This page has been archived on the Web

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.

Hurricane Hazel - Storm Information

Links To Storm Information:

  1. Storm Information
  2. Extratropical Transition
  3. Rainfall and Flooding
  4. Weather Bulletins

Storm Information

Between October 5 and October 16, 1954, Hurricane Hazel formed in the tropics, tracked through the interior of the United States and across the Great Lakes where it merged with a cold front and dropped record rainfall amounts in the Greater Toronto Area. The Toronto area had experienced above average rainfall that autumn, especially in the two weeks prior to Hazel’s arrival, saturating the ground and preventing the rainfall from infiltrating the soil. Instead, the water ran off the surface into local rivers, which rose very rapidly, wiping out communities built within their floodplains. The Weather Office warned of Hazel’s arrival throughout the day on Friday October 15, 1954, stressing the impending heavy rainfall. Their forecasts were quite accurate, yet, many people did not feel they were adequately warned of the risk posed by the rivers.

Hurricane Hazel did not strike Ontario... another storm did! In his 1955 paper in the Bulletin of the American Meteorological Society, Canadian Meteorological Service senior meteorologist, John Knox, described how the extratropical transition of Hazel resulted in southern Ontario being hit by a dangerous storm that was made more powerful by the hurricane... but it was a different storm. In describing the evolution of this new storm, Knox labelled Hazel the hurricane, as “Hazel I”, while the new storm centre which formed south of the Great Lakes was labelled, “Hazel II”. The development of “Hazel II” resulted in an apparent northwestward-jump of the storm centre and many storm track climatologies mistakenly show Hazel’s motion as continuous just south of Lake Ontario. The map below shows the tracks of Hazel I & II.

Tracks of Hazel I & II
Tracks of Hazel I & II

Here is an account of Hazel I & II

  • October 5 - A tropical disturbance near Grenada, at the southern end of the chain of islands in the Caribbean, was classified as a hurricane: “Hazel”.
  • October 6-9 - Hazel moves slowly westward in the Caribbean ocean: no land areas affected.
  • October 10 - Hazel turns abruptly northward.
  • October 11 - Hazel tracks northward toward Haiti.
  • October 12 - Hazel crosses the western tip of Haiti, killing more than 400 people.
  • October 13 - Hazel exits northward through channel between Haiti and Cuba then turns northwest toward the U.S. east coast.
  • October 14 - Hazel increases forward speed toward U.S. east coast.
  • October 15 - Hazel makes landfall in the Carolinas around noon as a marginal Category 4 hurricane. Hazel turns more northward and accelerates, entering Pennsylvania in the early evening, dissipating rapidly. Simultaneously, a rapid and continuous transformation of energy resulted in the formation of Hazel II, the extratropical cyclone, over New York State. By late evening, a vastly different storm existed... approaching Lake Ontario. Before leaving the United States, Hazel claimed 95 American lives. The storm centre moved directly over the city of Toronto near midnight.
  • October 16 - By morning, Hazel II was in Northern Ontario. Between the evening before and the early morning hours of October 16, 81 people in southern Ontario died from the flooding rains.

Nowhere is the saying, “timing is everything,” truer, than in meteorology. Three events took place with almost sinister timing to cause catastrophic flooding in Toronto: pre-Hazel rainfall which saturated the ground; a deep and intensifying atmospheric trough approaching from the west; Hurricane Hazel approaching from the south.

The following five figures show the evolution of the atmosphere over North America from October 12 to October 16, 1954. These images are snapshots in time (11 p.m. EDT each day) at the 700-millibar pressure level (approximately 10,000 feet). The contour lines on these maps can be interpreted as a flow of air at that level, where lines that are closer together represent stronger winds. On October 12 and 13, Hazel is well-removed to the southeast while a strong stream of air flows across much of Canada and the northern United States. On October 13, a weak trough develops near the west coast (red dashed line). This trough migrates eastward and becomes more pronounced over the next couple of days (October 14 and 15), eventually extending from the Great Lakes to Texas. Meanwhile, Hazel is still approaching the United States east coast. Between October 14 and October 15, a pronounced ridge develops over Atlantic Canada (blue dashed line). With Hazel approaching the US coast between the trough and the ridge, the storm is forced to accelerate northward.

Hazel at 700 millibar On October 12
Hazel at 700 millibar On October 12

Hazel at 700 millibar On October 13
Hazel at 700 millibar On October 13

Hazel at 700 millibar On October 14
Hazel at 700 millibar On October 14

Hazel at 700 millibar On October 15
Hazel at 700 millibar On October 15

Hazel at 700 millibar On October 16
Hazel at 700 millibar On October 16

A lot happens in the final 24 hours between the last two maps. The trough continues “digging” southward while the ridge over Atlantic Canada amplifies greatly. The development strengthens the northward flowing airstream over the eastern continent while also setting up the potential for a strong storm system to form somewhere near or south of the Great Lakes. Hazel moves inland and begins dying, as all tropical cyclones do when they move over land and are cut-off from their oceanic energy source. However, the dying process isn’t immediate and can take more than a day for a powerful hurricane like Hazel. The decaying remnants of Hazel move rapidly northward in the strengthening flow between the trough and ridge, arriving south of the Great Lakes about the same time that an extratropical cyclone was in the process of forming. The result was not unlike throwing kerosene on a newly-formed fire: an explosive atmospheric reaction.

The following three weather maps show the rapidity with which Hazel reacted with the trough on October 15.

This map (from NOAA) shows the surface pressure lines and weather features over North America at 11 a.m. EDT. Hazel is about to land in the Carolinas while a cold front extends from Ontario (just east of Toronto) down to the Gulf of Mexico. The shaded area depicts rainfall, which can be seen to connect from Hazel all the way to an area of low pressure north of Lake Superior. (October 15, 1954)
This map (from NOAA) shows the surface pressure lines and weather features over North America at 11 a.m. EDT. Hazel is about to land in the Carolinas while a cold front extends from Ontario (just east of Toronto) down to the Gulf of Mexico. The shaded area depicts rainfall, which can be seen to connect from Hazel all the way to an area of low pressure north of Lake Superior. (October 15, 1954)

Map (from Knox) shows a zoom into the area near the Great Lakes, from 6:30 p.m. EDT shows the position of the front (still near Toronto) and the elongated area of low pressure... with the dying centre of Hazel I still evident. (October 15, 1954)
Map (from Knox) shows a zoom into the area near the Great Lakes, from 6:30 p.m. EDT shows the position of the front (still near Toronto) and the elongated area of low pressure... with the dying centre of Hazel I still evident. When the cold front entered western Ontario near Windsor, it was advancing eastward at about 40 kilometres an hour. Thunderstorms along the front pushed rain through southwestern Ontario, saturating the ground. The evolution of the atmosphere caused a rapid deceleration of the front such that by the time it reached Toronto it came to a complete standstill. Toronto was getting drenched, and Hazel had yet to arrive. (October 15, 1954)

map from 10:30 p.m. EDT shows a more consolidated storm centre in the pressure pattern over the western end of Lake Ontario. This is Hazel II. (October 15, 1954)
Map from 10:30 p.m. EDT shows a more consolidated storm centre in the pressure pattern over the western end of Lake Ontario. This is Hazel II. (October 15, 1954)

Hazel II passed directly over Toronto near midnight, October 15 and 16, as evidenced by the contrasting weather between Malton Airport (now, Pearson International Airport) and Toronto City Centre. At midnight, Malton reported west-southwest winds while the Toronto Island reported east winds. Similarly at midnight, while Malton was reporting 8°C the Bloor Street weather office was reporting 16°C. This remarkable contrast pinpoints the storm centre over Toronto exactly at midnight. The airport remained in the cold air west of the centre with the temperature rising to no higher than 11°C while the wind backed in direction from northwest to west-southwest. On the other hand, the storm’s warm sector moved over the city with the wind veering from north through east while the temperature climbed to 17°C. Knox pointed out that the structure and behaviour of the storm that hit Toronto was much different than the one that existed earlier that day.

Extratropical Transition

An explanation of extratropical transition is in order; it is the process by which tropical cyclones transform into extratropical cyclones. Extratropical cyclones are the common low pressure centres which form along fronts in mid latitudes. They draw their energy from the imbalance that takes place when cold and warm air move into close proximity: greater temperature differences over shorter distances result in more intense frontal storms, or extratropical cyclones. Tropical cyclones, such as hurricanes, on the other hand, are warm air storms which draw their energy from the release of latent heat... the process when evaporated ocean water rises in the atmosphere as water vapour and subsequently condenses back into rain. This process releases heat into the centre of an already warm storm, making it warmer, and hence, more intense.

What happens when a tropical cyclone moves into the cooler and windier environment of the mid latitudes is a great forecast challenge and has been a focus of considerable scientific investigation since the mid 1990s. Tropical cyclones undergoing extratropical transition change structure, size, and intensity. The resulting cyclone--known now to meteorologists as ET storms--can be weaker or more intense than the incipient tropical cyclone, depending on the timing or phasing of a number of atmospheric components.

ETs are an ongoing concern for Canada in that they are amongst the most difficult meteorological forecast challenges and they can bring a surprising severity of weather. Typically, ETs are cyclones which would have likely formed anyway, but the injection of strong tropical ingredients, such as the strong low-level rotation of the atmosphere and the high moisture content and rainfall, make them more potent than they would have otherwise been. Hazel was a dramatic example of a powerful ET.

To give an appreciation of the forecast challenges with ETs, consider Hurricane Ivan from September 2004. The rainfall and upper atmospheric moisture from a dying Ivan (inland over Tennessee) fed northward into a developing storm over eastern Canada. The result was a much more intense storm system over Atlantic Canada than would have otherwise developed, with Newfoundland waters experiencing hurricane force winds and fatalities. Meanwhile, the lower atmospheric rotation of Ivan drifted back southward, through Florida, into the Gulf of Mexico, where it regenerated back into Tropical Storm Ivan, creating dangerous conditions all over again along some United States Gulf coasts.

Rainfall and Flooding

Rainfall over Southern Ontario associated with Hurricane Hazel was very intense, yet the cause of the severe flood was not solely Hazel. Residents recalled that the two weeks leading up to Hazel were the wettest they could remember with more than a month and a half’s worth of rain in the 16 days before Hazel. Drainage basins were saturated; therefore, when Hazel dropped the record amount of precipitation on the region, water ran off the surface directly into streams and rivers. Toronto is dissected by rivers and streams and in 1954 many people lived on the city’s floodplains, placing them directly in the path of the Hazel flood. Flooding occurred from Highland Creek in the east to the Credit River in the west. The Humber River, Highland Creek, Don River, Etobicoke Creek, Credit River, and 16 Mile Creek all overflowed their banks.

Reported rainfall amounts were:

Malton airport, 107 mm in 12 hours, end 137 mm in the 24 hours up to midnight. (Mason, 1955)

Toronto, 183 mm in 24 hours (Kennedy)

  • 76 mm east, south of the Danforth
  • 90 mm east of University
  • 127 mm in Weston and Islington (TS, Oct 23, 1954)

The highest reported 48-hour rainfall: 214 mm in Snelgrove. Brampton and Snelgrove both received an estimated about 90 mm in the 3 hours between 9 p.m. and midnight. (Mason, 1955)

The two maps here show Hazel’s rainfall over Ontario (from Mason).

Hazel’s rainfall over Ontario
Hazel’s rainfall over Ontario

Hazel’s rainfall over Ontario
Hazel’s rainfall over Ontario

Brampton received the greatest concentration of precipitation as this was where the cold front and Hurricane Hazel began to merge. Brampton received 210 mm in 48 hours, falling on land in the upper west end of the Humber watershed and the upper reaches of the river system itself, flowing 153 m down into the Humber River Valley. An escarpment approximately the same elevation runs north from Brampton parallel to the Humber River and also received heavy rains, which were funnelled into the river. The east side of the river from the Barrie highway to Woodbridge has a similar slope with a drop of 76.2 m.

The Humber River experiences a drop of 366 m between the farthest headwaters and the shores of Lake Ontario; so as water rushed overland in the Brampton area into the river, the river level rose 6 to 9 m in some areas and raced down the river over the course of the next several hours at speeds up to 1416 m3/s.

The flood hit Woodbridge first, but was waylaid by a bridge, which acted as a temporary dam. When the bridge failed to hold back the floodwaters, flooding cascaded down the river valley occasionally slowing at other bottlenecks, but hitting communities with tremendous force. There were only four dams on the Humber prior to Hazel and none were built for flood control purposes. Consequently, they may have held back a negligible amount of water but failed to have much effect. Two of the dams, Bolton and Woodbridge, were opened entirely and two, Pine Grove and Palgrave, were partly washed away.

The reason for the severity of the Humber flood was explained by Robert Campbell in the Toronto Star on October 16, 1954. “The Humber River rises in the Peel Plains about 113 kilometres north of Lake Ontario. The plains have a hard clay base and any heavy rainfall rushes off almost immediately. The Humber River takes it all.” The Humber River watershed is approximately 777 kilometres squared in area and received 229 mm of rainfall. One inch of rain on one square mile of land equals 14.5 million gallons of water; for the entire watershed, approximately 151 billion litres of water fell during Hazel or 200 million tons (Kennedy). “It was like dumping a lake the size of Lake Simcoe on the Humber River drainage area and having it all trying to get out by way of the river at once,” Turnbull told the Toronto Star (October 23, 1954). The Toronto Star reported that 246 billion litres of rain fell on the combined Don and Humber watersheds.

Weather Bulletins

Was Toronto properly warned about Hurricane Hazel? Special Weather Bulletins were issued by the Dominion Weather Office in Malton. However, many of the city’s residents did not take protective measures as is evidenced by the loss of life and widespread damage. For a complete review of the Weather Office’s management of Hazel, see the Evaluation of Hurricane Hazel.

Norman Grundy, a meteorologist at the Dominion Weather Office, developed a prognostic chart Friday morning that showed Hurricane Hazel over eastern Lake Ontario by 2300, on October 15, 1954, contrary to American meteorologists’ predictions that the storm would dissipate over the Allegheny Mountains. A forecast was issued at 9:30 a.m., calling for continuous rain throughout Friday and into the night, but Hazel was expected to lose strength as it crossed over the interior of the United States. Forecasters were also watching a cold front located west of Hazel’s projected track, recognizing that if the two systems combined, the rainfall would be more severe. Fred Turnbull, head of the weather office at Malton, issued a statement to the Telegram newspaper in which he stated that the rain they were expecting “could be the heaviest ever recorded in the city’s history,” exceeding the rainfall in 1887 that dropped close to four inches [101.6 mm] of rain. (Kennedy, 1979)

Turnbull’s post-Hazel report, summarizing the weather office’s handling of the storm stated, “ The issuing of a weather bulletin, particularly advising of the approach of a hurricane, is a serious step and not to be taken lightly. Since storms of this nature seldom follow our predicted course, and rarely travel 1100 kilometres [700 miles] overland retaining sufficient energy to be alarming, we decided to consult our Central Analysis Office in Montreal before issuing a special bulletin. In a telephone call to Montreal, Mr. Leaver of the Central Analysis Office agreed to have his staff give special consideration to Hurricane Hazel, and call us immediately after completion of their prognostic chart.” (Kennedy, 1979; p. 37)

After consulting with the Central Analysis Office, a Special Weather Bulletin was issued over a private teletype network, by telephone to local radio stations and to commercial interests that had identified themselves on a severe weather call list. The information was also sent to Marconi stations so that vessels on the Great Lakes could be warned of the approaching storm.

Special Weather Bulletin

“Hurricane Hazel moved inland near Myrtle Beach, South Carolina, this morning. Highest winds are estimated at 161 kilometres per hour [100 miles per hour] over a small area, with winds of 64 to 97 kilometres per hour [40 to 60 miles per hour] extending 129 kilometres [80 miles] to the North and East and 64 kilometres [40 miles] to the South and West of the centre. Hurricane Hazel is expected to continue at 35 to 40 miles [56 – 64 km/h] per hour [56 – 64 km/h] towards the Northwest for the next few hours, then to follow a more northerly course.

“Shipping interests on Lake Erie and Lake Ontario should take precautions against the possible occurrence of winds of 64 to 117 kilometres per hour [40 to 70 miles per hour].” (Kennedy, 1979; p. 38)

A second bulletin was issued around noon on Friday that read, “Hurricane Hazel moved inland just east of Myrtle Beach, South Carolina, at 9:15 Eastern. At 11:30 it was centred 48 kilometres [30 miles] west of Wilmington, North Carolina, and moving Northward with a speed of 48 kilometres per hour [30 miles per hour].

“The hurricane is expected to continue on its northerly course, reaching the eastern end of Lake Ontario about midnight tonight. In crossing the Allegheny Mountains the hurricane will decrease markedly in intensity with winds not expected to exceed 80 kilometres per hour [50 miles per hour] on the open water of Lake Ontario.” (Kennedy, 1979; p. 39)

Turnbull spent much of the afternoon and evening fielding calls for interviews from local media regarding the approaching storm and conducted four radio broadcasts to advise the public that Hazel would reach Toronto Friday night. The weather office’s switchboard registered on average one call every 20 seconds for a total of 2,300 calls.

At 9:30 p.m. the final official forecast on the storm was issued by the weather office and read as follows:


Hurricane Hazel, which moved in on the North Carolina coast Friday morning, continued to move northward and to accelerate during the day and by 9:00 p.m. it was centred between Buffalo and Rochester.

The intensity of this storm has decreased to the point where it should no longer be classified as a hurricane. This weakening storm will continue northward, passing east of Toronto before midnight. The main rainfall associated with it should end shortly thereafter, with occasional light rain occurring throughout the night. Winds will increase slightly to 45 to 50 MPH [72 – 80 km/h] until midnight, then slowly decrease throughout the remainder of the night.

For the sectors of Lake Ontario and Niagara Regions and Toronto and Hamilton Cities:

Rain tonight. Cloudy with occasional showers Saturday. Little change in temperature. Winds north 40 to 50 miles per hour [64 – 80 km/h] this evening decreasing overnight to northwest 30 Saturday. Low tonight and high Saturday at Toronto, St. Catherines, and Hamilton, 45 and 55 [7 and 13°C]. Outlook for Sunday cloudy and cool.

Date modified: