It happens all the time. 6 to 12 inches of snow is predicted. Blizzard warnings! Batten down the hatches, stay inside, close the schools WOO HOO SNOW DAY! *wake up the next morning* a dusting. Its enough to drive a person crazy and lose trust in every weatherman alive. But why is predicting snow so hard? Well there are dozens of reasons but I will highlight one of them in relation to a weather scenario we find ourselves in coming tomorrow.
Everyone knows the temperature has to be 32F (0C) or lower to get snow. Well use that for the example below. Temperatures in the atmosphere are never the same. Typically the higher you go the colder it gets, but since weather is complicated and likes to make our lives difficult that is not always the case. You can have warmer air the higher you go (known as an inversion.) This ALWAYS complicates the forecast no matter what season, but winter they are especially a pain because if can effect the TYPE of precipitation that occurs.
The simple way: Everytyhing from the ground up is below freezing, you get snow. Woo hoo easy.
The hard way: Not everything is below freezing, will we get rain, snow, sleet, ice, or graupel. ARRRGH.
To Illustrate this I am showing 2 forecast soundings for an event forecast to hit Chicago tomorrow (January 12th 2013) If you have been monitoring the forecast you may have noticed it changed a bunch of times and many meteorologists expressing uncertainties with the forecast. Not because they don’t know what they are talking about, but because THEY DO KNOW what they are talking about. Be prepared to get educated!
Below is a sounding, on the left, the NAM, on the right, the GFS. Both are taken from the Chicago area for the same time (3z, or 9pm in the winter.) Click for full.
Without getting too technical, you have your temperature at the bottom, and your height on the left starting with the ground at the bottom (1000) and working your way up. The red line is your temperature and the green line is your dewpoint. You may have noticed I colored and labeled some other things. Congratulations, you haven’t clicked off this little article yet and I applaud you for making it this far.
With the simple way of things in mind, check out the sounding on the right, from the GFS. All below the line (except the last few feet near the ground.) Guess what, SNOW! Yay easy life can go on. Wait a second, whats the NAM one on the left doing? Hmmmm, temperatures get above freezing for a while there. This is what complicates things. We all know what happens to snow once the temperature goes above 32, thats right folks, it melts. This is where it gets tricky. If it fully melts you will get rain, if the rain falls to the surface and its 29 degrees you get ice. If the rain falls to the surface and its 36 degrees, it’ll stay as rain. If the temperature gets below freezing again before the ground, the rain will start to refreeze again and you can get sleet, graupel or “ice pellets” (and yes there is a difference between the 3.)
Figuring out which scenario is going to unfold is the hard part. Furthermore, you have to take into consideration as to HOW warm and for HOW long the above freezing layer is. If its only 33 degrees for 1000 feet, you won’t get much melting. If its 36 degrees for 1000 feet, you will have much more melting. You can have 33 degrees for 5000 feet, or 40 degrees for 500 feet, what does what and how soon will the now melted or partially melted precip fall back into a temperature below freezing and again, for how long? If you got confused just reading that, imagine how a forecaster feels.
The models also show different things when it comes to whether the air is saturated or not. To again make it simple, if the red and green lines touch, the air is saturated and you have precipitation and/or clouds. If they are apart, the air is dryer and you could either have shallower, less intense precip or a layer of dry air to evaporate precip before it even hits the ground.
So which model is right? As if there wasn’t enough confusion already, no weather model is ever 100% correct. Some are better than others, but each of them are wrong at times. It is up to the forecaster to use his or her own understanding of how weather works, plus experience to try and figure that out. This is why you end up with different forecasts from different people. Each time a new model run comes out the scenario could change. Suddenly the sounding on the left disappears, and confidence in all snow can increase. Then the fun part is figuring out how much, but we’ll save that for another time.