Monday 29 December 2014

Philip Garrett visits GSB

For the last full week before Christmas, the Geological Survey of Belgium, Royal Belgian Institute of Natural Sciences, played host to Dr Philip Garrett, a Japanese scholar from the University of Cambridge. Dr Garrett is a Research Associate at the Faculty of East Asian and Middle Eastern Studies, where his research in the field of medieval Japanese history focuses on land and society in central Japan. His detailed historical knowledge, particularly of sites around the Kii Peninsula, is of great interest to the QuakeRecNankai project. The Kii Peninsula lies in the centre of the coastline affected by earthquakes and tsunamis from the Nankai-Suruga subduction zone. The southern tip of the peninsula, Cape Shionomisaki, marks the hypothesised boundary between the source areas of Nankai and Tōkai earthquakes.  
 
The Kii Peninsula, south central Japan. Yellow lines indicate plate boundaries. The lakes around Mount Fuji and the area around Lake Hamana are the main targets of the QuakeRecNankai project.
The productive visit included locating and translating numerous papers in a wide range of hard-to-find Japanese publications on the documentary and geological evidence of historical earthquakes and tsunamis. We're particularly interested in how the focus of paleoseismological research has changed since 2011, when inadequate anticipation of the size of the Tōhoku earthquake and tsunami led to a rapid reevaluation by the Japanese Cabinet Office of how earthquake and tsunami hazards should be assessed. New guidelines stress the necessity of evaluating the largest possible size of earthquake and tsunami from all available lines of evidence. With some notable exceptions, the majority of this research, and of research from before 2011 has been published in Japanese only. 

Dr Garrett also led a useful and entertaining session on professional manners and Japanese etiquette for QRN collaborators from the Geological Survey, Ghent University and the University of Liège. 

Tuesday 16 December 2014

X-ray computed tomography (CT scans)


All of our survey equipment and samples arrived in Belgium safely. Time to get the real work started! 

Before opening the cores from Lake Hamana, three-dimensional tomographic images were made with a medical CT scanner at the Ghent University Hospital (UZ Ghent). Indeed, the same device doctors use to look at bone fractures or detect other medical conditions. The X-ray CT technology is based on sending out a fan-shaped beam of computer processed X-rays from a helically revolving source. These rays become attenuated by a centrally positioned sample object (in this case, a sediment core) and collected again by a rotating array of detector cells. This setup allows to make full body scans of the objects of interest (Last & Smol, 2001). Resulting 3D images are a composite of series of 2D tomographic slices, in which every volume element (voxel) is represented by a specific shade of grey. These greyscales are a reflection of the measure in which the X-ray beam has been attenuated along its pathway and depend on the density as well as the composition of the sediment (Orsi et al., 1994). White voxels correspond to a complete attenuation of the signal, whereas black ones indicate no attenuation at all. Generally coarser grained sediment tends to attenuate the signal more severely than finer grained material. Also, every mineral generates its own characteristic attenuation, which is a function of the applied level of radiation energy.

Schematic side view and cross view of a medical CT scanner while imaging a sediment core. A rotating X-ray source continuously sends out a fan-shaped beam, which is received again by the synchronously revolving detector array. Meanwhile a motorised table gradually moves the core through the scanner, resulting in a helical scanning path. 

Within the QRN-project, CT images will be used mainly to identify depositional structures that are impossible or difficult to recognise on split surface photographs. They allow us to obtain a first impression of the content of the cores before even opening them. Moreover, processing software makes it possible to manipulate the data in order to enhance the visibility of specific features (e.g. coarse grained intercalations (tsunami deposits?), shells, rocks, laminations, cross stratification…).

Top: CT scanner at the Ghent University Hospital, illustrating the setup for X-ray core scanning on an already opened liner (not from Lake Hamana). Bottom: Example (core from Lake Kenai, Kenai Peninsula, south-central Alaska) of how resulting CT images (A) tend to look like, compared to original sediment surface pictures (B). This record is composed of an earthquake triggered event deposit or ‘turbidite’ (brown line) embedded within background laminations (light blue lines).

Timelapse of the Ghent University Hospital X-ray CT scanner in action. 


References

Last, W.M. and Smoll, J.P. (2001). Tracking environmental change using lake sediments. Volume 2: Physical and geochemical methods. Kluwer Academic Publishers. 515 pp.

Orsi, T.H., Edwards, C.M. and Anderson, A.L. (1994). X-ray computed tomography: A non-destructive method for quantitative analysis of sediment cores. Journal of Sedimentary Research 64A, 690-693.