Assessing the spatial and temporal characteristics of cyclone tracks along the east coast of Australia

East coast cyclones (ECCs) pose a significant threat to the communities of the eastern seaboard of Australia. Although an important contributor to freshwater resources, ECCs can also be disastrous and costly – not only for communities directly on the eastern seaboard, but also west of the Great Dividing Range and as far south as Tasmania. This region encompasses Australia's agricultural food bowl (the Murray Darling Basin), as well as some of Australia’s most rapidly expanding cities. Thus, understanding the historical ECC-related rainfall impacts and identifying emerging trends in ECC characteristics will be essential in managing future events, especially considering projected increases in rainfall intensity and associated flooding with a warming climate.

Previous studies on ECCs have explored many periods, inclusive of the satellite (1979 to current) and pre-satellite era. No meaningful relationship between cyclone occurrence and climate drivers (e.g. the El Niño Southern Oscillation) has been established. Failure to detect a meaningful relationship could be a combined result of a relatively brief time period, the complexity of potential connections (if any) or too much stochastic variability. A further limitation of ECC databases is the fact that they are not regularly updated and, as such, do not include the most recent events (e.g. post-2019). Capturing more recent events like the February 2020 east coast low and their impacts is important for forecasting how cyclones may conduct themselves in the current and future climate.

Rainfall impacts from ECCs occur along the length of their storm track. However, data on cyclone tracks have been notably underutilised in terms of their ability to identify environmental characteristics such as event rainfall along the east coast of Australia. As a result, track-specific studies that explore rainfall impacts (particularly sub-daily) for Australian ECCs are uncommon within the current literature. This thesis builds on the existing ECC knowledge base by examining groupings (clusters) of ECC tracks over the period of 1950 to 2019 and quantifies their rainfall impacts. The intended purpose of this work is to use the historical ECC characteristics to provide actionable information that aids in preparing for the impacts of future natural hazards.

An initial comparison of two Australian low-pressure datasets with coverage from 1950 to 2019 was used to determine the most suitable database for extending ECC track studies. The results conclusively determined that the NCEP1 (National Centres for Environmental Prediction) database provides a greater representation of cyclonic systems than the MATCHES (Maps and Tables of Climate Hazards of the Eastern Seaboard) database, with a two-fold increase in track length, duration and spatial coverage. Further, NCEP1 was found to have fewer inconsistencies with track configuration. In contrast, MATCHES had a large number of split tracks that required manual intervention to be comparable to NCEP1.

Clustering analysis using the NCEP1 database established a new typology of ECCs based on their geographical location and track shape. Six clusters were identified, with a further subclassification into three groups based on their proximal orientation to the coastline (i.e. coastal, continental or tropical). Each cluster has a distinct seasonality and impacts different regions of eastern Australia. The more common cyclone type is coastally orientated and produces less rainfall when compared to both tropical and continental ECCs. Conversely, continental tracks have the highest intensity and more extensive spatial precipitation impact because the tracks maintain a longer duration over land. In comparison, the tropical cluster had the least number of tracks but the highest rainfall quantity overall. The analysis of climate and environmental conditions established a greater understanding of potential, largescale drivers and associated spatial rainfall impacts for the different cluster types. For example, ECCs occurred more frequently in neutral phases of the El Niño Southern Oscillation, the Southern Annular Mode and the Indian Ocean Dipole.

Finally, sub-daily (1-hourly) rainfall for each cluster was amalgamated and mapped to outline hotspots of peak intensity precipitation at a radius of 500 km from the cyclone centre. The relationship between the intensity of the storms and associated rainfall was also investigated to determine the timing of peak rainfall occurrence for each cluster. A more extreme rainfall signature from cyclones was observed offshore than over land. No significant, decadal trends for cyclone intensity between or within the clusters were observed. Slower systems produced more severe precipitation, with higher hourly rainfall maxima more evident in the weaker ECC clusters. The drainage basins most impacted by elevated event and hourly maximum rainfall are located in the New South Wales and Victorian southeast coast and the eastern border of the Murray-Darling Basin. ECCs within both locations (especially the Murray– Darling Basin) can severely impact Australia’s food production, displace communities and damage infrastructure, highlighting the need to understand the mechanisms of precipitation from these events to increase the resilience of the communities in southeastern Australia.

The impacts of future climate and storm systems are a concern for Australian communities, with uncertainty around how the risk of extreme events should be managed in the current dynamic climate. Findings from this thesis are significant in providing context for current and future storm planning using historical ECC seasonality, potential climate drivers, rainfall footprint and the relationship of cyclone precipitation/intensity. These results are of practical use for Australian natural hazard authorities aiming to prepare for the adverse impacts of future ECC events.