EGU General Assembly 2020
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

How to reconcile OSL and TCN data: the potential of high-resolution sampling on the Choushui Tableland (West Central Taiwan)

Magali Rizza1, Brice Lebrun1,2, Lionel Siame1, and Valéry Guillou1
Magali Rizza et al.
  • 1CEREGE, Aix-Marseille University, Aix-en-Provence, France (
  • 2IRAMAT - CRP2A, Bordeaux-Montaigne University, Bordeaux, France

The determination of fault slip rate is often inferred from dating of Quaternary, deformed geomorphological surfaces affected by fault activity. For this reason, cosmogenic and luminescence methods now are widely applied to date the emplacement of geomorphic markers, but each method relates to different geomorphic processes. While the Terrestrial Cosmogenic Nuclides (TCN) method generally dates the exposure duration of the rock surface to cosmic rays, the Optically Stimulated Luminescence (OSL) method provides burial duration of the sediment after deposition. Age differences between these two methods may relate to the erosion-transport-deposition and aggradation processes experienced by the sediment prior its final deposition but combined may provide new insights into the processes affecting alluvial landforms.

Our case study is located in the Western Foothills, south of the Choushui River (Central Taiwan). There, slip on the Changhua blind thrust fault has caused the eastward tilt of a wide flight of fluvial terraces but slip rates on frontal faults are still debated due to large epistemic uncertainties in dating alluvial surfaces with OSL and TCN methods. To achieve a finer chronology of the deposits, a high-resolution sampling strategy has been deployed leading to a direct and unique comparison between OSL and TCN dating methods. Taking advantage of a natural exposure, we collected 10 samples for 10Be dating completed by 5 OSL samples along a 7 m depth profile. The depth distribution of 10Be concentrations show a complex depositional history with at least two depositional sequences, modelled to be older than ~38.7 ka.

As previous work has shown the difficulties of OSL dating in Taiwan, particular attention has been paid to luminescence characteristics of quartz and potential dosimetry issues. Our OSL analysis are in good agreement with 10Be and previous 14C dating and also reveal three depositional units, dated between ~9 ka and ~66 ka, that are evidenced by different OSL signal characteristics and variations in dosimetry.

This study shows that it is informative to have an exhaustive, detailed, and direct comparison between dating methods on a single depth profile to discuss the geomorphic processes and allow a more detailed understanding of the long-term rates of the Changhua Fault.

How to cite: Rizza, M., Lebrun, B., Siame, L., and Guillou, V.: How to reconcile OSL and TCN data: the potential of high-resolution sampling on the Choushui Tableland (West Central Taiwan), EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-4595,, 2020.


Display file

Comments on the display

AC: Author Comment | CC: Community Comment | Report abuse

displays version 1 – uploaded on 29 Apr 2020
  • CC1: Comment on EGU2020-4595, Ramona Schneider, 01 May 2020

    Personally, I think this is a very well-designed and intuitively understandable presentation! Very interesting, thank you!

  • CC2: Comment on EGU2020-4595, Sebastien Huot, 09 May 2020

    Interesting work, Magali!

    On slide 9, could you elaborate on what represent the difference in dose rate, between with or without gamma? Does it means that the dose rate, deduced from the portable gamma spectrometer, is substantially different (lower) from what you calculated from ICP-MS and ICP-OES? Could you speculate on what would be the cause?

    As for the lower half of the section, I agree also in your hypothesis of a change in sediments source. There is one more piece of evidence you could add, even small, to that puzzle. Display the K versus Th abundances for these samples, on the same plot. Look for a pattern that might show up.

    All clustered would imply the same mixture of source(s). All stretched along a common line imply a dilution, between two source (group) of sediment. It could be neither; in such as case, the difference in source area(s) is even greater, which would lend more support to a change in source (drainage) area, over time.

    You can repeat the same exercise with U vs Th: this should stretched on a common line, passing through the origin. Nearly in all situation. K and Rb should also correlate very closely, because they are usually associated together.

    • AC1: Reply to CC2, Magali Rizza, 10 May 2020

      Hi Sebastien! Thanks for your very useful comments. I will have a deeper look on your suggestions and try to reply here or by email later after having plotted the data.