Beach ridge records of late-Holocene tropical cyclone behaviour

Currently, there is a lack of detailed knowledge about the frequency of extreme intensity tropical cyclones in Queensland, Australia. As a consequence, it is difficult to conduct a robust assessment of the accuracy of current tropical cyclone (TC) risk estimates for this region. The objective of this study was to determine whether the coastal barriers in this region might contain long-term TC records that will prove useful in overcoming this difficulty. Such long-term records have been identified in both coral shingle ridges in the past as they have in similar landforms composed of shell and sand. However, a more geographically abundant type of record is required in order to gain a realistic picture of regional TC behaviour. Landform morphology, sedimentology and luminescence chronology were used in reconstruction of three beach ridge plains located along the northeast Queensland coast. These include a coarse-grained sand beach ridge plain at Rockingham Bay, a 'complex barrier' comprising fine- and coarse-grained sand ridges at Wonga Beach and a fine-grained sand beach ridge plain at Cungulla. Contrary to traditional interpretations about the origins of shoreparallel sand barriers, results suggest that aeolian processes are unlikely to have played a substantial role in the principal deposition of these ridges. Instead, it appears that marine inundations accompanying TCs are responsible for their deposition since the mid-Holocene. This is significant because these results suggest that sand beach ridges can be deposited by storm waves and surge and that the texture of these landforms need not necessarily be indicative of the processes responsible for ridge development. Records contained in the coarse-grained sand beach ridges at Rockingham Bay and Wonga Beach suggest that there has been considerable variation in both the intensity and frequency of TCs at these locations over the past 6000 years. This variability is characterised by extended alternating periods (centuries to millennia) of relative quiescence and heightened intense TC activity. These results differ from most previous studies of long-term sedimentary records conducted using coral shingle or shell and sand beach ridges that display relatively uniform patterns of deposition. Therefore, pure sand ridges seem to provide a more sensitive record of variations in long-term TC climatology. Consequently, sand beach ridge records may contribute to a considerable improvement in our ability to understand the hazard posed by TCs. There are chronological gaps in each of the ridge plains examined. The ridge plain at Cungulla offered an opportunity to determine whether these were caused by erosion or if they were due to quiescent phases in storm activity. Tropical cyclone records contained in the fine-grained sand beach ridge plain at Cungulla enabled characterisation of both erosional and non-erosional gaps. The erosional gaps are associated with changes in orientation between ridge sets and often a high ridge with hummocky topography that appears to have been disturbed by aeolian activity. River floods likely caused the partial erosion of ridge sets. Non-erosional gaps do not display these morphological characteristics and are likely associated with quiescence in severe tropical cyclone activity. These geomorphic and chronological signatures can be used to identify gaps of varying origin in other ridge plains and are an important tool in the reconstruction of long-term storm beach ridge records. Also, because sand beach ridges are widespread in northeast Queensland, it appears likely that further studies of this kind may provide a means of testing current risk assessments for that region. A comparison made between the northeast Queensland beach ridge records and long-term sedimentary TC records from elsewhere shows remarkable similarity in the behaviour of these hazards at the global scale. All of the records display extended alternating periods (centuries to millennia) of relative quiescence and heightened intense TC activity irrespective of both the resolution and type of record. Hence, patterns of TC behaviour in the beach ridge records are unlikely to be an artefact of the record type. And despite difficulty identifying the cause of punctuated activity, the identification of these patterns is vital for assessing the risk posed to coastal communities by TC hazards