This month, the paper from my second postdoc (with P.I. Uisdean Nicholson, Heriot Watt University) was published in the 500th Special Publication of the Geological Society, a volume dedicated to 'Subaqueous Mass Movements and their Consequences'. The paper, titled 'Indonesian Throughflow as a preconditioning mechanism for submarine landslides in the Makassar Strait' is, admittedly, quite heavy reading for anyone who is not from a sedimentology background. Therefore, this month, I look at what the postdoc was all about, the main results, and what it means for our understanding of geohazard risk in Indonesia.
Rationale for the study
This research is born out of previous work (see BBC video) by Uisdean Nicholson, who identified a link between deep ocean currents and submarine landslides to the south of the Falkland Islands. Looking for further examples where strong currents and tsunami hazards co-exist, Indonesia was a natural place to continue this research.
Southeast Asia is particularly prone to tsunamis due to its seismically active geology. In the last two years alone, the Palu Bay and Anak Krakatau tsunamis have claimed over 4,000 lives. Most tsunamis in the region are triggered by earthquakes, or volcanic activity. We know this as there are good historical records of past tsunami events and their triggers going back over 200 years.
The Makassar Strait (that separates the Indonesian Islands of Kalimantan and Sulawesi) is a narrow ocean gateway where ocean currents are focussed. Despite this, no historic tsunamis are listed as being caused by submarine landslide. This study, therefore, looked for pre-historic-evidence for submarine landslides that could be tsunamigenic.
Geological & Oceanographic Setting
The Makassar Strait is the main 'gateway' for the exchange of water from the Pacific into the Indian Ocean. As Pacific Ocean waters enter the narrow Strait, they are squeezed and accelerated to form a high velocity 'jet' called the Makassar Throughflow. Because of Coriolis Forces, the Makassar Throughflow is focussed to the west side of the Strait where is can erode and redistribute sediments entering the ocean from the Mahakam Delta.
We used bathymetric and seismic data (collected for oil exploration) to map the sea bed, and the most recent (last 2.6 million years) of sediment below the sea bed. The aim of this was to collect evidence for the Makasar Throughflow affecting the seabed, and look for ancient submarine landslides preserved in the rock record.
Results: Contourites & Mass Transport Deposits.
Seismic data allows us to image a slice of the ground below the seabed. This is done using a man-made energy source (dynamite, airgun or seismic vibrator) to send sound waves down into the subsurface. Boundaries between rocks of different properties cause a portion of the sound wave to be reflected back to the surface receiver. By understanding these changes in rock properties, a geophysicist can create an image of the rock layers in the subsurface. Different rock types give different seismic responses, and we can use this knowledge to say something on the processes they were deposited by.
Where ocean currents erode, redistribute and deposit sediments on the seabed, the resulting features are named 'contourites.' Contourites have characteristic shapes and seismic responses that can be clearly identified along the upper continental slope of the western side of the Makassar Strait. There is clear evidence in the seismic data for the Makassar Throughflow shaping the seabed forming erosional terraces and scarps, and deposition of contourite 'drifts' (Figure 3). These contourite features correlate well with the Makassar Throughflow jet, as seen in temperature and salinity data (Figure 3).
At the base of the continental slope, towards the centre of the basin, a series of mass transport deposits can be seen. These are identifiable where the seismic data is transparent or chaotic looking. These mass transport deposits were deposited by catastrophic submarine landslide events, many with over 100 km3 of sediment cascading down the slope. It is likely that such large submarine landslides caused a considerable tsunami wave like the one shown in the video below (video by GeoscienceAustralia).
Every mass transport deposit has been mapped out around the Makassar Strait (Figure 5). It is really clear that almost all submarine landslides are occurring on the western side of the Strait only, with the very largest (red) occurring in the southwest. This study shows that the largest submarine landslides are coming from a region that contourite deposition is most likely. The Makassar Throughflow picks up sediment coming into the basin at the Mahakam Delta, and piles it up along the margin in the southwest of the Strait making the seabed locally steeper, weaker, and more likely to collapse.
Conclusions & Future Work
Figure 5 highlights in red, the parts of the coastline that are most at risk should another submarine landslide occur and produce a tsunami. In blue are the regions of coastline that have historically seen relatively frequent tsunamis formed by earthquakes. Although no submarine landslide-induced tsunamis are seen in historical records, we now know that they were very likely to have occurred in the past, and need to understand the risk of them occurring again. We have identified a potential tsunami risk that and is triggered by an entirely new geological scenario. As such, the coastal communities at risk do not benefit from hazard mitigation or early warning systems that are currently in place elsewhere in Indonesia.
The next step is to quantify the risk. This research is ongoing at Heriot Watt University where tsunami modeling will assess wave propagation and run-up height for various scenarios. Geological sampling in Indonesia will look for evidence of pre-historical events across the region in the rock record, try to understand the frequency of events, and therefore the triggers. I hope to be able to share the results of this work (completed by Dr. Kirstie Wright, Dr. Urve Patel and others) soon!
Want to know all the details? Download the original paper here