Let’s Talk About The Weather

Tall, grinning Joe Klemp is one of these real weathermen. He’s spent 15 years developing the base code for the atmosphere of a simulated earth. His colleague George Bryan, with wide, quizzical eyes, models clouds. And bald, blue-eyed Greg Thompson? He makes rain – working on the particular challenge known as “microphysics.”

On a Tuesday this June, the three of them – and 200 or so other meteorologists – gathered in the sunlit atrium’of the National Center for Atmospheric Research in Boulder, Colorado, to talk about the weather. The four-day event was a kind of Davos for weather prediction: Discussed were both grand ideas about how it works, and more mundane concerns, like if the afternoon’s thunderstorms would delay their flights.

What the group had in common was their use of the Weather Research and Forecasting Model, commonly known’as the WRF, or the “worf,” an opensource digital model that meteorologists and physicists have been collectively building for the last 15 years. It is used by folks as varied as UCLA students (studying fog in California’s Central Valley) and the Air Force (customizing a version to plan missions). Back in 2006, space physicists even modified the WRF to model the atmosphere of Mars.

“We make it available as an open system, and we put no restrictions on who can get it or what they can do with it,” Klemp says, just before ringing a brass bell to start the morning session.

[mf_mosaic_container columns=”2″ captions=”yes”]
[mf_mosaic_item src=”https://modernfarmer.com/wp-content/uploads/2014/09/weatherman1.jpg” number=”1″ caption=”Chao Sun”]
[mf_mosaic_item src=”https://modernfarmer.com/wp-content/uploads/2014/09/weatherman2.jpg” number=”2″ caption=”Greg Thompson”]
[/mf_mosaic_container]

Weather models are computer programs that create a mirror of Earth’s atmosphere and then do a mindboggling number of calculations to run it forward in time. They dominate the world’s weather prediction, acting as the first source for forecasters around the world waiting eagerly for the release of each “run” online – and who, in some cases, pay dearly for the data.

Most models are developed and operated by big government bureaus or private companies. For example, the National Weather Service’s Global Forecasting System and the European Center for Medium Range Weather Forecasting’s model run on finely tuned (and fantastically expensive) supercomputers. These systems are tightly controlled – primarily to ensure that any improvements are, in fact, improvements. But that high protective fence makes it hard for researchers to get inside to tinker. The WRF is the weather predicting community’s attempt to build something collaborative outside those walls. Think of it as Linux to the government’s Microsoft.

In Boulder, the mood was more family reunion than formal academia. There were no sponsors’ tables brimming with swag, only simple clip-on badges, and a meeting organizer who greeted most of the attendees by name. The meteorologists came from 22 countries, but were almost all dressed in lab-casual, shorts and short-sleeved shirts. They talked a surreal kind of shop: “If there’s upward motion and it gets saturated, I’m going to make a cloud,” one says.

There are television weathermen, and then there are the real men and women behind your weather forecasts.

But beneath the backslapping (many scientists in this group have worked together for years) there’s a subtle competition for research dollars and bragging rights afoot. Like astronomers name stars, meteorologists can stake their claim on particular predictive algorithms, or “schemes.” At the afternoon poster session, for instance, Thompson (the rainmaker) circulated among the scientists, intently chatting with the young researchers about their experiments – especially if they’d made use of his work. For a decade, Thompson and colleagues have been tweaking the Thompson Microphysics Scheme to dictate when and how, inside the model, and hopefully out here on earth, water drops will freeze or ice crystals will grow. “You want to know if it’s going to rain or snow today,” he says.

The biggest buzz was about the change coming this fall. The first morning of the workshop, Steve Weygandt, a meteorologist at Boulder’s Earth System Research Laboratory, announced the imminent switch-on of the High-Resolution Rapid Refresh model (the “HRRR”), a version of the WRF designed to give a close-up (or “high resolution”) view of the weather in 15-minute increments. Significantly, it will run on the National Weather Service’s supercomputer, and become a key part of all forecasts in the continental U.S. One of the most noteworthy improvements the HRRR will offer is the ability to more accurately and quickly refresh its forecasts of severe storms.

[mf_mosaic_container columns=”2″ captions=”yes”]
[mf_mosaic_item src=”https://modernfarmer.com/wp-content/uploads/2014/09/weatherman3.jpg” number=”1″ caption=”Joe Klemp”]
[mf_mosaic_item src=”https://modernfarmer.com/wp-content/uploads/2014/09/weatherman4.jpg” number=”2″ caption=”Stan Benjamin”]
[/mf_mosaic_container]

“It’s a big honker,” says National Oceanic and Atmospheric Administration‘s Stan Benjamin, who’s been responsible for the HRRR’s implementation. “It’s something the community has really been asking for,” especially airlines and utility companies, who have been eager for its help in dispatching planes and planning power needs.

For many, the model’s graduation from “research to operations” (a favorite phrase of the attendees) was personal. For example, the HRRR features the aforementioned Thompsonmicrophysics scheme, and draws on information from the Grell-Freitas convection scheme of the weather power couple, Evelyn and George Grell, who were holding court at the “Information Exchange” (the tongue-in-cheek and funder-friendly word for party).

The number of WRF users has grown every year, as meteorologists struggle to comprehend and predict the weather of our changing planet. But the group is always aware of the simulation model’s limits. “There’s a real benefit in just being able to take a step back and say ‘OK, if we had all this to do over again, what would we do?’” Klemp muses. “Every now and then, you have to start with a clean sheet of paper.”

[mf_editorial_break layout=”twocolmedia” title=”Real Time Weather” image=”https://modernfarmer.com/wp-content/uploads/2014/09/weatherpic.jpg”]

Ever wonder where your weather forecast comes from? Head on over to the Weather Research and Forecasting – WRF, or “worf” – Model website (wrf-model.org). Here, on this open-source weather prediction system designed, according to their website, “to serve both atmospheric research and operational forecasting needs,” you can find real-time weather predictions. Pictured, for instance, is the severe storm potential predicted for July 25 and 26, 2014, live when this magazine went to press. Other options include predicted precipitation and wind.

The WRF is a result of more than 20,000 scientists and researchers from over 130 countries around the world who, using data both real and simulated, predict the weather.

[/mf_editorial_break]