Update: april 2008

CMC

WINTER SIGNIFICANT WEATHER

DESCRIPTION OF THE “WEB” CHARTS

1- SNOW ACCUMULATIONS - SNOW SQUALLS

2- FREEZING RAIN/DRIZZLE - HEAVY RAINFALL

3- BLOWING SNOW / BLIZZARD

4- WIND CHILL – SEVERE WINDS

ISSUE TIMES and FORECAST PERIODS

Charts issued at 0400Z- based on 00Z GEM regional

- 12-36h period - Valid from 12Z/d to 12Z/d+1

- 12-48h period - Valid from 12Z/d to 00Z/d+2

- 24-48h period - Valid from 00Z/d+1 to 00Z/d+2

Charts issued at 1600Z- based on 12Z GEM regional

- 12-36h period - Valid from 00Z/d+1 to 00Z/d+2

- 12-48h period - Valid from 00Z/d+1 to 12Z/d+2

- 24-48h period - Valid from 12Z/d+1 to 12Z/d+2

1- SNOW ACCUMULATIONS - SNOW SQUALLS

DESCRIPTION

a) Snow Accumulations

Areas where accumulations of 5 cm or more are forecast during the valid period of the chart are displayed.

Areas of accumulations are represented as follows:

b) Snow squalls

Snow squalls are indicated by magenta areas, representing all forecast occurrences of squalls during the valid period of the chart.

Areas are represented as follows:

DETERMINATION OF THE AREAS

a) Snow accumulations

Snow accumulations are generated from the regional model QPF and the “snow” type precipitation from the “Bourgouin” diagnostic scheme. This diagnostic is the one used operationally to generate precipitation types found on the 4-panel “pressure and pressure tendency” chart, at 6-hour intervals.

Hence, forecast snow accumulations for a given period are obtained as follows:

- the precipitation type forecast by the model ("Bourgouin” type) is examined at the two extremities of each 3 hour interval in the period under consideration. If it is possible to deduce that there will be no change of precipitation type during a given 3-hour interval, and only snow is expected during this period, then the QPF for this interval is considered to be “snow”. For this to happen, one of the following combinations of types must be forecast at the extremities of the 3-hour interval:

T T+3h

Nil Snow

Snow Snow

Snow Nil

where ... Nil = no precipitation

- the QPFs for the intervals where the above combinations of types are encountered during the integration period under consideration (12-36h, 12-48h, 24-48h) are thus added up, 3 hour period by 3 hour period. The forecast snow QPF is thus obtained for the entire period, and is expressed in cm, using the usual factor of 10 between water equivalent and snow depth.

The following adjustments, however, are considered:

Modification of the model “snow” type

There are 2 exceptions where the "Bourgouin" type will be modified by the program which produces the “Snow and snow squall” chart:

1- British Columbia and Yukon

In this sector, the resolution of the model topography is such that the terrain elevation in the valleys is estimated “too high”. This bias results in model surface temperatures which are, for the most part, too cold, and precipitation types have a decided bias toward solid precipitation. The result is that the regional model often forecasts snow in the valleys where rain would more reasonably be expected.

Empirical experience leads to the application of the following rule in this sector:

- where model terrain is below about 1600 metres in British Colombia and southern Yukon, i.e. in the "white" area of the chart below and then also including eastern pacific and northwestern corner of the United States:

- if the model forecasts snow ("Bourgouin" type), and

- if the 1000-850 hPa thickness is forecast above 131 dam,

then the program will change the “snow” type to “rain” at the grid point and projection time considered.

2- Snow squalls

The manner in which snow squalls are diagnosed by the program is explained below in section (b). It should be stated here, however, that snow squalls are obtained when the model forecasts type “snow” at a given location, under the conditions of low level instability described below. In this case the program artificially changes the “snow” type to “snow squall” type for the grid point and projection time under consideration.

As such, the program which generates the forecast snow accumulation charts takes into account the modifications of the precipitation types described above, and does not add the QPF for the 3-hourly intervals where these modifications have taken place.

Modification of the 10 times equivalence factor

The 10 times equivalence factor between model QPF and snow accumulation is used except in the following case:

- if the model forecasts snow over a 3 hour period, and

- if the average forecast 850-700 hPa thickness is above 154 dam for this time interval,

then the program will apply a rain-snow factor calculated according to an algorithm developed by B. Murphy (Ontario), based on average 1000 to 700 hPa thicknesses for the 3 hour period.

In such cases, the result will be a factor less than 10, thus lowering the total snow accumulations.

The algorithm is as follows:

RATIO = QSF / QPF (= 10 ... generally)

where... QSF = "Quantity of Snow Forecast"

QPF = "Quantity of Precipitation Forecast"

RATIO = 10 x [ (-0.1057 x DZ_moy) + 30.6 ]

(Murphy)

where... DZ_moy = DZ(1000-700) in dam

(averaged over 3 hours)

DZ(1000-700) = 1000-700 hPa thickness

b) Snow squalls

Two types of snow squalls are defined:

- snow squalls triggered by warm air flowing over relatively warmer bodies of water, such as the Great Lakes or the Gulf of St-Lawrence;

- snow squalls triggered by an intense cold front moving through an unstable airmass.

In fact, what is being looked for, in general, is to forecast the potential occurrence of the following meteorological phenomenon:

Heavy snow showers reducing visibility to near zero...

- visibility expected to drop to less than 1/2 mi;

- accompanied or not by blowing snow;

-* occurring suddenly and being of short duration (especially for frontal squalls);

-* may last for a much longer time in the case of lake/water effect squalls.

The forecast is thus independent of the duration of the phenomena...

* ANY OCCURRENCE (ONE OR MORE) OF SUCH SNOW SHOWERS EXPECTED DURING DE FORECAST PERIOD OF THE CHART SHOULD LEAD TO A FORECAST AREA OF SNOW SQUALLS.

N.B. THERE WILL BE NO FORECASTS OF SNOW ACCUMULATIONS ASSOCIATED WITH SNOW SQUALLS.

In all cases, snow squalls are produced when the surface temperature is near or below zero Celsius and, obviously, the temperature is below zero in the whole profile above the surface. To assure this, the following preliminary conditions are imposed upon the snow squall program:

- surface temperature <= 2°C

- 1000-850 hPa thickness < 129 dam

- forecast precipitation type = SNOW

The last condition above, that the forecast precipitation type be snow, also assures, in indirect fashion, that the humidity in the forecast profile is sufficient to produce precipitation.

However, the GEM regional model with 15 km resolution is more easily showing precipitation than the previous 24 km resolution model when it is relatively dry at the surface. So the following condition has been added:

- surface dew point depression <= 5°C

In addition to these preliminary conditions, conditions of one of the following three categories must be forecast:

1- Squalls - water bodies

Squalls caused by the circulation of cold air over bodies of relatively warm water are mainly related to the following parameters:

- low-level instability;

- low-level humidity.

As indicated above, for the purpose of the program used, low-level humidity is judged to be sufficient if the model forecasts precipitation (in the form of snow) and that the surface dew point depression (T-Td) is smaller than or equal to 5°C.

For the low-level instability, the following instability criteria are used:

· [ DZ(850-700) - DZ(1000-850) ] <= 20 dam

where... DZ(850-700) = 850-700 hPa thickness

DZ(1000-850) = 1000-850 hPa thickness

This criteria assures a certain level of instability in the 1000-850 hPa layer with respect to the 850-700 hPa layer.

· (Tsfc - T850) >= 10°C

where... Tsfc = surface temperature

T850 = 850 hPa temperature

This criteria assures that the temperature profile in the lower layer tends toward dry adiabatic.

2- Frontal squalls

In addition to the low-level instability and humidity criteria, frontal squalls require lift caused by the passage of a vigourous cold front.

This condition has been related to low-level convergence, normally associated with an isobaric trough. For the purposes of the program, a minimal value of 1000 hPa geostrophic vorticity is required to satisfy this condition.

As well as the preliminary conditions described above, the following conditions should be met:

· (Tsfc – T850) >= 5°C

· [ DZ(850-700) - DZ(1000-850) ] <= 19 dam

· QG-1000 >= 10 x 10E-5 /sec

OR

· (Tsfc – T850) >= 5°C

· [ DZ(850-700) - DZ(1000-850) ] <= 21 dam

· QG-1000 >= 25 x 10E-5 /sec

where... DZ(850-700) = 850-700 hPa thickness

DZ(1000-850) = 1000-850 hPa thickness

QG-1000 = 1000 hPa geostrophic

vorticity

3- Squalls due to strong winds

In certain cases where the low level instability is high, and the 850 hPa winds are quite strong, a snow shower occurring at the same time as a low level downdraft could result in a reduced visibility. What is required is a profile tending toward dry adiabatic between the surface and 850 hPa, under conditions of strong winds at the top of this layer.

The program thus finally looks for the following conditions to justify snow squall potential:

- (Tsfc - T850) >= 10°C

where... Tsfc = surface temperature

T850 = 850 hPa temperature

- UV-850 >= 40 kts

where... UV-850 = 850 hPa wind speed

=====================================================

=====================================================

2- FREEZING RAIN/DRIZZLE - HEAVY RAINFALL

ISSUE TIMES

Charts issued at 0400Z- based on GEM regional 00Z

- 12-36h period - Valid from 12Z/d to 12Z/d+1

- 12-48h period - Valid from 12Z/d to 00Z/d+2

- 24-48h period - Valid from 00Z/d+1 to 00Z/d+2

Charts issued at 1600Z- based on GEM regional 12Z

- 12-36h period - Valid from 00Z/d+1 to 00Z/d+2

- 12-48h period - Valid from 00Z/d+1 to 12Z/d+2

- 24-48h period - Valid from 12Z/d+1 to 12Z/d+2

DESCRIPTION

a) Rain amounts

Areas where amounts of 30 mm or more are forecast during the valid period of the chart are displayed.

Areas of 30 and 50 mm are represented as follows:

B) Freezing rain

Areas of forecast freezing rain in the amounts of 2 mm or more, or for a duration of 3 hours or more are represented as follows:

C) Freezing drizzle

Areas of forecast freezing drizzle for a duration of 6 hours or more are represented as follows:

DETERMINATION OF THE AREAS

a) Rain amounts

Amounts of rain are generated from the regional model QPF and from the precipitation type “rain”, from the “Bourgouin” diagnostic scheme. This diagnostic, done three hourly through the forecast period, is the one used operationally to generate precipitation types as found on the 4-panel “pressure and pressure tendency” charts.

In this way, rain amounts forecast for a given period of time are obtained as follows:

- the precipitation type forecast by the model ("Bourgouin” type) is examined at the two extremities of each 3 hour interval in the period under consideration. If it is possible to deduce that there will be no change of precipitation type during a given 3-hour interval, and only rain is expected during this period, then the QPF for this interval is considered to be “rain”. For this to occur, one of the following combinations of types must be forecast at the extremities of the 3-hour interval:

T T+3h

Nil Rain

Rain Rain

Rain Nil

where ... Nil = no precipitation

- the QPFs for the intervals where the above combinations of types are encountered during the integration period under consideration (12-36h, 12-48h, 24-48h) are thus added up, 3 hour period by 3 hour period. The forecast rain QPF is thus obtained for the entire period.

It is understood that if “snow” or "freezing rain" type was modified to “rain” type over British Columbia or southern Yukon due to the considerations in section (1a-1) above or (2-b) below, then the rain obtained is added by the program within the rain QPF.

b) Freezing rain

As shown in the above (2-b) description of the charts, 2 criteria are considered when generating a freezing rain area on one of the forecast charts (12-36h, 12-48h, 24-48h):

- a forecast amount of 2 mm or more;

- a duration of 3 hours or more.

Occurrence of freezing rain at 3 hours intervals is looked for as follows through a given forecast period (12-36h, 12-48h, 24-48h): for a given prognostic time "T", freezing rain is said to be forecast if one of the two following conditions is encountered:

- freezing rain is forecast by the diagnostic

"Bourgouin" scheme ("ZR" variable in

"regdiag");

- freezing rain amount of 0.5 mm or more is

forecast by the pronostic variable "FR" of the

model for the 3 hours period ending up at time

"T".

From the above diagnostic, freezing rain amount is obtained by adding the amounts given by the "FR" variable for each 3 hours period ending up at time "T" of a positive freezing rain diagnostic through a forecast period.

For the duration criteria, a minimum duration of three hours of freezing rain is said to be forecast if the above diagnostic is positive at both ends of any 3 hours interval within the considered forecast period (12-36h, 12-48h, 24-48h), by example:

T T+3h

FZRA FZRA

Modification of the above freezing rain diagnostic

However, in some cases where it seems justified to do it, the above freezing rain diagnostic can be artificially changed to rain of freezing drizzle by the program.

1- Modification of FZRA to RA – BC + Yukon

In this sector, the resolution of the model topography is such that the terrain elevation in the valleys is estimated “too high”. This bias results in model surface temperatures which are, for the most part, too cold, and precipitation types have a decided bias toward "solid" or "freezing" precipitation. The result is that the regional model sometimes forecasts freezing rain in the valleys where rain would more reasonably be expected.

Then, similarly as for the modification of snow to rain in this sector, the following rule is applied:

- where model terrain is below about 1600 metres in British Colombia and southern Yukon, i.e. in the "white" area of the chart below and then also including eastern pacific and northwestern corner of the United States:

- if the model forecasts freezing rain (according to the above described diagnostic), and

- if the 1000-850 hPa thickness is forecast above 131 dam,

then the program will change the “freezing rain” type to “rain” at the grid point and projection time considered.

2- Modification of FZRA to FZDZ - anywhere

For any diagnostic of FZRA described above, it is verified whether the model also forecasts a nose of warm air, i.e. temperatures above 0°C in the 850-700 hPa layer, according to the partial thicknesses technique.

If not, the “FZRA” so forecast is mainly associated with an above freezing layer only at low levels, i.e. in the 1000-850 hPa layer. It is then artificially converted to type “FZDZ”, freezing drizzle, at the grid point and projection time under consideration.

The rule applied is the following:

- if initial type forecast = FZRA, and

- DZ(850-700) < 154 dam

then resultant type = FZDZ.

One ends up with forecasts of "FZDZ" and "FZRA", which the chart production program considers separately to produce fields of potential freezing drizzle and freezing rain, respectively.

c) Freezing drizzle

For a freezing drizzle area to show up, only a duration criterion is consedered. There is no amount criterion as for freezing rain areas.

So the criterion for a freezing drizzle area is:

- a duration of 6 hours or more.

A 6 hour period of FZDZ is obtained if FZDZ is diagnosed for at least 2 complete 3 hours intervals for a given forecast period (12-36h, 12-48h, 24-48h), by example:

T T+3h T+6h

ZL ZL ZL

T T+3h T+6h T+9h T+12h

ZL ZL Nil ZL ZL

=====================================================

=====================================================

3- BLOWING SNOW / BLIZZARD

DESCRIPTION

a) Blizzard

Areas where blizzard or near blizzard conditions are forecast during de valid period of the chart are displayed.

Areas are represented as follows:

b) Snow and blowing snow (S/BS)

Areas where blowing snow associated with a snowfall is forecast during the valid period of the chart are displayed as S/BS (snow and blowing snow) or marginal S/BS areas.

Areas are represented as follows:

In both cases above, we want indeed to depict situations where blowing snow is reducing visibility to 1 km (0.5 mile) or less for 6 hours or more during the valid period of a chart. In the case of blizzard, no snowfall is necessarily occurring. A mask has been designed on the grid used by the program to delimit usual or known areas where blizzard conditions are expected to show up and these conditions are looked for only in these areas (see chart showing mask below).

It is to be noted that, if S/BS conditions are depicted in delineated blizzard areas according to the mask, then these S/BS conditions will show up as "blizzard" conditions in these sectors. This has been done to prevent overlapping of blizzard and S/BS areas on the forecast chart, as both represent anyway a similar event, namely blowing snow reduding visibility to 1 km or less. Thus, in these sectors, blizzard or S/BS conditions will show up as brown areas on the forecast charts.

Elsewhere, namely outside of the delineated areas of the mask, S/BS conditions will be represented, as indicated on the legend of the chart, by orange areas. And blizzard conditions are not depicted in these sectors, so no overlapping will show up.

DÉTERMINATION OF THE AREAS

Duration

In both cases, "blizzard" or "snow and blowing snow (S/BS)", we look for events that last for at least 6 hours. This is done as follows:

- we first verify if criteria described below for depicting "blizzard" and "S/BS" conditions are met at 3 hours time steps of the forecast period of the regional model (Gemreg-15 km) from T+6h to T+48h;

- then a period of 6 hours or more of "blizzard" or "S/BS" is depicted if the criteria are met at each end or any 6 hours period within the valid period of the chart.

By example, for the 12-36h forecast period (T+12h to T+36h), a miminum 6 hours period is depicted in any of the following situations:

- criteria met at T+12h and T+18h

- criteria met at T+15h and T+21h

- criteria met at T+18h and T+24h

- criteria met at T+21h and T+27h

- criteria met at T+24h and T+30h

- criteria met at T+27h and T+33h

- criteria met at T+30h and T+36h

We do it this way in view of the intermittent nature of the phenomenon and also to "give a chance" to the model. We are thus preventing to request that the criteria be met a 3 consecutive time steps. It would be more restrictive, for example for the T+18h to T+24h period, to request that the criteria be met at T+18h, T+21h and T+24h instead or requesting that they be met only at T+18h and T+24h as we indeed do.

a) Blizzard

For depicting potential blizzard conditions, we request a minimum snow depth on the ground, negative or 0ºC temperatures and a minimum wind velocity. It is not necessary that snow be falling, but if there is precipitation, it must not be liquid.

So the preliminary conditions are the following:

- Snow on the ground >= 5 cm

- Tsfc = 0ºC

- Precipitation: NIL or Snow or Ice pellets

For the wind conditions, the minimum thresholds are dependent upon local effects and topography. These thresholds are thus different from one sector to the other of the mask delineating areas where blizzard conditions are depicted.

The map below shows the areas delineated by the mask and where blizzard conditions are mainly observed. Blizzard conditions will be forecast only in these sectors.

We have grouped together in 5 regions the sectors showing up on the map:

10- Dempster: north of Yukon territory

highway toward WRR (Rock River)

20- WRN Nunavut: arctic islands

+ northwestern Hudson bay coast

+ southwestern Nunavut

30+40- Churchill (MB)

+ northeastern Hudson bay coast

+ Hudson strait

+ western Ungava bay coast

+ southwestern Baffin coast

50+60+70 Eastern Baffin coast

+ northeastern Labrador coast

80- Southeastern Saskatchewan

+ southern Manitoba

The critiria used to depict wind velocity are based on the following fields:

- Vgisal... modulus of the vectorial sum of the geostrophic and isallobaric winds

- MaxWin... low levels max winds

- UV (0.995)... model wind velocity at eta level 0.995

- Ugeos... "x" component of the geostrophic wind following the grid used, this orientation being mainly "east-west" in the area where this parameter is used.

The thresholds values below are "empirical" and have been mainly determined by experience and verification, and also by adjustment of the values for specific events.

Near Blizzard

- Région 10: Vgisal >= 90 kts

Ugeos <= 0 kt (easterly wind)

- Région 20: Vgisal >= 30 kts

UV(0.995) >= 15 kts

- Région 30+40: Vgisal >= 50 kts

- Région 50+60+70: Vgisal >= 40 kts

MaxWin >= 40 kts

- Région 80: Vgisal >= 60 kts

Blizzard

- Région 10: Vgisal >= 110 kts

Ugeos <= 0 kt (easterly wind)

- Région 20: Vgisal >= 40 kts

UV(0.995) >= 15 kts

- Région 30+40: Vgisal >= 60 kts

- Région 50+60+70: Vgisal >= 50 kts

MaxWin >= 50 kts

- Région 80: Vgisal >= 80 kts

It is to be noted that the Ugeos parameter is only used for the Dempster area (region 10). We need to make sure that the wind is blowing easterly in this area to prevent wrong blizzard diagnostics. Blizzard mainly shows up in this area with easterly winds north of an "east-west" trough central Yukon.

b) Snow and blowind snow (S/BS)

Cases of blowing snow within snow storms are indeed a subset of the blizzard cases described above. However a significant difference for depiction is showing up mainly for the minimum wind velocity requested. Visibility being reduced not only by blowing snow but also by the snowfall, it is then easier to obtain low visibilities than for "pure" blizzard where visibility is only reduced by the snow on the ground being blown up. As a result, the criteria for winds are somewhat relaxed for S/BS events compared to blizzard events.

On the other hand, S/BS events can of course show up anywhere and not only in the "mask" areas where we limit blizzard depiction. The only restriction in the program is for mountainous areas where depiction is not performed where model terrain elevation is 700 meters or higher (see chart below). Wind conditions are quite variable in mountainous areas where observations are also quite sparse.

Preliminary conditions for depiction of S/BS events are the following:

- Tsfc = 0ºC

- Precipitation type = Snow

- Vgisal = x where x = f(sector)

We then request that the model forecasts snow where and when we perform a diagnostic as we are looking for a "snow and blowing snow" event. On the other hand, temperature must be forecast to be 0ºC or lower to ensure for "solid" precipitation and to allow for snow to be more easily blown up.

The wind criteria is based on one field only for S/BS depiction, namely the "Vgisal" field which represents the modulus of the vectorial sum of the geostrophic and isallobaric winds. The thresholds are as follows for the different forecast sectors:

Marginal S/BS

- Région 10: Vgisal >= 70 kts

- Région 20: Vgisal >= 30 kts

- Région 30+40: Vgisal >= 40 kts

- Région 50+60+70: Vgisal >= 40 kts

- Région 80: Vgisal >= 50 kts

- Elsewhere (model terrain < 700 m):

Vgisal >= 70 kts

S/BS

- Région 10: Vgisal >= 90 kts

- Région 20: Vgisal >= 40 kts

- Région 30+40: Vgisal >= 50 kts

- Région 50+60+70: Vgisal >= 50 kts

- Région 80: Vgisal >= 70 kts

- Elsewhere (model terrain < 700 m):

Vgisal >= 90 kts

=====================================================

=====================================================

5- WIND CHILL – SEVERE WINDS

DESCRIPTION

a) Wind chill

Areas where wind chill temperatures of -35°C or less are forecast for a duration of at least 3 hours during the valid period of the chart are displayed.

Areas of different WCT (Wind Chill Temperature) are represented as follows:

b) Severe Winds

Areas where winds of 90 km/h or more, regular or gust, are forecast during the valid period of the chart are displayed.

Areas are represented as follows:

DETERMINATION OF THE AREAS

a) Wind chill

Wind chill is computed with the Osczevski-Bluestein formulation, which is used to produce the operational "difax" charts of wind chill temperatures within the CMC winter severe weather package.

Values of WCT (Wind Chill Temperature) showing up on these charts are computed only when the forecast surface temperature (Tsfc) is less than or equal to 0°C and the forecast surface wind (Vsfc) is of 5 km/h or more. Otherwise, values of WCT showing up are simply the forecast surface temperatures.

So the WCT values are obtained as follows:

IF Tsfc = 0°C AND Vsfc >= 5 km/h

Then WCT computed with Osczevski-Bluestein equation

ELSE

WCT = T°

On the "IWEB" charts described here, areas of WCT are only displayed for values of -35°C or less and with expected duration of 3 hours or more during the valid period of the chart. The 3 hours duration criteria is met if a looked for WCT value is forecast at both ends of any 3 hours period within the valid period of the chart.

By example, for the 12-36h forecast period (T+12h to T+36h), a minimum 3 hours duration is depicted in any of the following situations:

- WCT criteria met at T+12h and T+15h

- WCT criteria met at T+15h and T+18h

- WCT criteria met at T+18h and T+21h

- WCT criteria met at T+21h and T+24h

- WCT criteria met at T+24h and T+27h

- WCT criteria met at T+27h and T+30h

- WCT criteria met at T+30h and T+33h

- WCT criteria met at T+33h and T+36h

b) Severe winds

The following fields are used for depiction of severe winds:

- UV (0.995)... model wind velocity at eta level 0.995

- Vgisal... modulus of the vectorial sum of the

geostrophic and isallobaric winds

- Visal... modulus of the isallobaric wind

- MaxWin... low levels max winds

- Tsfc–T850... difference between surface temperature and 850 mb temperature

- Difdz... DZ(850-700)-DZ(1000-850)

- WW_850... 850 mb vertical velocity

Difference of partial thicknesses:

Thickness(850-700) – Thickness(1000-850)

- UV_850... model wind velocity at level 850 mb

Wind velocity being dependant upon local effects and topography, we then need to apply a mask on the computational grid to allow for different criteria to be applied as a function of the identified areas. These areas are showing up on the map below.

Some of the criteria described thereafter are thus applied as a function of these defined areas but it is to be noted that some other criteria are “universal”, i.e. they do apply as well within and outside of the defined areas by the mask.

Sectors #1: west coast + east coast + arctic

Sectors #2: Hudson bay (south of 60°N)

WHO + YDP (sector of Hopedale + Nain)

Sectors #3: Fraser valley (YHE-WLY-YWL)

(Hope-Lytton-Williams lake)

YYF (Penticton)

Sectors #4: Great lakes

This defined area of the mask is not currently

being used. There is then no specific rules for

this area where “universal” criteria are thus the

ones that do apply.

As for the blizzard and S/BS criteria, the threshold values indicated below for the severe winds criteria are "empirical" and have been mainly determined by experience and verification, and also by adjustment of the values for specific events.

The criteria are described below following their order of priority, from the one with the highest priority (criteria #1) to the one with the lowest priority (criteria #7).

The program checks for severe winds potential in the following way: for a given grid point and projection time, the program checks if the criteria described below are met, from criterion #1 down to criterion #7. As soon as one criterion is met and then that a positive diagnostic for severe wind is done, the program stops checking.

Such a verification of the criteria is performed for each grid point at 3 hours intervals of prognostic times, namely from T+3h to T+48h.

For a severe wind to be forecast at a given grid point for a given forecast period (12-36h, 12-48h, 24-48h), it is sufficient that a positive severe wind diagnostic be done for this grid point for any of the prognostic times of this period.

If a positive severe wind diagnostic results for more than one prognostic time within the forecast period at this grid point, then the criterion said to be met for forecasting severe wind at this grid point for the considered period is the one with the highest priority met during the period.

It is finally to be noted that a few criteria only apply in areas where model (Gemreg-15km) terrain height is less than 700 meters. The chart below enhances in red the areas where the model terrain height is 700 meters or more.

Criteria (from de highest to the lowest priority)

1- Model wind – applies anywhere

UV (0.995) >= 40 kts

2- Intense low – applies where terrain height < 700 m

Vgisal >= 125 kts

MaxWin >= 50 kts

3- Isallobaric winds – applies where

terrain height < 700 m

Vgisal >= 125 kts

Visal >= 50 kts

4- Instability – applies anywhere

UV (0.995) >= 30 kts

AND

Tsfc–T850 >= 10°C AND UV_850 >= 50 kts

OR

Maxwin >= 40 kts AND Difdz <= 19dam AND

Tsfc–T850 >= 9°C

OR

Vgisal >= 50 kts AND Maxwin >= 50 kts AND

Difdz <= 20 dam AND terrain < 700 m

5- West coast – East coast – Arctic

(sectors 1 of the above chart)

Vgisal >= 50 kts AND MaxWin >= 50 kts

6- Southern Hudson bay + WHO/YDP

(sectors 2 of the above chart)

Vgisal >= 75 kts AND MaxWin >= 50 kts

Fraser valley + CYYF

(sectors 3 of the above chart)

Vgisal >= 75 kts AND MaxWin >= 60 kts

7- Katababic wind or subsidence + instability

(applies anywhere)

UV (0.995) ≥ 15 kts

AND

Vgisal >= 75 kts AND MaxWin >= 50 kts AND

WW_850 <= -0.8 pa/s

OR

Vgisal >= 100 kts AND MaxWin >= 40 kts AND

WW_850 <= -0.8 pa/s

OR

Vgisal >= 50 kts AND MaxWin >= 60 kts AND

WW_850 <= -0.6 pa/s AND Difdz <= 21 dam

=====================================================

=====================================================