. Scientific Frontline: Predicting typhoon intensity using ocean surface temperatures

Thursday, May 14, 2026

Predicting typhoon intensity using ocean surface temperatures

Conceptual diagram of this study on future changes in typhoon characteristics. Top left: Model outline. Top right: Considered changes. Bottom left: Example of results for variance in typhoon intensity by SST pattern (blue) and global warming (red) signals for difference exceedance probability.
Image Credit: Kyoto University / Nobuhito Mori

Scientific Frontline: Extended "At a Glance" Summary: Predicting Typhoon Intensity Using Ocean Surface Temperatures

The Core Concept: A new probabilistic modeling framework that combines spatial sea surface temperature (SST) patterns with a global atmospheric climate model to quantitatively predict the intensity and frequency of severe typhoons under historical and future climate conditions.

Key Distinction/Mechanism: Unlike previous evaluations that insufficiently accounted for varying sea surface temperatures, this approach couples a slab-ocean model with the Global Atmospheric Climate Model to simulate atmosphere-ocean interactions globally. Running at high resolutions (up to 20 kilometers), the model reveals that SST patterns and climate-driven SST increases explain 50 to 60 percent of the variance in typhoon intensity.

Major Frameworks/Components:

  • Slab-ocean coupled Meteorological Research Institute Global Atmospheric Climate Model (MRI-AGCM).
  • High-resolution, global-scale ensemble experiments executed at 60-kilometer and 20-kilometer scales.
  • Spatial sea surface temperature (SST) pattern analysis.
  • Probabilistic extreme weather event modeling and risk assessment.

Branch of Science: Climatology, Meteorology, Oceanography, and Environmental Engineering.

Future Application: Providing highly reliable risk assessments for policymakers to guide climate adaptation, infrastructure development, coastal protection strategies, and disaster management planning.

Why It Matters: As global warming accelerates, extreme typhoons that historically occurred once every 100 years are projected to strike four to five times per century. This model significantly reduces uncertainty in climate projections, enabling proactive safeguarding of vulnerable coastal communities against devastating storms.

Every year in the western Pacific, as summer ends and September begins, typhoons are not far behind. Typhoons are the most impactful extreme weather events affecting Japan and East Asia, and due to climate change, extremely strong typhoons are becoming more frequent. To adapt critical infrastructure to these massive storms and protect coastal areas, accurately accounting for their future impact is essential.

Assessing the disaster risks of a typhoon season involves quantitatively predicting storm intensity and frequency, relying heavily on the characteristics of regional meteorological fields and the natural variability inherent in the climate system. Though varying typhoon characteristics are related to sea surface temperatures (SST), probabilistic evaluations that account for SST have been insufficient.

This gap motivated a team of researchers at Kyoto University’s Disaster Prevention Research Institute to analyze typhoon intensity through spatial SST patterns. They combined a slab-ocean model with the Global Atmospheric Climate Model developed by the Meteorological Research Institute of the Japan Meteorological Agency, resulting in a unique simulation that successfully represents atmosphere-ocean interactions on a global scale.

Using this model, the team conducted ensemble experiments specialized for typhoon evaluation, probabilistically assessing the relationship between spatial patterns of SST in the Pacific and typhoon intensity characteristics. They performed these simulations under both historical and future conditions to assess changes while accounting for natural variability. Furthermore, they ran the experiments at a conventional 60-kilometer resolution and on a horizontal 20-kilometer scale.

The team’s results revealed that approximately 50% to 60% of the variance in typhoon intensity can be explained by the combination of spatial SST pattern differences and increasing average SST due to climate change. The probabilistic typhoon evaluations enabled by this study also led the team to discover that while extreme typhoons occur once every 100 years in the present climate, they may occur four to five times per century under future scenarios.

"The impact of SST influence on the typhoon intensity in severe typhoons is clearer than we expected, as is the impact of global warming on the increase in the frequency of severe typhoons," says team leader Nobuhito Mori.

This study provides a framework for high-resolution, global-scale ensemble experiments using combined atmosphere-ocean models, reducing uncertainty in climate projections by providing a highly reliable foundation for risk assessment. These findings have particular potential for policymakers involved in infrastructure development, such as coastal protection and disaster management planning.

Next, the team intends to further refine their model to strengthen their results. "We want to understand how extreme weather is changing due to climate change," says first author Yoshiki Matsuo. "This is an important challenge from both an engineering and a social perspective."

Published in journal: Journal of Climate

TitleProbabilistic Assessments on Future Changes in Typhoon Characteristics Based on Fixed-SST Ensemble Experiments by Slab-Ocean Coupled MRI-AGCM

Authors: Yoshiki Matsuo, Tomoharu Okada, Tomoya Shimura, Nobuhito Mori, Takuya Miyashita, and Ryo Mizuta

Source/CreditKyoto University

Reference Number: as051426_01

Privacy Policy | Terms of Service | Contact Us