Global Climate Dynamics
Kosaka Laboratory

Exploring the formation and variation of the climate system with big data analyses and simulations

Investigating mechanisms of global linkages in climate variability and change to identify keys for climate prediction and projection

The Earth’ s atmosphere and ocean interact mutually and form a complex climate system. In addition to climate change driven by external forcings –such as variations in solar insolation and human-induced greenhouse gas emissions– the climate system also exhibits variations arising from its internal dynamics. This internal climate variability, which sometimes occurs on a global scale, is characterized by “teleconnections” and “atmosphere-ocean interactions” .

Large-scale atmospheric variability spanning thousands of kilometers often organizes into distinct spatiotemporal patterns, known as teleconnection patterns, which link climate conditions across distant regions. Compared to the atmosphere, the ocean varies more slowly, exerting influence on global climate from seasons to decades. Oceanic variability and its associated teleconnections are key to climate predictions from months to seasons ahead. Those internal climate variability can interact with human-induced climate change, often intensifying extreme events like heat waves, or conversely, causing temporary regional cooling even amid ongoing global warming.

We investigate climate variability and change by analyzing gridded four-dimensional data sets of the atmosphere and ocean from the past to present –called “reanalysis” data– as well as ensemble climate model simulations initialized with slightly different initial conditions. By leveraging those vast climate datasets and by designing and conducting new, targeted climate simulations, we aim to deepen understanding of global climate covariability and predictability, ultimately providing actionable climate information to support climate resilience and policy.

Producing regional atmospheric reanalysis data

Despite the continuous production of the global atmospheric reanalysis data by the Japan Meteorological Agency (JMA), no regional atmospheric reanalysis data in operational quality is thus far available for Japan. Under the funding by the Japan Agency for Science and Technology and close collaboration with JMA and Information Technology Center of our university, we are conducting a 10-year project “ClimCORE” (Project Leader: Senior Research Fellow H. Nakamura) to produce high-resolution atmospheric reanalysis data (RRJ-ClimCORE) over Japan and its surrounding maritime domain for broad business and community applications. Based on the latest version of the JMA meso-forecast system, the hourly RRJ-ClimCORE data will become available for the 20+ years for the current century at 5-km grid intervals.

Column
Our lab’s principal investigator (PI) is a devoted Doraemon fan. Once the lab members found out, a giant Doraemon plushie mysteriously made his way into the lab. Since then, students and staff have been steadily contributing Doraemon-themed items, and now the office is filled with Doraemon items. The plushie proudly sits at the heart of the lab, offering a sense of comfort and fun to everyone —from the PI to those dropping by for discussions.


The master of the laboratory. Left: Taken when he first arrived at the laboratory. Right: Taken in June 2025.

Member

• 
• • Associate Professor
    Yu KOSAKA
  Research Area: Climate variability, Extreme weather, Global climate simulation

• 
• • Project Associate Professor
    Takafumi MIYASAKA
  Research Area: Climate variability, Extreme weather, Air-sea interaction

• 
• • Senior Research Fellow
    Hisashi NAKAMURA

Laboratory Homepage