© 2024 Milwaukee Public Media is a service of UW-Milwaukee's College of Letters & Science
Play Live Radio
Next Up:
0:00
0:00
0:00 0:00
Available On Air Stations
WUWM’s Chuck Quirmbach reports on innovation in southeastern Wisconsin.

Wisconsin Meteorologists, Supercomputer Team Up To Improve Weather Forecasts

Chuck Quirmbach
Blustery conditions at midday Thursday prompted Riverwalk lights to be on, and flags to blow.

Typically, meteorologists can accurately predict what the weather will be like within a week. Beyond that time frame, the weather is too variable for forecasters to be all that confident.

But a recent federal law may bring some help to farmers, other business owners and anybody else who counts on reliable mid-or longer-range predictions —and a large Wisconsin computer may be part of the improvements. 

Weather forecasters make their predictions sound simple. But, of course, it’s not that simple.

Paul Roebber, professor of atmospheric sciences at the UW-Milwaukee, directs the Innovative Weather program that brings you forecasts on WUWM. He says the ease of forecasting varies.

“There are some days where a 24-hour forecast may be problematic. There are other days where you can see well out to 10 days," Roebber says.

One of the biggest problems, he says, is lack of data from the part of the globe where our weather fronts often start.

"Here, in the Midwest of the United States, if you think about a weather system that's going to arrive in 48 hours, it's probably coming west to east. It's probably going to be coming from the Pacific Ocean, so there's a data void right there," he says.

Roebber says weather balloons that collect information on atmospheric pressure, temperature, humidity and wind speed are less common over the ocean. Aircraft in flight can help collect additional data, and so can the increasing use of satellites. But there are still gaps, especially at some altitude levels.

Credit Computed in Fractint by Wikimol
A graph of the concept known as the Lorenz Butterfly.

Brad Pierce, who directs the Space Science and Engineering Center at the UW-Madison, says you could also try to understand the difficulty by considering the mathematical models and resulting graphs that became known as the Lorenz Butterfly.  

"Basically, the atmosphere is a very complex, very non-linear system, and very small changes in the current state of the atmosphere tend to grow and lead to uncertainty in the forecast. So, Edward Lorenz, who was actually a meteorologist, coined the phrase the Lorenz Butterfly — in that a butterfly in the Amazon might affect the weather over Wisconsin," he explains.

Pierce is optimistic that Congress has taken a step toward improving intense storm, mid-range and seasonal forecasts by passing the Weather Research and Forecasting and Innovation Act of 2017.  He says the aim is to bring a more unified prediction and weather observing system.

Plans are for a large machine in the Madison campus' space science and engineering building to help with some of that. It’s the supercomputer known as the S4.

Credit Chuck Quirmbach
Brad Pierce, director of the UW-Madison Space Science and Engineering Center, stands next to the S4 supercomputer.

"You're looking at the computing nodes, individual processors just like you have on your laptop, except that we've got thousands and thousands of them. You're looking at a bunch of wires. You have to connect each one of those nodes, so those can communicate with the other ones, very fast," he explains about the supercomputer.

The National Oceanic and Atmospheric Administration (NOAA) is paying for an upgrade to the supercomputer.

Pierce says the unit can bring together a lot of weather information and forecast models. "So that we can have all the satellite data we need to help constrain the forecast and all the forecast products that are spit out by the model. That requires all sorts of disc storage, so we can start looking at the results," he says.

Pierce says there are petabytes of data storage. Each petabyte is a million gigabytes.By contrast, many smart phones now have 64 gigabytes.

The aim, he says, is for the academic community to partner with the private sector to come up with better forecasts for users like the aviation industry or the farmer who wants to know when to plant, water and harvest. 

Credit Chuck Quirmbach
The Data Center at the UW-Madison Space Science and Engineering Center.

In late November, a key figure in the implementation of the Weather Research Act held a listening session in Madison. The price tag for the law isn't quite clear. But Neil Jacobs, an assistant secretary in the Commerce Department and Deputy Administrator of NOAA, says the goal is to get a good return on the investment.

"We'll have to watch the output for a year or two before you start to see any substantial improvements. But we're upgrading to a new dynamic core in our modeling system next year. That's going to be the first of a series of major upgrades," he says.

UWM's Paul Roebber says forecast improvements are likely to become more important during the more frequent extreme weather events expected under climate change.

Support is provided by Dr. Lawrence and Mrs. Hannah Goodman for Innovation reporting.

Do you have a question about innovation in Wisconsin that you'd like WUWM's Chuck Quirmbach to explore? Submit it below.

_

Related Content