“Once in a Generation”: The 2022 Buffalo Blizzard

Truck in snowdrift

A Research Spotlight from 32WAF/28NWP/20Meso

On 23 December, 2022, David Zaff of the National Weather Service’s Buffalo office walked out into a blank white world of howling wind. He headed to his car to get supplies, knowing there was no way to get home. He and his coworkers were trapped at the office, in the middle of one of the most deadly and disastrous blizzards Buffalo has ever seen.

Video by David Zaff, showing whiteout conditions outside NWS Buffalo office, December 23, 2022.

At the height of the 2022 holiday travel season, the four-day blizzard and lake-effect snow event knocked out power for more than 100,000 people, paralyzed emergency services and holiday travel, and left at least 47 dead. New York Governor Kathy Hochul described it as “the most devastating storm in Buffalo’s long, storied history.” Yet days earlier, Zaff and colleagues encountered skepticism from the public as they worked to warn the region.

Presenting at the J3 Joint Session at the 32nd Conference on Weather Analysis and Forecasting, the 20th Conference on Mesoscale Processes, and the 28th Conference on Numerical Weather Prediction, Zaff talked about the disaster and how the NWS countered accusations of hyperbole to get the word out.

Sounding the Alarm

The December 2022 snow was shocking, but not surprising. The pattern was easy enough to recognize, even 7–10 days earlier: a large high-pressure ridge forming over the western U.S., with a major trough in the east. “We knew something big was coming,” said Zaff. Five days before the storm, even low-resolution models suggested a major event. Four days ahead, the NWS started ringing the alarm bell. “We started saying, ‘A powerful storm will impact the region heading into the holiday weekend.’”

Three days out, the NWS issued an unusually emphatic Area Forecast Discussion (AFD):

“Some of the parameters of this intense storm are forecast to be climatologically ‘off the charts’ … One could certainly describe this storm system as a once in a generation type of event.”

NWS Lead Forecaster Robert Hamilton, Tuesday, December 20, 2022

That caused a stir, but many on social media dismissed it as hype. “People started saying, ‘There goes the weather service again,’” says Zaff.

He tried to find a way to show the science graphically, highlighting the forecast as “‘outside’ the climatology” for the time of year.

The graphic and its accompanying description got attention. By then, NWS Buffalo was communicating in earnest, including on social media. A tweet with a text-filled screengrab of the Winter Weather Message received 485,000 views. “A picture is worth a thousand words,” Zaff said, “except when people actually read the words, and see how impressive this event might be.”

Left: Graphic showing forecast surface pressure for Friday, December 23, 2022, with shading showing the relative frequency of the forecast MSLP values in the Buffalo region at that time of year. Source: David Zaff.

Surviving the Storm

Before noon on 23 December, visibility dropped to near zero, and it remained that way until around midnight on 25 December. 500 Millibar heights were “extraordinary” as the pressure trough moved into the Ohio Valley, and surface-level pressure was similarly unbelievable. A top wind speed of 79 mph was measured in downtown Buffalo at 10:10 a.m. on the 23rd, and winds in the 60–70 mph range lasted for 12 hours. “[It was] just an incredible bomb cyclone,” Zaff said. “An incredible storm.”

Zaff and some colleagues slept at the office; others attempted to drive in whiteout conditions using GPS alone, while some got stuck in drifts near the office and had to leave their cars to hike the rest of the way. Meanwhile, firefighters and airport employees worked to rescue motorists trapped nearby.

On December 24, the City of Buffalo issued “the scariest tweet I’ve ever seen,” said Zaff. The tweet stated that there were “no emergency services available” for Buffalo and numerous other towns.

“We knew by this time that there were fatalities occurring,” Zaff said. “And it just got worse and worse.”

Blizzard conditions lasted a full 37 hours–and lake effect snow wouldn’t stop for another two days. Three power substations shut down, frozen solid. Hundreds of power poles fell, and a significant percentage of locals were without power during the storm’s peak (some for days afterwards).

The 47 fatalities included people stranded outside, others who died from hypothermia in their homes, and some deaths due to delayed EMS response, according to Erie County. Hundreds of motorists were stranded on roadways during the storm. The Buffalo Niagara International Airport, with a proud legacy of operating under even the most horrific conditions, was closed for six days.

Zaff didn’t return home until late afternoon on the 25th, 18 hours after official blizzard conditions were over and having clocked 50+ hours at the office. On the drive, he saw iced-over buildings and trucks buried in snowdrifts. “It reminded me of [the movie] The Day After Tomorrow. … The impacts were tremendous.”

In his AMS presentation, Zaff compared the 2022 event to disastrous storms in 1977 (20+ fatalities, 69 mph winds, only 12” of snow yet drifts swallowed homes) and 1985 (5 fatalities, 53 mph winds, 33” snow), as well as the “Great Christmas Storm” of 1878, one of the first well-documented lake effect snow events, though lake-effect processes weren’t understood at the time. “This will likely be the storm of comparison now,” he says. “Once-in-a-generation” turned out to be right.

Future Lessons

Moving forward, said Zaff later, “Our intention is to further our relations with our Core Partners, including elected officials, emergency management, and the media [and] provide more probabilistic information that supports our ongoing Impact Decision Support Services. We hope to improve our outreach as well, instilling more confidence with the public.”

NWS will continue to provide improved decision support for partners, which may lead to more proactive road and school closures that could save lives in the future.

Photo at top: Buffalo roadways at 4 p.m. on December 25, 2022, 18 hours after blizzard conditions had passed. Photo credit: David Zaff.

About 32WAF/20Meso/28NWP

Predicting and understanding storms and other weather events is a complex business with real-world impacts. The American Meteorological Society’s 32nd Conference on Weather Analysis and Forecasting/28th Conference on Numerical Weather Prediction/20th Conference on Mesoscale Processes brought researchers, forecasters, emergency managers, and more together to learn about and discuss the latest scientific developments. The conferences took place in Madison, WI, and online 17–21 July, 2023. Recordings of the sessions are available here.

Melting Snow and Other Surprises on Mt. Everest

An AWS being installed. Note the tents of Camp IV in the background, and the exposed glacier ice visible behind. Photo credit: Baker Perry / National Geographic.
An automatic weather station (AWS) being installed on Everest’s South Col at 7,945 m (~26,066 ft). Note the tents of Camp IV in the background, and the exposed glacier ice visible behind. [Photo credit: Baker Perry / National Geographic.]


Despite freezing temperatures, snow is melting on Mount Everest. That’s just one finding in a recent study of weather data provided by a new network of five automated weather stations on Earth’s tallest mountain. The network includes two of the highest altitude weather stations on Earth, Balcony Station at 8,430 m (~27,658 ft) and South Col at 7,945 m (~26,066 ft), and offers “an unrivaled natural platform for measuring ongoing climate change across the full elevation range of Asia’s water towers,” Tom Matthews and his colleagues write in their new article published as an Early Online Release in the Bulletin of the American Meteorological Society.

Photos of the automatic weather stations installed during the 2019 Everest Expedition. Note the shovel handles used to mount the wind speed sensors on the Balcony weather station (upper right).
Photos of the automatic weather stations installed during the 2019 Everest Expedition. Note the shovel handles used to mount the wind speed sensors on the Balcony weather station (upper right).


The snowmelt is attributed to extreme insolation in the high altitudes of the Himalaya. It enables “considerable” melt up to Camp II at an altitude of 6,464 m (~21,207 ft), “despite freezing air temperatures,” the study reports. And modeling with the data the five stations are providing shows not only is melting occurring at South Col even with average air temperatures of -10°C—which means melting may be common at the tops of all but a small portion of the peaks in the Himalaya—but also is likely happening even at Everest’s peak, Matthews and his team report.

Uncertainties in the extrapolation are considerable, but we cannot rule out that limited melting during the monsoon may be occurring at the summit.

The authors note that while snow melting at the peak of the world’s tallest mountain may be “symbolic” as Earth continues to warm, sublimation of the snowpack appears to be a far greater contributor to its loss at such high altitudes. This finding has implications for the amount of snow that actually falls at extreme altitudes:

The amount of mass potentially lost by sublimation on the upper slopes of Everest, coupled with the presence of permanent snow cover over much of this terrain, raises the interesting prospect that snowfall at such altitudes in the Himalaya may be more substantial than previously thought. For example, the modeled sublimation of 128 mm at the South Col (in five months) is almost eight times greater than the predicted annual precipitation at such altitude. Windblown snow from lower elevations may account for much of the discrepancy, but the winds are also known to deflate the snow on Everest, sometimes to spectacular effect. Future work is clearly needed to rule out the possibility of a much more vigorous hydrological cycle at these extreme elevations.

Matthews and his coauthors conclude that the data the five AWSs have collected so far offer “rich opportunities” to adjust and improve mountain weather forecasting and melt modeling.