UCAR Staff Notes masthead
Home Our Organization Research News Center Education Community Tools Libraries
About Staff Notes
Past Issues
Favorite Photos
How to Subscribe


staff notes header

July-August 2006

A new eye on storm formation

REFRACTT tests out radar technique on the Front Range

Wilson and Roberts
Jim Wilson (left) and Rita Roberts discuss refractivity data.

People attending picnics, weddings, and ball games in the 2010s could get better advance notice of showers and thunderstorms than they do now, thanks to an observing strategy being tested by EOL and RAL. A field project in its second summer has put four Doppler radars to work in a new way, tracking low-level moisture across northeast Colorado in unprecedented detail.

RAL's Rita Roberts is the lead principal investigator for REFRACTT, the Refractivity Experiment For H2O Research And Collaborative operational Technology Transfer. The other PIs include Frederic Fabry (McGill University), Jim Wilson (EOL/RAL), Scott Ellis and Tammy Weckwerth (EOL), and Pat Kennedy and V. Chandrasekar (Colorado State University). The team hopes to show that tracking the pools of atmospheric moisture that slosh back and forth just above Earth's surface will help pin down the locations where storms might rage afew minutes to a few hours later. Up to now, the sparse network of surface weather stations has limited forecasters' ability to map these fields of moisture and predict exactly where heavy rains might develop.

Frederic hatched the scheme now being tested in REFRACTT while he was an ASP postdoctoral researcher in the mid-1990s. His idea was to measure the changes in the speed of radar signals, also known as refractivity, caused by changes in moisture and density of the air. To calibrate the refractivity measurement, it's necessary to point the radar at a fixed target, such as a silo or power line that would normally contaminate the radar map. Frederic and his colleagues at NCAR and McGill developed a way to measure the refractivity and extract moisture data from it. The result is a map much like any other radar display, but one that shows water vapor instead of blobs of rainfall or bundles of wind.

NCAR tested the refractivity approach during field projects in 2000 and 2002 using the transportable Doppler radar called S-Pol (now stationed at the Marshall Field Site just south of Boulder). REFRACTT is a larger study designed to show more definitively that the technique is worth adding to the national NEXRAD network of Doppler radars operated by the National Weather Service (NWS). If so, the method would benefit the hundreds of forecasters who monitor thunderstorms and warn the public about them.

After the 2002 results were analyzed, says Rita, "The technique looked so promising that we all thought, 'We'd better get this on the NEXRADs.' That was the real motive for the REFRACTT project."

REFRACTT is pulling data this summer from S-Pol as well as other radars across the Front Range operated by CSU, the University of Oklahoma, and the Denver-Boulder NWS office. Over the small region where the radar domains intersect, just northeast of Denver, the data can be easily crosschecked.

"Last year we demonstrated in post-processing that we could get refractivity fields from all the radars and get good results," explains Rita. "This year we're running multiple radars, 'mosaicing' the refractivity data in real time, and putting them in a display that forecasters can use."

Another PI, John Braun (COSMIC), is examining moisture measurements from radiometers and from ground-based sensors that intercept Global Positioning System signals and calculate moisture along slanted vertical paths. In addition to running S-Pol, EOL's radar technicians (Al Phinney, Mike Strong, Kyle Holden, Jonathan Emmett, Laura Tudor, and Brian Pereira) are launching radiosondes on either side of moisture boundaries using the Mobile GPS Advanced Upper-Air Sounding System (MGAUS). Behind the scenes, RAL's Eric Nelson has been providing daily weather forecasts, archiving data, and plotting and analyzing soundings.

“Nobody’s ever seen such high-resolution data on moisture before. There’s a lot for us to learn.”

—Rita Roberts

Although rainfall and high winds are what radar operators normally look for, REFRACTT needs standing objects such as buildings in order to analyze refraction in the radar signal. The lowest beam of a standard weather radar skims 0.5° above the horizon, which means that refractivity and moisture can be measured only out to about 30 miles before the beam overshoots even the tallest objects. Because of this limit, the refractivity technique can only provide data within a couple of counties of each radar. But many NWS radars are positioned close to the urban areas where the majority of Americans work and play (and dodge storms).

This summer hasn't yielded quite as much severe weather as usual across northeast Colorado, but Rita and her colleagues are nonetheless intrigued by what's showing up on the moisture display, which was designed by RAL's Frank Hage. During the hundred-degree heat of early June, bright patches of crimson and orange–denoting areas of extremely low moisture–flowed off the mountains on hot, dry west winds. In contrast, the monsoon of early July brought surges of green (high moisture) westward and upslope.

These general patterns are well known to local forecasters, but the details are new to everyone. That's because surface weather stations are separated by 30 miles or more, and vertical moisture profiles produced by twice-a-day radiosonde launches are typically separated by hundreds of miles. "Nobody's ever seen such high-resolution data on moisture before. There's a lot for us to learn," says Rita. One aspect of interest is how low-level patterns of water vapor are influenced by evaporation from wet fields and green vegetation.

NCAR's S-Pol radar gathers refractivity measurements from the Marshall Field Site south of Boulder as part of the REFRACTT project. (Photos by Carlye Cavin, ©UCAR.)

This fall the REFRACTT scientists will analyze the data in depth and make their pitch to the NWS. But the project is already a success in another sense. In order to make REFRACTT possible, Rita and colleagues convinced the NWS to allow an external connection to its Level I radar data—the rawest and most detailed form. Until now, it's been difficult for researchers to access Level I data due to security concerns. The approval "is a really major thing," says Rita. "It's going to open up a lot of doors." RAL and EOL engineers Nancy Rehak, Mike Dixon, Dennis Flanigan, and Tres Hofmeister have been responsible for accessing the Level I data in real time and computing the refractivity fields.

In EOL, a team led by John Hubbert has been helping the NWS improve the quality of NEXRAD data since the mid-1990s. The team, which includes Cathy Kessinger, Scott Ellis, Mike Dixon, Greg Meymaris, and Joe VanAndel, has worked to mitigate unwanted signals that can interfere with tracking rain and high winds—including, ironically, the "ground clutter" that's now being used by REFRACTT to measure moisture.

If the idea behind REFRACTT sounds vaguely familiar, it has a lot in common with COSMIC, the Constellation Observing System for Meteorology, Ionosphere and Climate, which was successfully launched this year (see the September 2004 issue of Staff Notes Monthly). Both systems measure refractivity and infer moisture from it. The key difference is that COSMIC takes vertical soundings throughout the lower atmosphere, using GPS signals, while the REFRACTT radars produce a much denser but smaller-scale horizontal map of moisture at very low altitudes. In both cases, scientists at UCAR and elsewhere have found a way to squeeze useful data out of once-annoying errors—a satisfying batch of lemonade, brewed from numerical lemons.

by Bob Henson

On the Web

For more information on REFRACTT

In this issue...

A new eye on storm formation

NCAR names five new senior scientists

New Delphi coordinator takes over

SOARS and RESESS protégés busy with research

Less load for the landfill: FREEcycle stores; ACD's "green" move

Short Takes

Delphi Question: Public events in the FL0 courtyard

Just One Look


Staff Notes home page | News Center