TP 10820 E



Prepared by

MacLaren Plansearch (1991) Limited
Suite 200, Park Lane Terraces
5657 Spring Garden Road
Halifax, Nova Scotia
B3J 3R4

Prepared for

Transportation Development Centre
Policy and Coordination Group
Transport Canada

March 1991


"The contents of this report reflect the views of the authors and not necessarily the official views or opinions of the Transportation Development Centre, Policy and Coordination Group, of Transport Canada."


The sponsorship and technical support given in the preparation of this work is acknowledged as follows:

  • The sponsorship and technical support given in the preparation of this work is acknowledged as follows:
  • Transport Canada - Transportation Development Centre
  • Transport Canada - Canadian Coast Guard-Ship Safety
  • Energy, Mines and - Program for Energy R & D
  • Resources Canada
  • Environment Canada - Canadian Climate Centre
  • Fisheries and Ocean - Marine Environment Data Service
  • Lloyd's Register of Shipping

Publication Data Form


Offshore exploration and shipping activities are affected by the climate conditions in the area. With an accurate description of the operating environment, ships and other marine structures can operate safely in these areas. This study was initiated by the Transportation Development Centre (TDC) on behalf of the Canadian Coast Guard, Transport Canada to develop a wind and wave climate atlas for the east coast of Canada, Gulf of St. Lawrence, and the Great lakes. The objective of this project is to collect, consolidate, and present the existing environmental data in a format which could be used to assess the strength and operational requirements of vessels and other marine structures for operation in these regions.

The atlas is divided into three volumes describing the following regions:

Volume I - The East Coast of Canada;

Volume II - The Gulf of St. Lawrence

Volume III - The Great Lakes.

Each region was divided into a number of subareas (or sites) to represent the various conditions within the region. The division of the subareas was based on the availability of data within the region, shipping lanes and fishing grounds, and the meteorological conditions within each sub area.

This volume presents the Great Lakes region. The Great Lakes represents five isolated bodies of water, interconnected by narrow constrictions, thereby eliminating any dynamic coupling between individual lakes. The region is divided into eight sub areas, as shown in Figure 1, where each of Lakes Ontario, Erie and Michigan is presented by a single point (or a sub area), Lake Huron by two subareas, and Lake Superior by three sub areas. The divisions are based on the major shipping routes and the consideration of wave climate in each of the Great Lakes.


An extensive review and evaluation of all data sources, user requirements, analysis techniques, etc. is provided in a separate report (Phase I Report, MacLaren Plansearch Limited, 1989, Transport Canada Publication #TP10867E). Examination of available measured data showed that there is a severe limitation in data coverage, both spatially and temporally. The review of previous hindcast studies for the Great Lakes showed that there are sufficient gaps in measured wave data to warrant the use of hindcast models. As concluded in Phase I Report, in spite of uncertainties of hindcasting methods, the immediate need for wave data in this area makes an approach via hindcasting the only viable alternative.

East Coast

Figure 1. East Coast Study Sub-Areas

  1. Lake Ontario
  2. Lake Erie
  3. Lake Huron South
  4. Lake Huron North
  5. Lake Michigan
  6. Lake Superior East
  7. Lake Superior Centre
  8. Lake Superior West

The data set used to produce the climatological statistics presented herein is from the U.S. Army Corps of Engineers, Waterways Experiment Station (WES) wind and wave hindcast model (Hubertz, 1989); a 32 year hindcast for the five Great Lakes. An extensive review and evaluation of the WES hindcast model, analysis techniques, etc. is provided in a separate report.

A database was compiled from the WES wind and wave hindcast model, archived at MEDS, Ottawa, covering the time period from 1956-1987. The database contains wind and wave hind cast data at three hourly intervals, for a number of the WES model grid point locations.

Data from each WES model grid point contained in the database was subject to an initial statistical analysis to find average and maximum significant wave height and wind speed, the ninety-five percent upper limit values, and prevailing wind and wave direction. These statistics, and the known wind/wave climate, were used to chose a single WES model grid point, for each sub area, to be presented in this publication. The location (grid point) which represents more severe climate was selected to represent the entire subarea.


Wind fields for the WES hindcast model are formed from data available for a number of wind recording stations. The stations used include land stations encompassing the lakes, and National Oceanographic and Atmospheric Administration (NOAA) buoys present in four of the lakes. NOAA buoy data are not available during the winter months (that is, from November-December to April-May). The wind speeds at the recording stations are reported to the closest knot; direction to the closest ten degrees. These data were corrected to an edited time series for each station. The buoy wind speeds were adjusted by a power law relationship, with an exponent of 1/7, to the 10 m level, from 6.1 m. For the land stations, an overland to overwater wind conversion was applied. It included correction for the difference in surface roughness between land and water, and a correction for the effects of air-sea temperature differences (Resio and Vincent, 1977). The wind fields produced in this way were then used to provide the input winds for each model grid point. The scheme used to interpolate the winds was, for each model grid point, to determine weighting factors, produced by an inverse power law, that depend on the physical location of the land and buoy stations with respect to the model grid points. For more information, see Hubertz (1989).

The wind data used in the analysis presented in this atlas present one-hour mean winds at 10 m above mean sea level. Following conversion factors may be used to convert wind speed to other averaging periods (U.K. Department of Energy, 1977):

1-hr mean 1.0
10-minute mean 1.05
I-minute mean 1.17
3-second gust 1.34


The 32 year WES hindcast database used in this atlas is based on the Wave Information Study (WIS) deep water wave model developed by Resio (1989). The WIS model is a discrete directional spectral model which simulates wave growth, dissipation, and propagation in deep water. Spectra are represented by energy in discrete bands of frequency and direction (20 frequency bands x 16 directions). The model is driven by a wind source term which is determined using the wind fields described above. A wave-wave interaction source term controls the transfer of energy across frequency bands. The model reproduces wave growth with fetch and duration comparable to that observed in the Joint North Sea Wave Program (JONSWAP) experiment (Hubertz, 1989).


  •  It should be emphasized that the WES model hindcasts used in this atlas represent deep water wave conditions, and therefore must be treated as such. Shallow water effects should be considered in coastal areas.

  • Wave height used in this atlas represents significant wave height (Hs or H1/3) which is defined as the average of the one-third highest waves in a wave record; it can be estimated from the energy spectrum (or total variance) by:
    TDC Atl1as

  • The ratio of Hmax/Hs (where Hmax is the maximum individual wave height in a wave record) is of great engineering concern, since Hmax and its associated period are required for engineering design (e.g. fluid loading calculation). This ratio can be estimated using suitable wave height probability distribution (e.g. Longuet-Higgins (1952), Longuet-Higgins (1980), Forristall (1978)). Assuming a Rayleigh distribution, which has been widely used for deep water waves, the following relation may be used as a first approximation:
        Hmax= 1.85 Hs
    This value can vary significantly depending on conditions such as water depth, wave period, etc. and should be used with caution.


Extreme value statistics were calculated by using the method of moments (MOM) to fit a Gumbel distribution to peak storm wind and wave values. The criteria for selecting wind and wave storms were wind speeds and significant wave heights surpassing a threshold of 30 knots and 2.5 m, respectively, for a minimum duration of 24 hours. From this, a preliminary list of potential storms was obtained and verified. The peak wind speed or wave height for each storm was then identified and used in the extreme analysis. The top 32 storms in each area were used to estimate the design values for a given recurrence interval or return period (e.g. 100 year or probability of exceedance of 0.01). See Canadian Climate Centre (1991) and Swail et al. (1989).


The WES wave hindcast assumes a median ice coverage of the Great Lakes for the period mid-December through mid April (Hubertz, 1989). The median ice cover was defined by ice concentration of 5/10th or greater (which is treated as solid land) from 20 years of ice observation (1960-1979).


any marine engineering applications require knowledge of the shape of the wave spectra. The analysis used in this atlas is based on the six-parameter model of Ochi and Hubble (1976). To generate the representative spectrum for a given sea-state value, the model function was fitted to the measured or hindcast data. Statistical analysis of the fitted parameters leads to a family of spectra, i.e. most probable spectrum and a set of 95% confidence spectra. Six sea-state classes were considered: Hs=0.5-2 m, 2-3 m, 3-4 m, 4-5 m, 5-6 m, and greater than 6 m.


The following is a list of statistics provided for area xx (and the corresponding page number):

Wind Speed Statistics
Annual Percentage Occurrence xx-1
Annual Percentage Exceeding xx-1
Annual Frequency of Occurrence by Direction (Wind Rose) xx-1
Annual Percentage Exceeding for Given Time Durations (Persistence) xx-1
Extreme Analysis xx-1
Monthly Frequency of Occurrence by Direction (Wind Rose) xx-2,3
Monthly Percentage Occurrence xx-4,5
Monthly Percentage Exceeding  xx-6,7
Monthly Statistics (mean, maximum, minimum, 95% limits, etc.) xx-8
Annual Percent Frequency of Occurrence by Direction (Table) xx-8
Significant Wave Height Statistics
Annual Percentage Occurrence xx-9
Annual Percentage Exceeding xx-10
Annual Frequency of Occurrence by Direction (Wave Rose) xx-9
Annual Percentage Exceeding for Given xx-10
Time Durations (Persistence) xx-10
Extreme Analysis xx-10
Monthly Frequency of Occurrence by Direction (Wave Rose) xx-11,12
Monthly Percentage Occurrence xx-13,14
Monthly Percentage Exceeding xx-15,16
Monthly Statistics (mean, maximum, minimum, 95% limits, etc.) xx-19
Annual Percent Frequency of Occurence by Direction (Table) xx-19
Most Probable Spectra xx-21
Wave Spectral Coefficients (Table) xx-22
Wave Peak Period Statistics
Annual Percentage Occurrence xx-9
Annual Percentage Exceeding for Given Time Durations (Persistence) xx-10
Monthly Percentage Occurrence xx-17,18
Joint Probability Statistics
Annual Occurrence of Significant Wave Height and Wind Speed xx-20
Annual Occurrence of Significant Wave Height and Peak Period xx-20


Canadian Climate Centre, 1991. Wind/Wave Hindcast Extremes for the East Coast of Canada. Report prepared by MacLaren Plansearch Limited and Oceanweather Inc. under DSS contract #KM169-7-6678.

Forristall, G.Z. 1978. On the Statistical Distribution of Wave Heights in a Storm. J. Geophysical Res. 83: No. C5, 2353-2358.

Hubertz, J.M., 1989. A Wave Hindcast for the Great Lakes 1956-1987. Proceedings of the 2nd International Workshop on Wave Hindcasting and Forecasting, Vancouver, B.C., April 25-28 1989, pp. 171-181.

Longuet-Higgins, M.S., 1952. On Statistical Distribution of the Heights of Sea Waves. J. Mar. Res. 11: pp. 245-266.

Longuet-Higgins, M.S., 1980. On the Distribution of the Heights of Sea Waves; Some Effects of Nonlinearity and Finite Band Width. J. Geophysical Res. 85: No. C8, pp. 1519-1528.

MacLaren Plansearch Limited, 1989. Preparation of a Wind and Wave Climate Atlas, Phase I Interim Report - Initialization, Planning, Data Compilation, and Methodology. Report submitted To Transportation Development Centre, Transport Canada, Montreal, Quebec, March 1989, TP 10867E.

Ochi, M.K. and E.N. Hubble, 1976. Six Parameter Wave Spectra. Proceedings of the 15th Coastal Engineering Conference. Honolulu, pp. 301-328.

Resio, D.T. and C.L. Vincent, 1977. Estimation of Winds Over the Great Lakes. American Society of Civil Engineering Waterway Port. and Coast. Ocean Div. J. 102, 265-283.

Resio, D.T., 1989. A revised Deep Water Wave Model. Coastal Engineering Research Centre, U.S. Army Engineers Waterways Experiment Station, Vicksburg, Mississippi.

Swail V.R., V.J. Cardone and B.M. Eid, 1989. Wind/Wave Hindcast Extremes for the East Coast of Canada. Proceedings of the 2nd International Workshop on Wave Hindcasting and Forecasting, Vancouver, B.C.,

Rose Key