Snow Event Rating Scale: Critique and Comment

[Jeff Miller]

Good day all,

I have attempted to produce a snow storm rating scale in the past. I ran into many problems producing such a scale as I ran into a great deal of barriers based on regional variabilities. For instance and example, as we know, a 3" snowstorm in Brownsville or Dallas is quite significant. This same amount of snow in Minneapolis is hardly worth batting an eye at. And that is just looking at basic accumulation amounts. You also must factor in travel impacts, economic impacts, visibility impacts, and life-threat impacts.

I have found that in order to produce a successful snowstorm rating scale one must focus on impact and not variabilities. With that said, I have devised a "snow event rating scale" that I shall post here.

The way this works is in this manner:

The rating system is a 4-point rating scale ranging from "Low" to "Extreme". Each rating is examined by the following impacts: Visibility, Travel, Economic and Life Impact. Therefore, the scale still has a great deal of variability depending on one's exact location and impact level.

To effectively "rate" a storm on this 4-point tier, the method is simple: The majority level of all 4 combined levels completes the total level of the storm, or an average of the four levels. For instance, let's take a storm that is high visibility impact due to ground blizzard conditions, moderate on travel impacts due to not being highly urban or rural, and low on economic impacts and low on overall life-threat. We have "visibility" as high, "Travel" as moderate, and two "low" ratings on economy and life threat. This could create a "LEVEL 2" snowstorm out of the 1 to 4 possible ratings. If you have Two "HIGH" ratings and Two "EXTREME" ratings out of the 4 categories, the lower rank of the two would be the "official" rating of the snow event: Three "EXTREMES" and one "HIGH" rating would be an "official" rating of extreme.

LIFE THREAT LEVEL is determined by the total combination of all events including temperature, chance of disorientation, windchill (hyperthermia chances), visibility, and accessibility to shelter and food.

The point of this scale is to

• Eliminate snowfall totals due to regional variabilities and focus on impact
• Create an easy to understand, impact-driven snow impact scale for the rating of winter weather events
• Create a simple to understand impact scale that can be understood by the public and the meteorologist alike
• Create a simple, 4-scale storm impact scale that can give appropriate emphasis to a storms' predicted intensity.
• Create a scale that can be applicable to all areas regardless of regional variability.

Your comments and feedback, critique, and analysis are welcomed. This is NOT a perfect scale, and is a first attempt I've posted here in some time. I welcome your criticism and this is intended to jump start conversation on something that is needed but avoided due to the problems of regional variability.

Thank you!

 

[Jeff Duda]

I think you could use snowfall measurements in a storm rating scale by comparing point measurements by the history of measurements at that point. For example, obtain all of the 24-hour snowfall records from all records between October 1st and April 30th for Minneapolis, MN and Dallas, TX. Make a frequency distribution (i.e., histogram) of these records and say a 24-hour snowfall above the 90th percentile is "extreme", above the 75th is "significant" etc. Obviously you are right that a straight-up comparison between cities would not work because the 90th percentile of 24-hour snowfall at Minneapolis is very different from that at Dallas.

It seems that your classes of impacts overlap each other. For example, visibility and travel impact share a lot of common symptoms. Same goes for threat to life...it's kind of a combination of the three other classes.

Also, is your scale an attempt to rate a storm overall (i.e., apply just one number to all areas that the storm affected) or is it just for a point location? If so, then you'll have to account for the spatial variability in impacts such as visibility and road quality.

I think you are onto something great, Jeff, but I think some improvements can be made. Good luck with this. I'd love to see how it turns out.

 

[Mike Kovalchick]

Nice. Thanks.

Just a few thoughts. I don't think your system addresses variability in a local area very well. For example, lake effect snow in the Great Lakes. The threat of power outages in certain situations such as wet snow on full foliage can be so significant that it should not be buried under a generic "economic" category. Is "economic" defined? How about storm duration? I see this every year..a 5 minute snow squall has far more impact than a 2 day moderate snowfall. Another example..the impacts of a given storm are far greater when they are unseasonal..for instance..the first storm of the season when travelers aren't use to icy weather. Also, this is a snowstorm index..what about mixed precip? Most storms in many parts of the country run the full gambit of precip types in a given scenario. How is that addressed?

 

[Matthew Munsey]

I like the idea of a point scale that measures the severity of winter storms. I agree with Jeff that visibilities and travel impacts overlap in some ways; but in general, where I differ, they are separate categories that have distinct effects. My only concern: how do you quantify the words "little to no impact"? If there is no impact, why even rate the specific storm? If there is little impact, why would it not have an occasional or intermittent impact?

 

[Jeff Miller]

Also, is your scale an attempt to rate a storm overall (i.e., apply just one number to all areas that the storm affected) or is it just for a point location? If so, then you'll have to account for the spatial variability in impacts such as visibility and road quality.

.

Response: When I was composing this scale, I was imagining a comparative rating between forecast and actual. Applying a number to all areas the storm has affected would be counterproductive as the southern areas of the storm may not have the same impacts as the northern areas. Therefore, I believe the only choice is to have it on a point - or regional - rating system which would then be calibrated to "actuals".

It seems that your classes of impacts overlap each other. For example, visibility and travel impact share a lot of common symptoms. Same goes for threat to life...it's kind of a combination of the three other classes.

I agree and thought of this - however, I believe there is a distinct difference between what I define travel impacts (road conditions, depth of snow on roads as well as quality of snow [i.e. packed vs powder] and visibilities due to blowing and drifting with falling snow. Therefore, I figured this was a wide enough differential to be two categories of rating. If these were blended together, I'm not sure the ratings would be as accurate.

Just a few thoughts. I don't think your system addresses variability in a local area very well. For example, lake effect snow in the Great Lakes. The threat of power outages in certain situations such as wet snow on full foliage can be so significant that it should not be buried under a generic "economic" category. Is "economic" defined? How about storm duration? I see this every year..a 5 minute snow squall has far more impact than a 2 day moderate snowfall. Another example..the impacts of a given storm are far greater when they are unseasonal..for instance..the first storm of the season when travelers aren't use to icy weather. Also, this is a snowstorm index..what about mixed precip? Most storms in many parts of the country run the full gambit of precip types in a given scenario. How is that addressed?

Response: Lake Effect snowfalls that create a threat of power outages due to wet snow on wet foliage on this system would rate under a high or exteme "economic impact" due to financial loss and property damage. Economic is defined as loss of finances or property due to storm impact. The only way to define this is on a local or regional level as the economy in Boston would be a much higher impact level then if the same storm went over rural western Nebraska, for instance.

Response: Mixed precipitation or ice is not accounted for in this system. Ice Storms could probably be addressed with their own rating system - it is intended for true snowstorm situations with this scenario as currently written.

Response: Unseasonal storms would have a higher economic impact then seasonal storms due to issues such as foliage and time of year - so it would bump up the "economic" level most likely a notch over normal, potentially increasing the overall rating of the storm.

My thoughts on this so far, again thanks for the early thoughts and keep them coming, debate the points and this can be refined, as Jeff Duda said, I also believe this is "on to something" and many minds are far better than one (being mine).

 

[Jeff Miller]

I like the idea of a point scale that measures the severity of winter storms. I agree with Jeff that visibilities and travel impacts overlap in some ways; but in general, where I differ, they are separate categories that have distinct effects. My only concern: how do you quantify the words "little to no impact"? If there is no impact, why even rate the specific storm? If there is little impact, why would it not have an occasional or intermittent impact?

No impact storms (example, heavy snowstorms with above-freezing roads and immediate melting - grassy only accumulations) still should be in my opinion rated. This kind of storm may cause low visibilities (moderate to high) but low to no impact on travel (due to immediate melting on roads). However, such a scenario is rare as wet-bulb cooling frequently lowers the temperature in such a situation to 32 to 34 degrees within hours anyway if snow persists and does not turn to rain. Yet such a snowfall can quickly end or change to rain if it is north of an approaching warm sector.

 

[beaudodson]

I like the idea that you focus on impact. Most people could care less about whether it snows 3" or 5" - they just want to know how it is going to impact them.

Remember that a light coating of freezing drizzle can cause more "impact" problems than 8" of snow. Our biggest event a couple of years ago was a short duration freezing drizzle event. Numerous people were killed and injured. There were thousands of wrecks in our area associated with the storm. All from just a trace amount of freezing drizzle.

Impact - impact - impact.