Taichi Sano, Yusuke Yokoyama, Kanon Kino, Taikan Oki
Received 31 July, 2025
Accepted 10 October, 2025
Published online 18 July, 2026

Taichi Sano¹⁾, Yusuke Yokoyama¹⁾, Kanon Kino¹⁾, Taikan Oki¹⁾

1) Department of Civil Engineering, Graduate School of Engineering, The University of Tokyo, Japan

Climate change has intensified extreme weather events, raising urgent concerns for densely populated regions like Japan. This study applies the Climatic Risk Boundary framework, a recently proposed concept suggesting that adaptation can be difficult when future extremes exceed historical human experience. We applied it to Japan using two bias-corrected datasets of different resolutions and the large-ensemble d4PDF experiment. SSP-consistent 1 km population projections were integrated to assess exposure. Results show consistent warming, with temperature extremes rising about 5°C and precipitation extremes by 0–50 mm by late century under SSP5-8.5. High-resolution datasets revealed localized Climatic Risk Boundary transgressions that coarser products in earlier studies could not capture, exposing risks at finer scales. Across datasets, broad agreement exists that major Japanese cities will face intensified risks, yet differences remain: the CMIP6-based datasets project most major cities – including Tokyo, Osaka, and Nagoya – will transgress the boundary, whereas d4PDF’s large ensemble yields smoother extremes and no transgression at the city scale. By 2100, many urban areas, especially along the Pacific coast, are projected to face unprecedented conditions. These findings highlight the need to consider resolution- and method-dependent uncertainties and inform adaptation planning in Japan’s urban and regional contexts.

[Full Text]

Copyright (c) 2026 The Author(s) CC-BY 4.0