- Snowfall relative to Upper Air Features (Synoptic Climatology Methods)
- Cook Method
- Garcia Method
- The Magic Chart
- Satellite Techniques
- LEMO Technique
- Snowfall relative to Upper Level Features (Synoptic Climatology Methods)
Summary:
These methods predict the distribution and type of precipitation
relative to the position of the 850mb, 700mb, or 500mb vorticity
maxima, the thickness patterns, and temperature fields. The most
common of these rules of thumb dates back to the 1960's when
a study by Goree and Younkin (1966) compiled a synoptic climatology
from 81 cases between 1963-1965 and concluded that the most favorable
location for the occurrence of heavy snow is 2.5 degrees to the left
of the track of the 500mb vorticity maximum during the following
12 hours.
For a more complete description of these techniques and the steps
involved in forecasting snow amounts relative to upper level features,
see the procedure page.
850mb low and Precipitation Pattern
Cook Method
(Cook, B.J., 1980)
Summary:
The Cook method predicts the areal extent of snowfall
for a 24-hour period based on the 200mb temperature field.
The temperature at 700mb is used to modify the forecast.
This is based on the idea that the thermal pattern at 200mb
reflects the strength of the system occurring at lower levels,
a concept which has since been illustrated nicely by the
Potential Vorticity perspective. This method also makes assumptions
about the influence of adiabatic cooling with ascent on the
upper air temperature distribution.
Ingredients Table for Cook Method
|
Forcing/Vertical Motion |
No |
| Instability |
No |
| Moisture |
No |
| Temperature |
Yes |
| Efficiency |
No |
Forecasts for 24 hours
Considers features through 200mb
Predicts areal extent of snowfall
Technique:
The steps involved in forecasting snow amounts with the Cook method
are presented on the procedure page.
Assumptions and Limitations (from Cook, 1980):
- Unreliable technique in Spring and early Fall and should not
be used before October 15 or after March 10. Late spring storms
have been observed with little temperature advection at 200mb and
others have occured under the strongest 200mb warm advection rather
than the coldest air, as used in winter.
- Heavy snow should not be forecast beneath a strong confluent
200mb flow. The heavy snow will usually be south of the confluent zone.
- Heavy snow should not be forecast south of the surface low, even
if the 200mb warm and cold pockets are south of the surface low.
Comments and Suggestions for Use:
It is interesting to relate this technique to the concept of
Potential Vorticity. A maximum in Potential Vorticity (PV)
corresponds to upper level temperature anomalies and is
associated with surface cyclones. From this perspective,
this technique is more useful for the synoptic "big picture"
than mesoscale precipitation processes.
If the PV perspective is used in conjunction with some of the Upper Air
Relationships to snowfall, a result similar to that predicted by Cook
is obtained. It will be interesting to look into a relationship
between areas of vertical motion and isentropic ascent associated with PV max or otherwise (TROWAL).
[pv/temp at 200 show good agreement for jan 19, but in stronger apr1 storm
pve is deeper than 200mb so don't see great agreement. instead of 200mb, use tropopause.
pv at 700 can reflect lhr causing low pv max and temperature advection. temperature
"advection" vs temperature misused].
example: 19 January 1995
-
Garcia Method
(Garcia, C., 1994)
There is now an updated version of the Garcia paper:
Garcia, C., Jr., 2000: Forecasting snowfall using mixing ratios on an isentropic surface: an update. NWS Technical
Memorandum CR-116.
Summary:
The Garcia Method predicts snowfall totals for a 12-hour
period using mixing ratio on an isentropic surface. The forecaster
decides on an area of concern where forcing for ascent is expected
and chooses an isentropic surface which intersects the 700-750mb layer
over this area. The average mixing ratio on this surface over the area
of concern multiplied by two is the amount of snow in inches predicted
by the Garcia method. Average mixing ratio is defined as the average
between the mixing ratio at the initial time and the maximum mixing ratio
predicted by advection during a 12-hour period.
Ingredients Table for the Garcia Method
| Forcing/Vertical Motion |
Left to Forecaster |
| Instability |
No |
| Moisture |
Yes |
| Temperature |
No |
| Efficiency |
No |
Forecasts for 12 hours
Considers features through 700mb
Predicts amount of snow for a given location
Technique:
The steps involved in forecasting snow amounts with the Garcia method
are presented on the technique page.
Assumptions and Limitations:
Not applicable to lake or orographic effects, possibly not applicable
to storms on the East or West coasts because of the additional ocean moisture
and mountains which were not accounted for in the development of the technique.
The paper emphasizes that this technique should be a part of a comprehensive approach
and that it is intended to answer the question of how much but not where.
Comments and Suggestions for Use:
The forcing component of this technique should be enhanced, perhaps
by incorporating a Q-vector analysis in choosing the area of concern (see
Garcia Gempak script page for details on this). Also, a complete forecast
would include consideration for instability, perhaps through the use of
Equivalent Potential Vorticity (saturated, geostrophic), see discussion in
the ingredients section.
Using this method, it might be tedious to find full extent of snow band.
It forecasts for the snowfall for a given cross-section only, and a number
of cross-sections
are needed to get a picture of the distribution. This is important to know
to determine how sensitive the forecast is to slight changes in storm
track, for example, if the forecast area is near on a boundary between high snowfall
amounts and little snow.
It might be useful to shorten the time of consideration from 12 hours to
6 hours. This would require a re-evalution of the mixing ratio to snow ratio,
perhaps it would simply be half. By shortening the time of consideration, it will
be easier to make sure all ingredients are in place for the next 6 hours
before deciding how much total accumulation.
The empirical relationships were developed with a particular
technique in mind so care should be taken with any modifications.
One element, the moisture advection calculation might be investigated further.
It has been found that the current method over
or under estimates the actual change in mixing ratio because it doesn't account for the storm motion. This could be remedied by using a moving
coordinate system, but we do not yet know if "correcting" this element would
be appropriate since the technique was derived empirically with a specific
procedure in mind. Some things to think about throughout the season in
computing the mixing ratio: Should we use the average between mixing ratio
at t=0 and the amount advected in 12 hours? Or, the average between the
mixing ratio now and that at t = 12 hours? Or, the maximum mixing ratio
over the 12 hour period?
Gempak Script:
A complete description and tutorial of a UNIX/Gempak script which
automates the forecast process of the Garcia method can be found on the
Garcia gempak script page.
Magic Chart
Summary:
The Magic Chart uses 700mb Net Vertical Displacement (NVD) and 850mb Temperature
to predict inches of snowfall in a 12-hour period.
Ingredients Table for the Magic Chart
| Forcing/Vertical Motion |
Yes |
| Instability |
No |
| Moisture |
Left to Forecaster |
| Temperature |
Yes |
| Efficiency |
No |
Forecasts for 12 hours
Considers features through 700 mb
Predicts both extent and amount
Technique:
Overlay:
- Net vertical displacement (NVD) in mb for air that will
arrive at the 700mb level at 24 hours. NVD is calculated
from LFM data by a trajectory model as an AFOS product.
- 850mb 12 hour temperature from the NGM.
Where the greatest NVD coincides with a temperature region
between -3C and -5C is where the heaviest
snowfall is likely to occur.
Furthermore, snowfall amounts may be computed from the
table below. The snowfall given is a forecast for
the 12-hour period between 12 and 24 hours after the initial
forecast time. Notice that the snowfall amount in inches
is the millibars of vertical displacement divided by 10.
| Net 12-hour Vertical Displacement |
12-hour Snowfall |
| 20-40mb |
2-4" |
| 40mb |
4" |
| 80mb |
8" |
| 100mb |
10" |
| 120mb |
12" |
| > 140mb |
> 14" |
Assumptions and Limitations:
- Does not apply to mesoscale snowstorms, including topographic and lake effect.
- Adequate moisture should be forecast or available. This is defined as a Temperature-Dew Point spread of no more than a few degrees at 850mb and 700mb or a 1000-500 relative humidity of 90% or greater.
- LFM and trajectory model must be trusted through 24 hours, and the NGM through
the first 12 hours.
Comments and Suggestions for Use:
This technique assumes that all precipitation-generating upward vertical motion
occurs below 700mb and in a vertical column directly above the point where the
trajectories cross 700mb. A trajectory analysis with Vis5D shows that this air
parcels do not rise directly up from the surface to 700mb but often traverse
considerable distances in the horizontal.
- Satellite Techniques
Summary:
These techniques predict the location of snow bands using satellite
infrared images. Based on the location of the vorticity maximum,
coldest cloud tops, and the dry intrusion, these techniques provide
a means to improve short term (a few hours in advance) forecasts
without the reliance on model data.
Technique:
Below is a short description of common techniques employed in
forecasting snow amounts with satellite imagery. The forecaster
is referred to Johnston (1995), Morrison (1989), Beckman (1987)
for the full technique description (or to the in-house
authority Ned Johnston).
- For a weaker system without a strong surface low or other significant
synoptic characteristics,
expect the heaviest snow to develop 1 degree of latitude (100km)
to the left of the axis of greatest cyclonic shear ("shear zone" in
image below).
from Johnston (1995)
- The south edge of the coldest cloud tops in IR satellite
imagery approximates the southern boundary of the heavy snow band.
- A line drawn through the center of the coldest cloud tops
approximates the northern boundary of the most significant snowfall.
from Morrison (1989)
- LEMO Technique
Maximum Snowfall in Inches =
(Maximum Absolute Vorticity - 10) * (30/s)
where s is the speed of the vorticity maximum at 500mb in knots.
Return to Winter Forecast Ingredients Page
