Atmospheric and Climate Science Directorate

[Contents] [PDF] [Previous] [Next]

A Community Approach to Building a Better Meteorological Model

The Meso-scale Compressible Model Mc2:

In 1994, researchers at the Meteorological Service of Canada MSC undertook the development of a new type of meteorological model, the Meso-scale Compressible Community model Mc2. “Community” in this context reflects the collaboration of MSC scientists with others in Canada and internationally to develop the world's fastest, most efficient meso-scale to local scale numerical model. Applications for this model include improving wind forecasts in mountainous terrains, driving wind power models, and generating air quality forecasts.

Because the source code for MC2 was made available to other researchers (known as “open source code”), MSC scientists have effectively leveraged their intellectual investment. Over 50 researchers around the world, in universities and national labs, have adopted MC2 and developed enhancements for it.

This model has been designed to take full advantage of the power of parallel processors - meaning that the calculations of atmospheric dynamics that form the basis for all subsequent calculations can be run, in parallel, on any number of processors. There are plans to run MC2 on Earth Simulator, a massively parallel computer system in Japan, which is 64 times more powerful than the supercomputer currently used for Canada's daily weather forecasts. The goal of the exercise is to produce an unprecedented high-resolution simulation of one of nature's most powerful weather phenomena – a hurricane. Norway uses MC2 to forecast winds at airports – an extremely challenging task because of the wide variations in wind speed and direction in their mountainous terrain. The biggest challenge for MC2 was the head-to-head competition with several European models in one of the most challenging places to forecast weather on the globe – the Alps. The combination of mountainous terrain and the moisture of the Mediterranean Sea, make this area too difficult for the current global and regional weather forecasting models to handle well, and an exciting challenge even to meso-scale models. Sophisticated airborne and ground measurements were used to evaluate the performance of the models. In this competition, MC2 produced unprecedented realism, especially for intricate low-level wind flow patterns.

MC2 is also used to generate air quality forecasts. Researchers in Canada and elsewhere begin with MC2, and then add atmospheric chemistry calculations to predict the transport of smog. All the experience gained in developing and applying MC2 worldwide is now being applied in Canada to improve wind forecasts in mountainous terrains, to drive wind power models, and to generate the air quality forecasts delivered jointly by Environment Canada and the provinces. This “community” approach led by MSC is a very effective way to access the scientific knowledge of the entire atmospheric science community.

The community model MC2 run over Vancouver Island at 2km horizontal resolution: 17H forecast valid 26 June 1997 2000 UTC. Near surface flow (arrows with scale in knots in lower left corner) superimposed over topography (gray shades every 500m). Only one arrow for every other grid point is displayed for each direction.

“The MC2 model is one of the few models available that allows conducting real-time high-resolution numerical weather prediction in a cloud-resolving mode.”
Christoph Schär
Institute of Atmospheric and Climate Science
ETH Zürich, Switzerland

The Green Lane^TM,
Environment Canada's World Wide Web Site.