Why are snow forecasts less accurate than rain forecasts?

Last week, as I’m sure most of you know, up here in the Northeast we had the Storm of the Century that wasn’t. It got me thinking – at least around here, snow forecasts are almost always inaccurate. Other forecasts – rain, sun, etc., are usually right on, at least the day before. Meterology has advanced to the point that, if the forecast calls for rain in five days, we’re gonna get rain. What is so hard about accurately forecasting snow?

Subsidiary questions:

  1. Why do meterologists almost always err on overestimating snow amounts rather than underestimating them? I can only recall one snowstorm in my 30-odd years where we ended up with significantly more snow than forecast. In constrast, I’d say 80% of snowstorms underproduce. My sneaking suspicion is that this is driven by the media – higher forecasts of snow accumulation is much more dramatic.

  2. What is so hard about forecasting the rain/snow line? (For you unitiated, the rain/snow line is the line on the map where the precipitation starts coming down as rain instead of snow, IOW where the temperature changes).

  3. Is this just a Northeast problem? Are snow forecasts more accurate in Denver?



Snow forecasts are very accurate here in Charleston, SC. :smiley:
However, a couple of weeks ago, they predicted snow flurries which we did not get, and in 1989, after Hugo, we had a 16-inch snowfall which was not predicted.

My understanding is that, in general, the NE got plenty of snow, over 2 ft in some places. Other areas did not get as much, primarily because it started as rain. It is obvious to me why rain is easier to predict. Hey, if it’s warm enough to preclude snow and precipitation is going to fall, it’s going to be rain.

In the storm in question, the path of the low pressure area was crucial to the rain-snow line. It deviated a little further north than predicted. That resulted in the warmer area coming further north than predicted. So, the rain-snow line depends on fine tuning the path of the Low.

I think snow estimates are always on the high sign for the same reason hurricane warnings are given even to those areas with a low percentage of getting one. It’s better to warn the public about the oncoming danger than to underestimate and the danger takes the public by surprise. Better safe than sorry.

This will only partially answer your question…

A rule of thumb is that one inch of rain is roughly the equivalent of 10 inches of snowfall.
Not many people really care or notice if a 1/4 inch of rain versus one inch of rain falls, but the difference is definitely noticeable when this amount of moisture falls and accumulates as snow.
Cheers Mike30

damn, finally a science question I have an idea about and some body beats me to it. At this rate, I’ll never get my pie completed.

anyhow. I believe that MikeG has it right - the issue generally isn’t where the rain/snow line is (that they can get pretty well). the problem is they can forecast to a general degree the amount of precipitation, except that if it comes down as snow, the resulting variance is quite high (like the comedy central newscast on the storm of the century “we’ll get some where between 2 and 35 inches of snow”).

here in mid MI, I’ve often seen them get it wrong by quite a bit both ways.

Another thing to take into account: when they predict snow, they’re functionally predicting two different things, rain and the temperature. While they can generally forecast pretty accurately for either one, forecasting for both obviously increases the likelihood that they’ll be wrong.

The OP asked specifically why is it so difficult to predict the snow-rain line and why do they overestimate the amount of snow. The prior posts are not responsive to those questions and are inaccurate.

One inch of snow does not equal 10" of rain. That is just an approximation and doesn’t hold true in all cases. About half a foot of “wet” snow, which is the texture of snow ehen the temperatures are around freezing (near the rain-snow line) equals about an inch of rain. Moreover, meteorologists are very accurate in predictiong the amount of rain that is going to fall.

This is easy.

First, the conditions for predicting snow are the same as that for rain, except the temperature has to be below zero as well. This is more often the case in Canada than in most States.

Second, the amount of snow is generally overpredicted since it has more impact on making travel plans, etc. People are generally happy if the amount is less; if it is more than predicted they tend to remember that more often and more bitterly. So it pays to do a worst-case scenario.

Third, media hype in the States being what it is, you tend to get terms like “The Storm of the Century” batted around. Weather is hard to predict, and terms like this before the fact are asking for trouble. I mean, Newfoundland is having the worst winter now that they’ve had in seventy years and you don’t hear them whining. Pussies.

Fourth, the wet snow packs quite differently from drier snow and this plays a role in how much water you get from snow.

Fifth, weathermen LIKE predicting large amounts of snow. You know what they’re like. You just know the whole lot of them spent their entire childhoods crushing ants and setting fire to paper towels in the bathroom. They love bad weather and it shows in their beady little eyes near the any of every newscast.

I’m with Mike30 and wring. To extend this further, the statement in the OP that

is completely untrue. This perception is solely due to the fact that people don’t pay too much attention to predictions of rain in five days. I know this because there have been times when I have followed these rain predictions with greater interest (e.g. when I planted new grass) and found them to be every bit as unreliable as predictions of snow.

What is true is that it is rare that a prediction for no rain will be untrue. But it is very common for predicted rain to fail to materialize.

That’s below zero Centigrade. (32 degrees F to us Yankees.)
And that has to be the temperature at cloud level. The temperatures at the surface can be well above freezing and it can still snow. A lot of it will melt quickly, or if warm enough, upon landing, but it can approach the ground without melting even 10 degrees above freezing. When I lived in Illinois, I’ve seen snow falling in the low 40s. On the other hand, it it’s above freezing in the clouds, it will not snow. Snow is caused by the precipitation if ice crystals from the clouds.

It is easy to predict fair weather when a high pressure system is present. Precipitation depends upon a low pressure system, and that does not necessarily mean it will rain (or snow, sleet, etc.). Other factors have to be present, such as enough moisture in the area. Also the path of the system may be such so it will not rain, and if the path is incorrectly predicted, the rain forecast will not materialize.