Japan Geoscience Union Meeting 2023

Presentation information

[J] Oral

S (Solid Earth Sciences ) » S-CG Complex & General

[S-CG52] Ocean Floor Geoscience

Tue. May 23, 2023 3:30 PM - 5:00 PM 301B (International Conference Hall, Makuhari Messe)

convener:Kyoko Okino(Atmosphere and Ocean Research Institute, The University of Tokyo), Keiichi Tadokoro(Research Center for Seismology, Volcanology and Earthquake and Volcano Research Center, Nagoya University), Chairperson:KanHsi Hsiung(Japan Agency for Marine-Earth Science and Technology), Kyoko Okino(Atmosphere and Ocean Research Institute, The University of Tokyo)

3:30 PM - 3:45 PM

[SCG52-16] Age constraints of chemostratigraphic units in REY-rich mud around Minamitorishima Island with a deep learning-based image detection system for ichthyolith

*Takahiro Kitazawa1, Kazuhide Mimura2,1, Kazutaka Yasukawa1, Junichiro Ohta1,2, Koichiro Fujinaga2,1, Kentaro Nakamura1,2, Yasuhiro Kato1,2 (1.School of Engineering, The University of Tokyo, 2.ORCeNG, Chiba Institute of Technology)

Keywords:ichthyolith, biostratigraphy, chemostratigraphy, deep-sea sediment, deep learning, REY-rich mud

In recent years, the demand for rare-earth elements and yttrium (REY) has increased globally with the promotion of renewable energy and the spread of electric vehicles. As a new promising resource for these industrially critical elements, Kato et al. [1] discovered deep-sea sediments enriched in REY, or REY-rich mud, in the Pacific Ocean. Furthermore, Takaya et al. [2] estimated that several hundred times the annual world demand for REY lie in the promising area of about 2,500 km2 in the southern part of Japan's exclusive economic zone (EEZ) around Minamitorishima Island.
Tanaka et al. [3] established a chemostratigraphy based on multi-elemental features in bulk chemical composition of 49 sediment cores collected in the area, resulting in a classification of the sediments into five units and three REY peaks. This chemostratigraphy enabled the apparently homogeneous pelagic clay layers including REY-rich mud to be correlated with each other.
To decipher the entire depositional history of the Minamitorishima REY-rich mud, age-determination of each chemostratigraphic unit is essential. However, the depositional ages of each chemostratigraphic unit remain poorly constrained. The common dating methods such as siliceous and calcareous microfossil biostratigraphy or palaeomagnetic stratigraphy are hardly to be applied to the REY-rich mud, or barren pelagic clay in lithology, due to a great water depth that dissolves most fossils and a low sedimentation rate that obscures geomagnetic polarity reversals. For this reason, it has not even been verified whether each chemostratigraphic unit in this area were temporally coincident with one another or not.
The only biostratigraphy that can be applied to REY-rich mud is the ichthyolith stratigraphy. Ichthyolith is a type of microfossils composed of fish teeth and denticles [4]. They consist of calcium phosphate that is less dissolvable than carbonate or silica (biogenic opal), resulting in being well preserved in pelagic clay [5]. In order to efficiently determine the depositional ages of pelagic clay, Mimura et al. [6] recently established a method to automatically detect ichthyoliths in microscopic images using a deep learning technique. We applied this method to several cores collected from the Minamitorishima EEZ to examine the depositional ages of the chemostratigraphic units. The preliminary results will be discussed in the presentation.

References: [1] Kato et al. (2011) Nature Geoscience 4, 535-539. [2] Takaya et al. (2018) Scientific Reports 8, 5763. [3] Tanaka et al. (2020) Ore Geology Reviews 119, 103392. [4] Doyle and Riedel (1985) Init. Repts. DSDP 86, 349-366. [5] Sibert and Norris (2015) Proceedings of the National Academy of Sciences 112(28), 8537-8542. [6] Mimura et al. (2022) Applied Computing and Geosciences 16, 100092.