracker.earthsci.unimelb.edu.au/. Similar datasets are also available for a small selection of CMIP6 GCMs (https://doi.org/10.6084/m9.figshare.28887851.v1) and reanalyses (https://doi.org/10.6084/m9.figshare.28887851.v1).

Cyclone tracks are provided based on the weakest intensity threshold, with stronger thresholds typically identified to select lows for use in the paper. For more details on the parameters used for detection and tracking see the files *incycloc and *intrack which are used as inputs for the tracking scheme. The regions used for tracking differ slightly between the three modelling methods (BARPA, CCAM and NARCliM), and are apparent through the relevant incycloc file.

A separate track file is provided for each year, beginning on 1 December of the previous year, to allow for the identification of cross-year tracks when relevant. Each .tar.gz file includes all tracks for one simulation, e.g. $model_$member_$agency_$rcm_$ssp_$level_$details.tar.gz

Column headings:

• ID Track ID
• Fix Instance within track
• Date Date* (UTC)
• Time Time (UTC)
• Open/closed Is there closed circulation (0) or an "open" ridge.
• Lon Longitude
• Lat Latitude
• MSLP Central pressure (hPa)
• Laplacian Laplacian of MSLP (hPa. Deg.lat^2), which represents the curvature of the MSLP. This is a better indicator of the strength of a system
• Depth Depth of cyclone (hPa) between central point and outer radius
• Radius Radius of cyclone (degrees)
• Up Steering velocity (U)
• Vp Steering velocity (V)

*Dates are two digit years, and in some cases are wrong because of the way the tracking scheme works (e.g. 1999-2001). To correct these, check the number of unique years in the file - if there is only one year, then that is the same as given in the filename; if there are 2 years, the first month will be the December of the previous year.

References:

• Murray, R. J., and I. Simmonds, 1991: A numerical scheme for tracking cyclone centres from digital data. Part I: Development and operation of the scheme. Aust. Meteorol. Mag., 39, 155166.
• Simmonds, I., R. J. Murray, and R. M. Leighton, 1999: A refinement of cyclone tracking methods with data from FROST. Aust. Meteorol. Mag., (special edition), 3549.
• Pepler AS, Di Virgilio G, Dowdy AD, Goyal R, Su C-H, Thatcher M, and Syktus J. (2025) Projections of Australian low pressure systems in downscaled CMIP6 models Submitted to JSHESS