BEEKMANTOWN — Clinton County experienced its first recorded tornado in 21 years last Tuesday. There were no injuries or deaths, though several structures sustained damage.

The National Weather Service’s Burlington, Vermont office, which serves northeastern New York, confirmed that an EF-0 tornado, with winds between 65 and 75 mph, touched down at 3:38 p.m. in a wooded area west of Durand Road, about 0.8 miles southwest of the Beekmantown Central School District. The GPS coordinates of its starting location are 44.763/-73.497, according to NWS Burlington.

The tornado moved northeast, crossing Durand Road and state Route 22 before dissipating at 3:40 p.m. The GPS coordinates of its ending location are 44.767/-73.488. The tornado’s path length was about 0.5 miles and its maximum width was 100 yards, according to NWS Burlington. Damage included trees that were uprooted and snapped, a roof that was partially ripped apart — with the detached portion getting wrapped around a nearby powerline — and a trampoline were uplifted and flung from their previous locations.

NWS Burlington was able to confirm these figures through a Damage Survey, which took place that evening.

It’s something that National Weather Service offices are equipped to do when a tornado or other severe weather event is suspected to have occurred in its service area. It involves a team of meteorologists on the ground looking for specific visual indicators that can help to confirm what type of weather event took place. NWS Burlington Lead Meteorologist Maureen Hastings said the office will typically send a team out when meteorological tools, such as radar, as well as reports from people on the ground, seem to indicate that there was a strong possibility of a tornado having occurred.

“We need to have reasonably decent suspicion that there’s been some kind of tornado,” she said. “Whether we see a (wind rotation) signature on radar or some kind of debris signature, we usually have that collaborated with some kind of ground truth. So, we start seeing reports of some kind of damage in the area where we saw the radar signatures.”

When surveying for a possible tornado, meteorologists will look to see if trees fell in a pattern. This indicates that rotational winds have occurred, as opposed to trees that all fell in the same direction, which could indicate strong straight-line winds from a powerful thunderstorm, but not necessarily a tornado.

“You want damage that’s converging,” Hastings said. “With a tornado … you’ll have trees that are kind of overlaying on top of one another or kind of leaning toward one another.”

There are also near misses in which funnel clouds can form, making it look like a tornado is occurring, but the cloud does not actually reach the ground — which must occur in order the storm to be considered a tornado. These situations can be confusing, often requiring these surveys to definitely establish whether or not a tornado took place.

“Just seeing pictures of a funnel in particular, if we can’t see if it’s touching the ground, we still have to go out and verify that it was torndadic damage,” she said. “Even sometimes a picture of a tornado and damage is not enough to say ‘Yes, that’s a tornado!’ We have to go out and get eyes on it.”

If it has been established that a tornado occurred, the next step is typically to determine how strong it was.

Meteorologists will also look for the significance of damage, such as how far objects were moved from their pre-tornado locations and the degree of degradation buildings sustained relative to their preexisting structural integrity. They compare it with photos from past tornadoes from around the country of varying strengths, and determine where the damage at hand seems to match closest to.

“We can compare it to damage indicators,” Hastings said. “We have different levels. So, if there’s this kind of roof damage to this kind of structure, this is the kind of wind strength that might cause that kind of damage. Are the trees just tipped over? Are they uprooted? Are they debarked? This is what wind speed would correlate to each of those kind of damages to trees.”

NWS Burlington surveyed 41 instances of damage — both to property and natural objects, such as trees — in and around the Beekmantown tornado’s path during its damage survey before making its determination.

“Based on that wind speed is how we declare what EF level the tornado is,” Hastings said.

Beekmantown’s EF-0 tornado is the lowest level of intensity on the Enhanced Fujita Scale, named in honor of Japanese-American Meteorologist Dr. Tetsuya Theodore “Ted” Fujita, who developed the scale’s predecessor in 1971 and produced groundbreaking research on tornadoes and several other severe weather events. The EF scale has been in use since 2007, and contains several categories, based on estimated wind speeds derived from damage surveys.

EF-0 tornadoes have winds between 65 and 85 mph. EF-1 tornadoes are between 86 and 110 mph. EF-2 tornadoes are between 111 and 135 mph. EF-3 tornadoes are between 136 and 165 mph. EF-4 tornadoes are between 166 and 200 mph and EF-5 tornadoes exceed 200 mph.

There is also an EF-Unknown rating, when a tornado is confirmed but there is no damage to survey. This typically correlates to a weaker tornado, but could also be indicative of a strong tornado that occurred in a large, open and unpopulated area, more typical on the Great Plains, where there is little in the way of potential debris to damage.

A tornado’s windspeed is far from the only factor that can play a role in its impact on life and property. Tornadoes that occur in populated areas or at nighttime — which are less common they daytime twisters — tend to be particularly devastating, even if they are less powerful. A tornado’s duration is also important here. While most only last a few minutes, some can twist for more than an hour on rare occasions, or dissipate and redevelop as the storm cell progresses along.

–

The bigger picture

–

New York state broke its yearly tornado count last year, with 32 confirmed tornadoes, topping the previous record of 25 set in 1992. They were all EF-2 or less in intensity. Of the 32 last year, 17 occurred during outbreaks on July 10 and July 16.

Most occurred in central and western New York, and there were five that touched down in the southern Adirondacks. These locations included the towns of Chestertown, Wells, Lake Pleasant, Ohio and Hadley. NWS Burlington’s service area had one tornado last year. That occurred in Pikes Corner, approximately five miles southwest of Gouverneur, in St. Lawrence County. The complete list of confirmed New York tornadoes for last year can be found at tinyurl.com/3n9bep3n.

While last year’s tally was a record-breaker across the state, and the second highest overall in the U.S., it’s unclear if tornado activity overall is becoming more frequent, and if it is, if that’s attributable to human or natural causes. Tornadoes are a less reliable weather metric to evaluate climatological trends than conditions that can be measured with a fixed instrument, such as temperature or rainfall. That’s largely because tornadoes often occur in such fleeting and unpredictable instances, and usually require human confirmation.

This partially explains why overall tornado reports have increased in recent decades: there is greater radar and meteorological coverage over a wider area, especially in rural areas where tornadoes may have been occurring at similar rates historically, but weren’t reported as frequently.

To tease this out somewhat, meteorologists look to annual counts of larger, EF-3 or greater, tornadoes, whose greater scale and damage mean that they were more likely to have been reported in years that predate today’s widespread radar coverage. According to National Weather Service data from 1955 through 2020, the number of EF-3 or greater tornadoes nationally has not exhibited a discernible increasing or decreasing trend. That data and further discussion is available at tinyurl.com/mww6tn42.

A 2018 study by Vittorio A. Gensini, of Northern Illinois University, and Herald Brooks, of the National Oceanic and Atmospheric Administration’s Severe Storms Labratory, found that while there were no trends in the national frequency, there were regional differences, with a downward trend in the number of tornado occurrences over the Great Plains, the traditional “Tornado Alley” and an increase in frequency over the Midwest and Southeastern U.S.

The study did not establish, or rule out, a connection between this eastern frequency migration and climate change. Its authors noted that further research was necessary to determine whether or not the shift was due to natural or human causes, or a combination thereof. The study can be viewed at tinyurl.com/2w2rfmje.

–

Staying safe and knowing the different alerts

–

At times, there is confusion between the two different predominant alerts that the NWS issues for tornadoes. Tornado watches are listed for a broad area, such as an entire county or set of counties, several hours in advance and indicate that weather conditions favor the possibility of tornado development, but not that one is occurring at that time. Tornado watches are usually issued for a four to eight-hour period of time.

A tornado watch does not mean a tornado is occurring imminently, or that one is guaranteed to form during the watch period. When a watch is issued, there’s no need for people to take shelter right then and there. Rather, they should stay tuned, keep an eye on the skies and make sure they have a way to receive updates from the NWS.

“This is kind of like a ‘heads up,'” Hastings said. “Keep an eye out. If you’re going to be outdoors, stay tuned.”

On the flip side, a tornado warning is issued when one is imminently occurring. A tornado could either be visually observed by people in its vicinity or live video footage on the ground, or when Doppler radar — which is able to track wind direction and rotation — indicates that a tornado is likely occurring at that specific moment in time, or will begin imminently. These warnings are issued for a specific area within a county or set of counties that the tornado is impacting, typically much smaller than a watch area.

When a tornado warning is issued, people need to take shelter immediately, Hastings said. While meteorologists can determine an area where conditions will be favorable for potential tornado development, and issue tornado watches accordingly, forecasting technology that can predict exactly when and where a tornado will occur does not exist.

Therefore, tornado warnings can typically only be issued minutes or moments in advance of it impacting an area. With this in mind, people should pay close attention whenever a tornado watch is in effect for their area to give themselves as much notice as possible if a tornado warning is subsequently issued.

Hastings said that part of the problem, especially locally, is that tornadoes form and dissipate so quickly that it’s difficult to issue a warning with much advance time. She added that sometimes, the NWS will instead issue a severe thunderstorm warning for tornadic potential within that storm. This would specifically be noted in the severe thunderstorm warning’s text. While not the same as a tornado warning, people should shelter and treat severe thunderstorm warnings similarly.

“A lot of times, (tornadoes) can happen between radar scans,” she said. “It’s on the ground for just a couple of minutes, and our radar takes usually a couple of minutes to get a full scan in.”

While meteorologists still can’t pinpoint tornado paths ahead of time, advances in computer forecasting models, radars and communication technology have notably reduced the loss of life and injuries from tornadoes across the country in recent decades.

People have a variety of ways to monitor the latest weather information during a tornado watch, including online at weather.gov, over NOAA weather radio and local broadcast stations.