- Our Facility
- User Services
The Earth is a rich and varied system, made up of countless different parts that interact with each other, sometimes in ways that seem chaotic due to their extreme complexity. Sunlight, ocean currents, vegetation, land and sea ice, weather and climate, volcanoes, and even humans are all parts of this interconnected system that is our home.
The laws of physics govern the natural world, and we can describe the interactions within the Earth system in mathematical terms. Because there are so many variables, and because we cannot gather every single bit of data for every moment, we rely on numerical simulations from computer models to help us understand how the system works. These models provide scientists with a virtual laboratory in which they can test hypotheses and conduct experiments that would not be possible any other way.
Climate is a part of the Earth system, and researchers study and model how the climate works and interacts with other parts of the Earth system. Only supercomputers are able to run the models that integrate the many parts of the Earth system, or to run the models in great detail and for long time periods. This means that as computers become more powerful, we can get a clearer picture of what the climate is doing—and how it might look in the future.
How do we know if climate models are accurate?
To verify a model's accuracy, scientists simulate past conditions and then compare the model results to the actual observations. If the models can simulate what was observed in the past, this gives us confidence that they tell us something useful about the future.
What is the difference between a climate model and a weather forecast?
Climate models cover far longer periods than weather models, and they include more components of the Earth system like oceans, vegetation and sea ice. This means they are far more complex and computer intensive. In order for researchers to simulate global climate over long time periods like years, decades, or millennia, they reduce the geographical detail—resolution—to regional or global scales, rather than local scales. Weather forecasts do the opposite, running for very short time scales, but at very high resolution.
Climate and weather models rely on a three-dimensional mesh that reaches high into the atmosphere and deep into the oceans. At regularly spaced intervals, or grid points, the models use laws of physics to simulate winds, temperature, precipitation, moisture, pollution and energy exchanges in the climate system.
How can we predict climate over future decades if the weather for next week is uncertain? Climate models are very similar to weather models, but they look at averages. A weather model tries to predict specific weather events over the next few days. A climate model cannot predict which days will be rainy, but it can predict the average number of rainy days over months, seasons or years. This is typical for complex systems: you cannot predict the details far into the future, but you can predict the general trend.
Observations and climate models both show that the most dramatic warming is in the Arctic region. Since snow and ice are highly reflective surfaces, they absorb very little of the Sun’s heat. When they melt, they expose the much darker underlying ground or ocean, which then absorbs most of that heat. This leads to warmer temperatures, which in turn melts more snow and ice. This is known as a feedback mechanism.
Learn more about climate change research at:
The NWSC is the result of a partnership between University Corporation for Atmospheric Research (UCAR), the State of Wyoming, the University of Wyoming, Cheyenne LEADS, Wyoming Business Council, Cheyenne Light, Fuel & Power Company. The NWSC is operated by the National Center for Atmospheric Research (NCAR) under the sponsorship of the National Science Foundation.
The NWSC’s first supercomputer is called Yellowstone, named after the world’s first established national park, and in honor of Wyoming’s important role in making the NWSC a reality. This machine is capable of 1.5 quadrillion calculations per second—or, in computing terms, 1.5 petaflops. “FLOPS” stands for “floating point operations per second.”