The science behind the storms

UNIVERSITY PARK, Pa. — On their first day of a six-week campaign that began in May, hail researchers from Penn State followed expert forecasts that took them to the plains in central Texas, south of San Angelo. They parked their instrumented, armored Chevy Tahoe near a U.S. highway, set up an array of instruments and retreated to the vehicle for safety as the forecasted marbles of hail began to fall. The vehicle, named Novel Atmospheric Research on High-impact Weather Applications Laboratory, or NARHWAL, was converted to withstand damage from hail.

Soon — to their surprise — the chunks of ice grew larger and so did the crosswinds, gusting up to 50 miles per hour, firing three-inch hail stones — like buckshot — under the hail protection cage, breaking the vehicle’s front windshield.

After that first experience, the field team — including two Penn State faculty and three students — regrouped, repaired NARHWAL and again took to the field during the first U.S. hail research campaign in four decades, named the In-situ Collaborative Experiment for the Collection of Hail In the Plains, or ICECHIP. It was the largest, most sophisticated international collaborative hail research campaign to date.

As another storm picked up, the team launched balloons equipped with sensors into the storm. A vehicle-mounted Mesonet-style instrument mast recorded temperature, relative humidity, wind speed and direction, and pressure. A roof-mount probe captured hail size, frequency and impact forces. Another team out of Australia launched Hailsondes — hail-sized styrofoam GPS sensors — hoping their imposters would join the others, so they could track the physics at play during a hailstorm. Coolers were ready to gather stones bound for analysis after they fell, including physical measurements and examining their internal layered structures to understand their growth history. The scientists angled NARHWAL so that the armor would protect the windshield. They were ready.

And then it happened again.

“All of a sudden the storm started getting really intense,” said Kelly Lombardo, associate professor of meteorology and atmospheric science at Penn State and co-principal investigator on the study. “I think there was a lot of frustration, because we’re like, ‘again, you’re kidding me.’ But then there was also joy that we had exactly what we were trying to capture with all the instruments in place, which was fantastic.”

Damaging storms

A chief focus of the campaign is determining when hailstorms can change from a nuisance to a threat, said Matthew Kumjian, professor of meteorology at atmospheric science at Penn State and principal investigator who also serves on the ICECHIP’s steering committee.

One thing that came as no surprise to the researchers was hail’s ability to inflict damage. The Insurance Institute for Business & Home Safety (IBHS) estimates hail damage in the U.S. costs tens of billions of dollars each year, including a record-setting $60 billion in damage to cars, crops, roofs and more in 2023. The U.S. National Science Foundation (NSF) approved nearly $11 million for the three-year field campaign that includes nine teams with roughly 70 researchers in total from four countries. Two years of analyses will follow this year’s field campaign. IBHS also contributed funding, personnel and instruments to the project.

A thematic approach

ICECHIP was designed to improve hail detection, modeling and forecasting by collecting ground-truth observations of hail size, structure, and impact and addresses five research themes: hailstone growth and fall behavior, improving hail trajectory models, improving hail forecasting, how environmental factors influence hail production, and linking hailstone growth and damage to radar observations. Penn State’s team is working on all five themes.

Kumjian said technological advancements are going to drive discovery. Things like high-resolution video drones, Hailsondes, impact-recording probes — sampling at a rate of over 1,000 times per second — and mobile radars didn’t exist during the last campaign in the 1970s. Advances in radar technology, computer modeling, and other tools have also improved.

Radar is good at locating hail, he said, but it’s not as good at indicating expected hail damage. One avenue the team is exploring is the hail kinetic energy flux from a storm. If a radar tells us a storm is producing a few larger hailstones, it’s not as useful as telling us about the frequency of smaller hailstones, which can be more damaging due to their sheer volume.

“We’ve been stagnant in this area, and ICECHIP can potentially give us the data we need for a breakthrough,” Kumjian said. “If we can better predict and warn for a hailstorm’s damage potential, that will be a big advancement for science and the public.”

Another goal, Kumjian said, is chipping away at the massive cost of damage caused by hail. A better understanding of hail physics can lead to more resilient structures. Additionally, partners at IBHS deployed different roofing materials in storms to test how well they stand up to the barrage of ice.

The project is already gaining interest outside of the science community. An upcoming conference featuring the research is expected to be attended by emergency managers, forecasters, private sector insurance companies, contractors and others.

The next step, improving the predictability problem, will involve two data assimilation experts at Penn State: Yunji Zhang, assistant professor of meteorology and atmospheric science and assistant director of the Penn State Center for Advanced Data Assimilation and Predictability Techniques, and co-principal investigator, and Xingchao Chen, assistant professor of meteorology and atmospheric science and co-principal investigator. They will use ICECHIP observations to improve how computer models represent and predict hailstorms.

"We want to learn the storm scale processes that change the hail development characteristics," Lombardo said. "We want to be able to better predict when benign hail will suddenly shift into a real threat."

The student experience

Some of the data scientists will be analyzing is from Allie Scarlett, a meteorology and atmospheric science student at Penn State and one of 32 undergraduate students tapped to work on ICECHIP.

Twenty graduate students, including doctoral candidate Lydia Spychalla, also worked in the field.

For six weeks, Scarlett worked in various areas but mostly on the C-Band on Wheels, a semi-mobile radar truck called CoW for short. Scarlett also deployed SuMHOs — or Super Mobile Hail Observatories — which are large vessels developed by the team’s Canadian colleagues that funnel and capture hail stones for later research. Those hailstones, some measuring a couple inches in width, were shipped off to the NSF’s National Center for Atmospheric Research for analysis. Another team captured a stone nearly six inches in diameter.

Because the CoW operates with a longer wavelength than most mobile radars, the large dish makes setup time consuming. That meant Scarlett was often positioned far out of the storm path. Each morning, her team would often set up shop on a quiet crop of farmland as other teams trekked closer to the storms. One surprisingly interested party to the CoW were cows, Scarlett said, mistaking the bulky piece of scientific equipment for a feed trough.

“Working on ICECHIP really solidified my interest in research,” Scarlett said. “This campaign ground truthed a lot of the radar data we were gathering, and I’d love to be able to follow through on a project like this.”

Next steps

NARHWAL was down two windshields from hail damage, but the team is rich in new data. All told, the team deployed in 11 states, logging more than 13,000 miles. Kumjian said the field research is giving the team some of the data it needs to answer key questions hail researchers have been asking since the last campaign.

“We’re fundamentally interested in a number of questions that only observing actual storms can tell us,” Kumjian said. “ICECHIP is a once-in-a-career opportunity to get all of those pieces of the puzzle on at least a small sample of storms, so we can begin to dig into and understand the processes and how they play out in hailstorms at large.”

Editor's note: This story first appeared in the Earth and Mineral Sciences' research magazine "Impact."