Global model goes local, zooming in on hurricanes and other fine-scale features


Click image to view animation (requires free Quicktime player).
An AVI version is available, but image quality is not as high.
NCAR experimenters boosted resolution fourfold in two key components of the Community Climate System Model to produce the fine-scale details in this animation, which captures one month of a five-year simulation. The animation shows the global circulation pattern that transports water vapor from the tropics toward the poles, an atmospheric feature generally well simulated by climate models. But by dialing up the resolution, regional weather phenomena also emerge, including fronts and storm systems not previously visible in global models. For example, several tropical depressions form in the western Pacific Ocean, heading toward the east coast of Asia. Starting around August 8, two of these storms spin up into hurricanes (called tropical cyclones or typhoons in the western Pacific). Even the eyes of the hurricanes are visible.

The breakthrough simulation demonstrates the CCSM's ability to reproduce the behavior of these fine-scale features with remarkable fidelity. For the experiment, high-resolution configurations of the atmospheric (CAM) and land-surface components (CLM) of the CCSM were free to evolve over time, while specifications of sea-ice extent and sea-surface temperatures came from observations made in 1997.

Since climate models are used to explore the statistical properties of the climate system, not day-to-day weather, the simulation illustrated by this animation is not intended to recreate specific events. It's noteworthy, however, that two super typhoons actually formed in the Pacific during August 1997.

Data points for the atmospheric model were about 37 kilometers (23 miles) apart on a three-dimensional grid, compared with a resolution of about 150 km (93 mi) for CCSM simulations prepared for the Intergovernmental Panel on Climate Change Fourth Assessment Report. (Image courtesy Earth and Sun Systems Laboratory, NCAR.)