In order to elucidate the transition of insect fauna of the Japanese archipelago, Pliocene to Pleistocene fossil insects were surveyed in the Japanese main islands. As a result, it was found that extinct and extant species coexisted in the Pliocene to Early Pleistocene. Most extinct species are thought to have been extinct at the end of the Early Pleistocene to Middle Pleistocene. In particular, the abundant fossil records of Donaciinae leaf beetles (Coleopetera, Chrysomelidae) show that the records of the extant species include “old extant species” dating back to the Pliocene to Early Pleistocene and “new extant species” appearing in the Middle Pleistocene. The transition of the beetle fauna was divided into four ages. The time when the recent Donaciinae fauna were almost completed is after the Middle Pleistocene. Also, the time axis of the molecular phylogenetic tree was set, and the speciation age was estimated, based on the fossil record of the extant species.

The 2015 Kanto-Tohoku Torrential Rain that hit the Kanto and Tohoku regions of eastern Japan from September 9 to September 12, 2015, resulted in a major riverine and slope disaster. Overbanking and breaching occurred along the central part of the Kinu River and formed crevasse splays on the flood plain around Kami-Misaka, Joso City, Ibaraki Prefecture. We clarified the distribution of crevasse splay deposits based on an interpretation of aerial photographs taken just after the disaster and a field survey. In addition, we performed an excavation survey and grain size analysis of the crevasse splay deposit. These resulted in the following findings.

1) Crevasse splay deposits mainly comprise fine to medium sand and can be divided into the following three units : Unit A, sandy silt to fine sand layer showing upward coarsening ; Unit B, fine to medium sand layer showing upward fining ; and Unit C, fine to medium sand layer with parallel lamination and low-angle cross-stratification.

2) The difference between these units probably corresponded to the changes in flow conditions during the flooding. Unit A, deposited at the early period of flooding, was derived from heavy rain and overbank of the river water ; Unit B, deposited during hydraulic bore, was generated by the breach of the river bank, flowing down the flood plain ; Unit C was formed by the flooding flow from the Kinu River via the breach point during the flood.

3) Typical stratigraphy of crevasse splay deposits in this study is characterized by the upward-fining unit (Unit B), which shows a difference from the known flood deposits.

Using thermoluminescence (TL) technique (multiple aliquot regeneration method), we dated White Pumice tephra (WP), an unconsolidated tephra on Middle Pleistocene (MIS 7) marine terrace deposits at the Kamikita plain, Aomori Prefecture. We collected 6 and 10 samples from WP on outcrops at Kacchi and Shichihyaku, respectively for evaluation of chemical leaching. Concentration of radiometric elements (U, Th and K₂O) was measured for all 16 samples. Among them, five samples were selected for TL age determination. The TL ages of three samples (187±29ka, 230±23ka and 229±38ka, weighted mean : 216±15ka) well coincide with a reported age of WP (MIS 7b ; 205-210ka), correlated with δ¹⁸O stratigraphy of deep-sea sediments. The TL ages of other two samples (165±28ka and 265±33ka) are beyond the error range to 205-210ka at mid MIS 7. The chemistry of former three samples falls in the concentrated zone of U-K₂O and Th-K₂O diagram. Other two samples were plotted far from concentrated zone of the same diagram. For TL age determination of unconsolidated tephra, annual dose evaluation through simple screening of chemical character is critical to obtain reasonable results.