Contact & FAQ
For media inquiries & other information, please reach out to us via:
We are located in:
Amit Chakma Engineering Building (ACEB), Room 4470
Western Engineering, Western University
London, Ontario, Canada, N6A 3K7
Frequently Asked Questions
A tornado is a violently rotating column of air that extends through the lower part of a storm cloud to the surface (land or water). A rotating, typically funnel-shaped cloud often develops when lower pressure and temperature in the core of the tornado cause water vapour in the air to condense. This funnel cloud may extend fully or partly from the storm cloud to the surface with a tornado. Note that a waterspout is simply a tornado that occurs over water.
For NTP to verify a tornado from a photograph, video or detailed eyewitness description, it should include a well-developed funnel cloud extending fully from the storm cloud to the surface, or a funnel cloud extending partly to the surface plus an area of rotating debris near the ground / spray ring near the water surface.
Effectively, there needs to be evidence that winds associated with the vortex at least approach the lower bound of EF0 on the Enhanced Fujita Scale. It is possible for a storm to generate a vortex that is too weak at the surface to be considered a tornado. Such vortices may be accompanied by a funnel cloud aloft.
It is also possible for a short-lived, shallow vortex to develop along the leading edge of a rain-cooled storm downdraft (known as the ‘gust front’), well away from the storm updraft region. Such a vortex can generate an area of rotating debris near the ground or a spray ring over water, but is not considered a tornado. It is often referred to as a ‘gustnado’. Other vortices not considered to be tornadoes include dust devils and steam devils.
There are three main types of thunderstorms that can generate a tornado.
The first is a supercell thunderstorm, a long-lived storm with an intense, rotating updraft (known as a mesocyclone) that can often be detected by Doppler radar. The tornado develops near the right-rear flank of the storm beneath a rotating wall cloud. Because the thunderstorm's updraft can be so powerful, most significant tornadoes (EF2+) and all violent tornadoes (EF4-5) develop with supercells.
The second type is a thunderstorm complex organized roughly along a line, known as a Quasi-Linear Convective System or QLCS. Here, the storm updraft is located along the leading edge of rain-cooled storm downdrafts that often push out ahead of the storm. Tornadoes that develop along this gust front are generally weaker than those with supercells but can be strong (EF2-3).
Third, a thunderstorm can move over or develop along a local air mass boundary that has pre-existing pockets of vertically oriented rotation along it. The storm's updraft stretches the rotation into the vertical and strengthens the vortex. These tornadoes, often called 'landspout' tornadoes, tend to be weak (EF0-1) and brief, and develop as the storm is intensifying.
Sometimes the storm that produces a tornado is a hybrid that does not fit nicely into one of these categories. Note that all of these storm types can produce a weak (EF0-1) tornado, or a tornado over water (a waterspout).
Most showers and thunderstorms produce downdrafts that descend toward the surface then move outward. In certain atmospheric environments, such as those that lead to severe thunderstorms, these downdrafts may be strong enough to cause damage at the surface or interfere with the ascent/descent of aircraft, and are known as downbursts. A particularly brief, intense downburst that affects an area less than or equal to 4 km is known as a microburst.
While the rotating winds of a tornado converge at the surface then rise into the storm, often resulting in narrow paths of chaotic damage, downburst winds diverge beneath the storm and result in outward burst patterns of damage or wide areas with damage mostly from the same direction. Multiple downbursts may also combine to form squall lines, or even derechos that cause widespread wind damage along a path hundreds of kilometres long.
For NTP to verify the occurrence of a downburst, there needs to be at least EF0 damage.
NTP has a multi-step process for detecting, characterizing and documenting tornadoes.
Field teams are trained and ready to respond at a moment's notice. Tornado potential outlooks are created by NTP and sent to the teams to ensure they are prepared. Once potentially tornadic storms occur, they are tracked using available radar. NTP follows up on any reports of wind damage or visual sightings of tornadoes (usually via social media).
Ground survey teams are deployed once an area of potentially tornadic damage has been identified. The team will also typically collect drone imagery (resolution less than 2 cm) in the area of interest. While ground surveying activities are underway, other team members scour satellite imagery (resolution 3-5 m) to map out the extent of damage and determine if other unreported damage exists nearby.
In events with significant damage that cannot be fully sampled by drone, aircraft-based aerial imagery (resolution 5 cm) is collected usually some time after the event. A preliminary classification (typically tornado or downburst) and rating (using the EF-scale) is usually determined shortly after the event in coordination with Environment and Climate Change Canada.
However, NTP then undertakes a more thorough investigation incorporating all collected data to determine a final classification and rating. This typically takes from weeks to months, depending how quickly all data are received and analyzed.
Canada adopted the Enhanced Fujita scale in April of 2013 after using the original Fujita scale for many years. Like the F-scale, the EF-scale uses observed damage to rate the intensity of a tornado and estimate associated wind speeds (https://www.canada.ca/en/environment-climate-change/services/seasonal-weather-hazards/enhanced-fujita-scale-wind-damage.html).
The EF-scale however has improved relationships between observed damage and estimated wind speeds, and many damage indicators that list degrees of damage, from lightest to heaviest. The EF-scale adopted in Canada in 2013 is a slightly modified version of the original that was implemented in the United States in 2007.
The main differences are the addition or modification of a number of damage indicators (e.g., farm silos and grain bins - new, trees - modified). The minimum wind speed for EF0 was also lowered to 90 km/h in order to maintain consistency with ECCC warning criteria.
The EF-scale is used in two ways.
First, it is used to rate the intensity of damage to individual damage indicators along the damage path. This allows contouring of damage intensity, particularly useful for strong to violent tornadoes.
Second, the maximum intensity found along the entire damage path using the EF-scale is also used to characterize the strength of the tornado as a whole. EF/F0 and EF/F1 tornadoes are by far the most frequent in Canada (over 90%), though some of these might have been rated higher if more damage indicators had been present. This is particularly true for some parts of the Prairies.
The percentage of Canadian tornadoes that have been rated EF/F4 or higher is less than 0.5%. In fact, there has only been one recorded EF/F5 tornado in Canada, the Elie, MB tornado of 2007. Some NTP scientists are involved in the creation of a new and better version of the EF-scale that will eventually become a standard under the American Society of Civil Engineers for the first time.
High-quality tornado data are important in a number of areas:
(1) Verification – the occurrence of a tornado is used for ECCC alert verification, and allows forecasters to see how well their warnings perform and how they may be improved.
(2) Climatology and risk analysis – many organizations depend on knowledge of tornado risk, including municipalities, provinces, emergency managers, companies with electrical transmission lines, nuclear plants, insurance and reinsurance companies, etc.
(3) Science process studies – scientists that study how tornadoes form and what clues to formation are apparent in radar, satellite and lightning data need to know where and when tornadoes occurred and how strong they were.
(4) Climate change studies – to have an observations-based understanding of trends in tornado occurrence due to climate change, a high-quality, long-term tornado data set is required.
(5) Raising awareness – many Canadians may not be aware that they live in tornado-prone regions, so better tornado data helps to raise the level of awareness and increase knowledge about what actions to take in the event of a tornado.
Tornadoes have been recorded in every province and territory in Canada. However, tornadoes occur most frequently in two areas - from southern Alberta across southern Saskatchewan and southern Manitoba to northwestern Ontario, and from southern Ontario across southern Quebec to New Brunswick. These areas are extensions of tornado-active areas in the United States, though separated by an area of low frequency caused by the stabilizing influence of the relatively cool Great Lakes.
While the area or areas experiencing the most tornadoes can change from year to year, on average it is extreme southern Saskatchewan, extreme southern Manitoba and southwestern Ontario that record the most tornadoes. These are areas that have also experienced some of the most intense tornadoes and largest tornado outbreaks as well.
Our Open Data site is constantly updating its tornado data.
Check out the Environment and Climate Change Canada 30 year map
Check out the Environment and Climate Change Canada Fact Sheet
Across Canada, the month with the most tornado activity is July (about a third of all tornadoes). Slightly more than half of all Canadian tornadoes occur between 4:00 pm and 7:59 pm (data drawn from analysis of the ECCC 30-year database).
Considering all tornadoes, the average path length is 10.6 km. That increases to 23.9 km for strong to violent tornadoes (EF/F2 to EF/F5). The average path direction is 260⁰ (from the west). If you took the paths of all tornadoes that occurred in Canada from 1980 to 2009 and added them up, the total path length would be 3,058.2 km (roughly the driving distance from Ottawa, ON to Saskatoon, SK!) (all data drawn from analysis of the ECCC 30-year database).
Considering all tornadoes, the average path width is 249 m. That increases to 391 m for strong to violent tornadoes (EF/F2 to EF/F5) (data drawn from analysis of the ECCC 30-year database).
The following are the seven worst single tornadoes in Canada in terms of the number of fatalities and injuries, and often associated damage costs as well (values unadjusted):
• Regina, SK F4 tornado, 30 Jun 1912 – 30 fatalities, hundreds injured, $4M in damages
• Edmonton, AB F4 tornado, 31 Jul 1987 – 27 fatalities, over 300 injured, over $250M in damages
• Windsor, ON F4 tornado, 17 Jun 1946 – 17 fatalities, over 100 injured, $1.5M in damages
• Pine Lake, AB F3 tornado, 14 Jul 2000 – 12 fatalities, at least 130 injured, $12M in damages
• Valleyfield, QC (not rated) 16 Aug 1888 – 9 fatalities, 16 injured, nearly $3M in damages
• Windsor, ON F1 tornado, 3 Apr 1974 – 9 fatalities, 30 injured, curling arena damaged
• Barrie, ON F4 tornado, 31 May 1985 – 8 fatalities, 155 injured, $150M in damages
There have been a number of impressive tornado outbreaks as well, with the eight largest listed below:
• Southern ON, 20 Aug 2009 – 19 tornadoes up to F2
• Southern ON, 2 Aug 2006 – 17 tornadoes up to F2
• Southern ON, 31 May 1985 – 14 tornadoes up to F4
• Southern QC, 5 Sep 2018 – 14 tornadoes up to EF2
• Southern SK / AB, 29 Jun 1984 – 13 tornadoes up to F2
• Southern AB, 31 Jul 1987 – 13 tornadoes up to F4
• Southern SK, 30 Jun 1912 – 12 tornadoes up to F4
• Southern AB, 8 Jul 1927 – 12 tornadoes up to F3
The ECCC/MSC Library at the Downsview facility in Toronto holds all of the country’s physical tornado archives including reports, event summaries, maps, newspaper clippings, photos, slides, video and film. This material is in the process of being digitized and some will soon be available online via www.uwo.ca/ntp.
The NTP website is also a source for recent tornado data and includes interactive mapping, access to damage photos, drone video, and aerial imagery.
On the Government of Canada Open Data portal, tornado data compiled for the period 1980-2009 can be found in multiple formats: https://open.canada.ca/data/en/dataset/f314a39f-009d-430b-97b9-d6e0cae22340
There have been a number of publications that have been written following significant tornado events. Some are out of print and hard to find:
• Not Like Any Other Sunday by Elizabeth Bundy-Cooper and Cathy Cove (focusing on the 21 Aug 2011 F3 Goderich, ON tornado)
• Under the Whirlwind by Arjen Verkaik and Jerrine Verkaik (a general book on tornadoes with focus on the F3 tornadoes of 20 Apr 1996 in southern Ontario)
• Black Friday by the Edmonton Sun (focusing on the F4 tornado of 31 Jul 1987 that hit Edmonton, AB)
• Tornado Town by Laura Lennox and Brock McKinney (focusing on the F4 tornado of 31 May 1985 that hit Grand Valley, ON)
• Ontario Tornado by the Barrie Banner (focusing on the tornado outbreak of 31 May 1985 that produced the F4 Grand Valley, ON tornado and the F4 Barrie, ON tornado)
• Tornado by John Toll (focusing on the F4 Woodstock, ON tornado of 7 Aug 1979)
• The Tornadoes of Western Canada by A. B. Lowe and G. A. McKay (focused on tornado events in western Canada up to 1960)
They can actually be either depending on their intensity. ECCC defines severe weather as hail having diameter 2 cm or greater, 50 mm of rain in an hour or less, wind gusts reaching 90 km/h or greater, or tornadoes. These phenomena occur many times per year across Canada and are considered a regular part of our warm season climate.
However, climatologically extreme weather can also occur; that is, severe weather phenomena having a size or intensity that has been very rarely if ever observed. Examples are violent tornadoes (EF/F4 to EF/F5) and softball-sized (or larger) hail that can only be generated by extremely intense supercell thunderstorms.
Despite the awesome power of tornadoes, they are excruciatingly small in scale when compared to other phenomena that meteorologists must forecast. They are also difficult to detect.
Doppler weather radars are not typically able to resolve a tornado, and tornadoes cannot be seen using satellite or lightning data. This is why visual reports of tornadoes and their precursors from the public have historically been so critical, with volunteer networks in place in Canada since the 1980s (an early example of crowdsourcing).
Increasingly, forecasters have been able to predict the larger-scale conditions conducive to the development of tornadic storms with the help of ever-improving computer models, sometimes days in advance. While this will not allow a prediction of exactly where a tornado might form, it does allow for some lead time and preparedness, with the issuing of watches as necessary.
Many scientists continue to explore the question of what exactly makes a tornado form, and how a tornado might be detected better in real time. Tornado prediction is likely to improve due to this ongoing work.
Our warming climate will most likely have an impact on tornado activity in Canada and other parts of the world. What exactly that impact might be and how quickly such changes might be observed is currently a topic of research.
The tornado climatology may be modified in time, space and/or intensity and the direction of such changes may be regional in nature. It is important to continue to build a high-quality national tornado data set in order to be able to detect such shifts.