Archived Content

Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats on the Contact Us page.

Data Source and References

Frequency Tables and Maps of Tracks and Intensities

The National Hurricane Center (NHC - Miami Florida) has the responsibility for maintaining a database of historical Atlantic tropical cyclones. This “Best Track” database or HURDAT is an archive dating back to 1851. The database is maintained in a highly formatted text file and includes six-hour track positions, maximum winds, central pressures, etc.  The database is updated yearly, at the end of the hurricane season. A hurricane forecasting specialist is assigned a particular storm and reviews all observational data with respect to that storm.  This late analysis utilizes all data that was available at the time of the storm and also includes any information that became available later. Thus, the final product represents the Best Track that can be produced for each storm and becomes part of the official record.

Wind and Rainfall Data Sources and Methodologies

A 29-page detailed report (independent of more than 250 analyses of wind or rainfall) outlines the data sources and analysis methodologies for, and results of the tropical cyclone wind and rainfall climatology. The results of that study are scattered about this website under targeted headings for convenience and ease of access. For a more complete background on the methodologies used as well as the full results, a copy of the original report is available upon request.

Data sources that helped to determine precipitation occurrence, amounts, and extent to complete the rainfall analysis portion of this study were acquired through a several sources including:

  • Environment Canada, National Climate Data and Information Archive (NCDIA), Canadian Daily Climate Data (CDCD) total daily rainfall dataset for specific locations
  • Daily 10 kilometre Raster-Gridded Climate Dataset for Canada 1961-2003 from the National Land and Water Information Service (NLWIS). This dataset developed in coordination with Environment Canada (EC) in 2008
  • NOAA/NWS Historical Weather Viewer
  • HURSAT Microwave Retrievals 1993-2006
  • NOAA/NEDSDIS satellite imagery
  • NCEP reanalysis model data
  • Archived Canadian North American Surface Analyses
  • NOAA Daily weather maps
  • NHC best track data

A total of 125 storms were used for the wind analysis for 1979-2008, with an additional 17 storms for the period 1953-1978. This is two more cases than used for rainfall due to two events that produced gale force winds within the forecast area without any rainfall over lands areas. Wind data was more of a challenge to obtain and sort out than the rainfall data. The factors that contributed to this challenge include:

  • Reporting of 1-minute vs. 10-minute wind speed averages
  • Inconsistent coverage and sparse data especially over marine areas
  • Inconsistencies in reporting techniques and requirements among the various sensor platforms such as ships, buoys, and land-based
  • Changing and developing wind asymmetries associated with transitioning Tropical Cyclones.

Wind averaging reporting is a significant issue. Forecasters and other users of wind data should be aware that the maximum wind speeds in Tropical Cyclones, as well as the radial decrease of wind from the maximum wind zone, are seldom measured with a degree of precision implied by standard wind definitions. Tropical cyclones rarely pass directly over measurement devices and when they do, the devices are often incapacitated by the strong wind speed. Thus, there is heavy reliance on indirect estimates of surface Tropical Cyclone wind speeds and directions such as wind data provided by aircraft, satellite cloud imagery, radar, etc. These platforms also introduce another level of error into the measurements.

Since surface winds and gusts can change dramatically over short time intervals, it is necessary to define the length of time over which the winds are to be measured. For a cyclone of some given intensity, longer wind averaging times will yield lower maximum winds. Unfortunately, different meteorological services use different averaging times. Following World Meteorological Organization (WMO) guidelines, most regions use a 10-minute average. However, WMO Regional Association IV (United States and Caribbean area) and therefore the NHC, use a 1-minute standard average. A Tropical Cyclone defined as a hurricane using a 1-minute standard may not be defined as a hurricane using a 10-minute standard. Winds averaged over periods of at least 1 minute are referred to as sustained winds.

By examining a large number of recorded wind speeds vs. time traces and damage reports, conversion factors for going from one averaging time to another have been derived by (Fujita, 1971; Simiuand and Scanlon, 1978; Krayer and Marshall, 1982). Depending on the methodology, there are some small differences in recommended conversion factors. For marine interests, the factor 0.88 is commonly used in going from a 1-minute system to a 10-minute system such that TEN-MINUTE MEAN = 0.88 * ONE-MINUTE MEAN or ONE-MINUTE MEAN = 1.14 * TEN-MINUTE MEAN. There are many variations depending mainly on the frictional characteristics of the surface area and the atmospheric stability.

For the purposes of this study, the conversions for the various platforms were not taken into account and a 10-min average was assumed. This is mostly because of the number of different platforms used, the large number of storms analysed, and the emphasis on the winds over the marine areas where the 10-min averaging is typical. Therefore, the user of these wind data and the charts, graphics, and statistics generated from them should keep in mind that some cases may indicate lower wind speeds and threshold winds of lesser extent than found in the NHC best track data which uses a 1-min average.

Storm Surge Data Sources and Methodologies

A 172-page detailed report  outlines the data sources and analysis methodologies for, and results of the tropical cyclone storm surge climatology. The results of that study are scattered about this website under targeted headings for convenience and ease of access. For a more complete background on the methodologies used as well as the full results, a copy of the original report is available upon request.

Once specific tropical cyclones of interest were identified, hourly water-level data were obtained from the Marine Environmental Data Service (MEDS). Water levels are given with respect to local Chart Datum (CD), which is defined by the Canadian Hydrographic Service (CHS) in the Canadian Tide and Current Tables as the plane of lowest normal tides. CHS also has the responsibility of generating Canada’s official tidal predictions and these were obtained from CHS for period of record of a given tide gauge to 2008. A full discussion of data quality control, error identification and caveats regarding trend-analysis is discussed in the full report.

Predicted astronomical tides were subtracted from the adjusted observed water-level data to obtain maximum surge values for each storm. For surge peaks above 40 cm, the maximum observed water level during the peak was also recorded, (which tended to be different from the water level at the time of high tide), as was the duration of the surge above 40 cm, in hours. Sometimes the time series of the residuals will rise and fall through the 40 cm threshold, therefore to qualify as separate events, it was decided that peaks needed to be separated by at least 12 hours and the trough needed to pass through a value less than 40 cm. Storms which had peaks less than 12 hours each had their peak durations recorded. For storms that appeared to have no evident peak and were indistinguishable from the general variability of the residual, the surge was considered to be zero.

A storm surge was considered to be very significant if it reached 60 cm or more and more statistics were performed on those events. Surge profiles were plotted for all storms of 60 cm or more with hourly observations when available (after 1988).

Several sources were reviewed to find records of flooding caused by storm surge:

  1. Part 2: Storm Impacts from “A Climatology of Hurricanes for Canada Improving Our Awareness of the Threat.”  Most of the data on flooding data came from this source (which is itself a literature review) as it lists impacts known to have affected areas in Atlantic during 1901-2000 from all tropical cyclones.
  2. “Historic Seismicity and record of severe storms with coastal flooding for Western Newfoundland,” was used for events affecting St. John’s, Argentia and Port aux Basques.
  3. “Flooding Events in Nova Scotia a Historical Perspective,” was used for events affecting Halifax, Yarmouth and North Syndey.
  4. “Severe Storms of Canada’s East Coast: A Catalogue Summary for the Period 1957-1983,” was used for events affecting the Maritime coast.
  5. Gulf of St. Lawrence flooding found in “Impacts of Sea-Level Rise and Climate Change on the Costal Zone of Southwestern New Brunswick.”

Storm Impacts Summaries

(ET)Evening Telegram, St. John’s, NL.
(G)The Guardian, Charlottetown, P.E.I.
(HH)Chronicle Herald, Halifax, N.S.
(TJ)Telegraph Journal, Saint John, N.B.
(MG)The Gazette, Montreal, QC.
(TS)Toronto Star, Toronto, Ontario
(GM)Globe and Mail, Toronto, Ontario
(DFP)Dartmouth Free Press, Dartmouth, N.S.
(KWS)Kingston Whig Standard, Kingston, Ontario
(BB)Bridgewater Bulletin, Bridgewater, N.S.
(LPE)Lunenburg Progress Enterprise, Lunenburg, N.S.
(OC)The Ottawa Citizen, Ottawa, Ontario
(TR)The Register, Berwick, N.S.

(SNL)Shipwrecks of Newfoundland and Labrador
Frank Galgay and Michael McCarthy,
Creative Book Publishing, 1987(volume 1), 1990(volume 2), 1995 (volume 3), 1997 (volume 4)
(SNS)Shipwrecks of Nova Scotia
Jack Zinck,
Lancelot Press Limited, 1975(volume 1) & 1977(volume 2)
(IPS)In Pearl On The Sea: Shipwrecks of Nova Scotia
Robert C. Parsons,
Pottersfield Press, 2000
(ADG)Atlantic Diver Guides
David N. Barron,
Atlantic Diver, 1988(volume 1-4)
(TOS)Toll of the Sea: Stories from the Forgotten Coast
Robert C. Parsons,
Creative Publishers, 1995
(WOS)The Wake of the Schooners
Robert C. Parsons,
Creative Publishers, 1993
(WW)Wind and Wave: Sea Tales from Around Our Coast
Robert C. Parsons
Creative Publishers, 2003

(GGHC)The Great 1900 Galveston Hurricane in Canada
(LAS)Lost at Sea
(NLR)Newfoundland Roots

(CDD)Canadian Disaster Database, Ottawa, Ontario
(CHC)Canadian Hurricane Centre, Dartmouth, N.S.
(NSDNR)Nova Scotia Department of Natural Resources, N.S.
(USCG)United States Coast Guard, Washington D.C., U.S.
(CSC)Community Services Council, NL.
(NSD)Northern Shipwreck Database
(MTB)The October 15-16, 1954 Storm, “Hurricane Hazel” in Ontario
Meteorological Service of Canada Archives, CIR-2606, TEC-210, Jan 21 1955
A. H. Mason, M. K. Thomas, and D. W. Boyd
(D)A Study of Blowdown in Nova Scotia, October 1958
G. D. Dwyer

A list of additional sources was compiled for Hurricane Juan
(CBC)Canadian Broadcasting Corporation
(HDN)Halifax Daily News, Halifax, N.S.
(TDN)Truro Daily News, Truro, N.S.
(CCG)Canadian Coast Guard
(NR)Nature’s Resources (N.S. government magazine)
(HOC)Halifax Outdoor Club, Halifax, N.S.
(DL)Dartmouth Laker, Dartmouth, N.S.
(HRM)Halifax Regional Municipality, Halifax, N.S.
(TR)The Rucksack, Halifax, N.S.
(CNR)Canadian National Railway
(NSCC)Nova Scotia Community College, Halifax, N.S.
(MSVU)Mount Saint Vincent University, Halifax, N.S.
(DAL)Dalhousie University, Halifax, N.S.
(SMUT)Saint Mary’s University Times, Halifax, N.S.
(IBC)Insurance Bureau of Canada
(NSPI)Nova Scotia Power Incorporated, Halifax, N.S.
(NSEMO)Nova Scotia Emergency Measures Organization, Halifax, N.S.
(LHD)Lighthouse Digest
(TEN)The Evening News, New Glasgow, N.S.
(CYCN)Charlottetown Yacht Club Newsletter, Charlottetown, P.E.I.
(TBE)The Bulletin Enterprise
(GPEI)Government of Prince Edward Island
(ADNSPEI)Anglican Diocese of Nova Scotia and Prince Edward Island, Halifax, N.S.
(CFIA)Canadian Food Inspection Agency
(DFNS)Dairy Farmers of Nova Scotia
(COAST)Coast Magazine, Halifax, N.S.
(TB)Truro Bureau, Truro, N.S.

Additional sources for 2004 storms
(TWS)The Windsor Star, Windsor, Ontario
(SS)Sudbury Star, Sudbury, Ontario
(NP)National Post, Don Mills, Ontario
(SP)Star-Phoenix, Saskatoon, SK
(TC)Times-Colonist, Victoria, B.C.
(CTV)CTV News Online
(SLRB)St. Lawrence River Board of Control, Ottawa, Ontario
(TTWS04)Top Ten Weather Stories of 2004, MSC

Information that is categorized by date is the summary information gleaned from newspapers published on that date.


One objective of the Canadian Hurricane Centre Climatology Project was to compile Canadian hurricane related damage and mortality statistics. Four broad sources were consulted, one more extensively than others. Sources include newspapers, the internet, and other organizations.


Newspapers provide very detailed real time accounts of the impact of a storm; therefore they are the most useful source. Monetary damage estimates cited in newspaper articles immediately following a storm may be misleading due to incomplete damage reports, failure to include an entire region, or failure to specify what statistics are included. Revisions of these storm costs may be neglected in future editions. Often a newspaper will fail to update stories reported in previous editions overlooking revisions on a storm’s impacts. One example witnessed were reports that a person had gone missing in a storm, but with no further information. In these situations missing persons cannot be added to a death toll for a storm as it is unknown whether the person actually perished.

Time constraints in the newspaper search necessitated the narrowing of the search criteria at the risk of missing extra statistics. For each of the six eastern provinces that have had to cope with hurricanes the largest newspaper was chosen to be the primary reference as it was not feasible to consult more than one publication in each province, except as noted below. The newspapers consulted were the Chronicle Herald (Halifax, Nova Scotia), Telegraph Journal (Saint John, New Brunswick), the Guardian (Charlottetown, Prince Edward Island), the Evening Telegram (St. John’s, Newfoundland), the Gazette (Montreal, Quebec), the Toronto Star (Toronto, Ontario), and the Globe and Mail (Ottawa, Ontario).

All tropical cyclones were analyzed to determine if they would likely have affected Canadian land or people. Dates that storms might have impacted Canada were recorded and organized by province. Often, several provincial newspapers were consulted for a single storm because of the far-reaching effects of a hurricane. The newspaper was consulted from the date the storm entered the response zone to one week after the storm had passed or until articles ceased being published on a storm. With more time the search period would have been extended to attempt to gather more storm details, such as insurance statistics, that often take longer to calculate. The Toronto Star allowed a more complete search as the archive was digitized allowing a keyword search to be conducted.

Most of the newspapers were investigated at the National Library in Ottawa except the Chronicle Herald and the Toronto Star, which were investigated in Halifax, at the Nova Scotia Archives, the Halifax Regional Library, and from the online Toronto Star database. All articles located were printed and filed according to year. These articles will be retained by the Canadian Hurricane Centre for any future reference.


An extensive internet literature search was also conducted for additional information, although many of the statistics that the search yielded were not referenced and therefore deemed less reliable. Several valuable sources were discovered and the details were printed and added to the newspaper files. For example, Public Safety and Emergency Preparedness Canada, maintains a disaster database with damage and death statistics related to any natural or human induced event. Also, a website titled Lost at Sea contained many applicable reports of deaths due to tropical cyclones as recorded from smaller east coast publications.

Shipwreck Methodology

The Hurricane Climatology Project included significant research into shipwrecks along the Atlantic coast of Canada. This region has been home to a hypernate fishing industry, major shipping routes, and an avenue for immigration from European countries over the past couple of centuries. When unexpected tropical cyclones moved into the region, many of these vessels fled for safety. Unfortunately, many lives still have been lost as a result of these storms.

Valuable sources of shipwreck data are available throughout different locations in Nova Scotia. These include: theMaritime Museum of the Atlantic in Halifax, theFisheries Museum of theAtlantic in Lunenburg, theHalifax Regional Library and the Dalhousie Killam Memorial Library in Halifax. Shipwreck data was obtained from the major newspapers that were collected for other components of the climatology project and from a variety of shipwreck reference books composed by several authors within the Atlantic provinces. Also incorporated were several archives such as the Northern Shipwreck Database CD obtained from the Halifax Regional Library, an information kiosk station at the Maritime Museum of the Atlantic, and a previously compiled shipwreck database at the Maritime Museum of the Atlantic.

These sources of information were investigated at their respective locations within the Halifax region and relevant information was either printed or recorded and filed by year. A list of reference books containing substantial information was recorded and copies of each text have been placed in Environment Canada’s Library located in Dartmouth, Nova Scotia. To avoid research duplication in future years, a database of all reviewed sources has been created, with annotations noting the data’s relevance to this project.

All vessels which were found to have been affected by a tropical cyclone were compiled within a spreadsheet database. Shipwreck fatalities were considered to be any death within Atlantic Canada’s and the Great Lakes' marine forecast areas. For all notable shipwrecks, the date, location, and number of fatalities were recorded, as well as the vessel name, reference source, and comments concerning the vessel.  The spreadsheet was organized by year and storm date and sorted alphabetically by vessel name. Note that the nationality of the vessel or of lost persons was not deemed significant to this project.

Listed here are some significant obstacles frequently encountered in the research:

  1. The number of fatalities was unknown or several sources reported different values; this discrepancy was noted and a range of fatalities was recorded.
  2. Vessels that were lost were reported without any indication of their exact location.
  3. Vessels were reported missing but no further indication of status was given (i.e. it is unknown if the vessel returned safely or was subsequently confirmed to be lost). This became a major issue in the 1900 Galveston Hurricane. The range of fatality statistics therefore incorporates people who are classified as “missing.” The lower number within a range is a confirmed number of fatalities. The higher number includes the confirmed fatalities plus the additional missing and other unconfirmed fatalities (confirmed fatalities would be best defined as a report, whether it be a newspaper or reference book stating that a vessel was lost with a known number of crewmembers, with no contradictory articles being found).
  4. Some databases were organized only by vessel name and had no information of the shipwreck date. This frustrated the research, making it extremely difficult the determination as to whether or not the vessel was lost by a tropical cyclone (in many cases the determination was impossible).
  5. Misspelling of vessel names created ambiguity. Several cases were found where vessels were given different spellings of their names; however, further study confirmed the ships were the same. In such cases, both names were given.
  6. Ambiguity also arose due to the discrepancy between shipwreck dates based on meteorological data and data cited in the literature. In several instances, a shipwreck was reported to have been due to a hurricane on a given date, however, meteorological data (such as reliable track maps) show that the storm likely did not affect the area until the following day. Sometimes the loss of a vessel and the passage of the tropical cyclone may have been out of phase by as much as two days. Despite this, these vessels were recorded and noted that the date of wreckage may be incorrect.
  7. Much of the information gathered at the Maritime Museum of the Atlantic and other sources focussed on vessels registered in Nova Scotia or those lost in Nova Scotian waters. Few reference books covered the other Atlantic provinces in similar detail.

Other sources

The final sources consulted within this phase of the investigation were other organizations that have infrastructure to cope with hurricanes. For example, Search and Rescue maintains a database including statistics on operations they have participated in, such as hurricanes. The Red Cross is another potential source that may be consulted as the organization is normally the first non-governmental agency to respond to recovery efforts and the Institute for Catastrophic Loss Reduction specializes in research that lessens the cost of natural hazards.

Legend for Hurricane Track Information

Legend for Hurricane Track Information

Date modified: