Pan-Arctic Climate Change Projection Research Group
Developing a Highly Accurate Pan-Arctic Regional Model and Making Detailed Predictions for Several Decades into the Future
Under the overarching theme of “prediction,” our Pan-Arctic Climate Change Projection Research Group is using JAMSTEC’s “Earth Simulator” supercomputer in an effort to develop a highly accurate numerical model for the Pan-Arctic region. What will the Earth, or Japan, be like 10, 20, or 50 years from now owing to global warming and other climate change? To find out, it’s important to increase our understanding of the Arctic and its surrounding regions.
The Arctic is one of the world’s regions where climate change impacts are most pronounced, as symbolized by the rapid decrease in the sea ice that once covered the Arctic Ocean. Over the last few years, attention has focused on questions such as, “How will the melting of
the sea ice affect other regions?” The Arctic is strongly affected by climate change, but because of the great magnitude of change, it could at the same time have strong impacts on other regions, including Japan. For that reason we are building a model that covers not only the Arctic, but also the “Pan-Arctic region,” which includes the peripheral regions around the Arctic.
Models for global-scale, long-term simulations, called climate models, have been created at research institutions in many countries with collaboration by many researchers. The results of those models have been collected, examined, and released through international frameworks such as the Intergovernmental Panel on Climate Change, an organization that assesses research results on global warming and other climate changes caused by human activities. But because performing simulations for the entire Earth requires enormous computing resources, there is a limit to detail even with the power of a supercomputer. For example, many of the climate models provided for the IPCC’s most recent report performed their calculations by covering the Earth with grids about 100 km on a side. Although climate models provide an indicator for understanding changes over the Earth as a whole, 100-km grids are too big to reproduce local changes and meteorological events. We will therefore create a model that uses large grids for everywhere except the Pan-Arctic region, which is divided into small grids about 5 km on a side, to obtain results that are highly accurate for the Pan-Arctic region despite limited computational resources. Considering the fact that rapid changes in the Arctic are greatly affecting other regions of the Earth, it is our expectation that the results obtained from such models will lead to a new understanding of, and more detailed predictions for, climate change not only in the Arctic but also the entire planet.
One more thing that distinguishes our unit is modeling of the ice sheets in Greenland and other places. Even Greenland with its vast ice sheet is beginning to experience previously unseen phenomena, such as the large scale and rapid melting of ice in summer. Melting of the Greenland ice sheet feeds immense volumes of fresh water into the ocean: one cannot overlook the impact. Our incorporation of a Greenland ice sheet model into our climate model is internationally cutting-edge research through which we hope to be helpful in further improving the accuracy and reliability of predictions from the Pan-Arctic to mid-latitude regions.