Forecasting snowfall amounts along the I-95 Corridor from Washington DC to Boston always presents a challenge. It is of course the most populated stretch of highway in the United States, so there is an extra reason to get it right. When so many millions of people are going to be hanging on your every word, accuracy is highly desired. By nature of its location, each winter storm, or Nor'easter, presents a set of variables that could have a large range of results. In order to produce significant snowfall amounts, a few factors must be in place; cold air, abundant moisture supply, and a prolonged precipitation event. Many winters, including the 2007-2008 winter have seen these ingredients fail to come together at the right time. Sometimes there is an extended arctic outbreak, but it is simply a cold dry air-mass and no storms are present. The other snow-less scenario takes place when there is cold air in place, and a storm develops, but the Low Pressure tracks too far to the west, bringing in copious moisture, but also abundant warm air advection because the ocean is typically much warmer than land during the winter. Low pressure that tracks too far to the west, it brings an onshore flow, setting the stage for a cold rain from Philadelphia to Boston. Another snow-buster that can disrupt a snowfall forecast even if all the ingredients seem to be coming together is the dry slot. The dry slot is defined as the cloudless region between the head and tail is produced by dry air descending from aloft that causes evaporation. Depending on the size and location of the dry slot, a snowfall forecast in the Northeast can be drastically affected.
Every winter, meteorologists are given the chance to forecast snowfall around the big cities of the Northeast. Mid-latitude cyclones, also known as "Nor'easters", are more common in the winter months than any other time of the year. Snowfall can happen from fall to spring in the Northeast, but typically most snow events occur between December and February. Each mid-latitude cyclone presents this problem, because the dry slot is a feature of all Mid-Latitude cyclones. Sometimes the dry slot can quickly "fill in" if the low intensifies and moisture wraps all the way around the center of circulation, but this is not always the case.
The consequences of a dry slot invasion are broad and can truly be a forecast buster. The problem with forecasting snowfall is that the public wants to know exactly how much snow they will have to shovel off their cars, or if they should stay home from work because of an 8" snowfall during the day. Meteorologists will use various model data, including QPF data to come up with a snowfall amount. The potential problem exists when all the models point to over a foot of snow for a given area, and it makes sense to agree with this forecast and put it out to the public. If a dry slot happens to form over the mid-Atlantic coastline during the intensification of the low, which is often the case, than instead of over a foot of snow, the most populated regions might only see 6-8". This forecast would be a bust, and the public would be angry. The problem is that dry air can affect a small area, so that over Long Island where the dry slot might not have an influence, the storm totals would pan out and the 12"+ forecast would be spot on, but in Manhattan and Brooklyn, where the dry slot appeared during some of the heaviest precipitation, the residents would think that the storm never happened.
A Nor'easter is a fun and challenging storm to watch develop for several reasons. It usually involves two low pressure systems, one that moves northeastward from the gulf region, and a secondary low that forms somewhere off the coast of the mid-Atlantic. This transference of energy makes is a very dynamic event. Besides the energy aspect of the storm, the rapid intensification is often very impressive. When pressure levels drop more than 24 milibars per 24 hours, we say the storm is "bombing" out. This can mean huge snowfall totals for lots of folks. What makes the storm so fun to watch is also what makes the dry slot effect more pronounced. Although these winter storms can seem like a long drawn out event, sometimes the actual snow event is rather brief, when considering the amount of snow forecasted. There is usually a window of a few hours when the heaviest snow will occur over any given location. If a dry slot happens to form over the forecast area during this window, the effect can be detrimental to a storm. Snowfall rates of 2-3" per hour are common with these storms, and sometimes it only takes a good 4-6 hours of snow to leave a huge impact, but if dry air moves in to the storm for even only 2 hours of the heaviest snow, this can cut final accumulations in half.
It is never easy to forecast snowfall totals for these winter storms along the I-95 corridor. One has to be constantly checking radar and satellite to monitor the formation of the low and the structure of the storm. The radar will aide the meteorologist in determining where the heaviest snowfall is occurring and the satellite imagery will point out where any dry air is invading the storm. When dealing with the dry slot one must constantly monitor the storm in order to produce an accurate snowfall forecast. A lot of "Nowcasting" is involved. If I were to forecast snowfall totals for an intensifying Nor'easter and I noticed a dry slot showing up around the DC area, I would have to take this into account. One must constantly check satellite and radar to see if the dry slot is eroding the precip away, or if it is filling in. The trickiest thing about this is that the dry slot behavior could rapidly change. The forecaster must adapt his/her forecast as the storm develops, so if a dry slot seems to be expanding, one might cut forecast totals in half. The problem with this is that the dry slot could fill in and then the totals might rebound towards the higher original amounts. It is definitely a "wait and see" situation where intuition and experience would be a huge benefit.
I sat down with Meteorologist Jim Nichols to discuss his experience forecasting with winter storms and dry slots. When I asked him when he has ever had to deal with the issue of a dry slot affecting snow totals his response was "it is a problem with every winter storm in New Jersey". He recalled a winter storm during his college days where fistfights broke out regarding snowfall forecasts. Jim has been forecasting in the Philadelphia area for several years, and he has experience with dry air completely busting a forecast. The Nor'easter of March 2001 was originally forecast to be a monster snowmaker from North of DC all the way to Boston, but dry air worked its way into the storm and the snow never materialized. In the end instead of picking up over a foot of snow as predicted, residents were left with just over an inch. Furious, viewers demanded blood and blood was let, jobs were lost and this blunder will be remembered for a long time in Philadelphia.
When forecasting around the dry slot, there are a couple of tools a meteorologist can use. The fact that the public is used to receiving ranges for snowfall totals is beneficial. In some cases a range of 3-6" will be sufficient to account for the dry air limiting snow totals. The problem with these ranges is the public often focuses on the upper limit, and ignores the lower limit. So even if a storm is forecast to dump 4-8" and most people see 4", there is a tendency for the public say, "I thought we were supposed to get 8"!"
To better forecast in this situation, understanding and using satellite data are essential. It is not enough to make your snowfall forecast and let the storm unfold. One must constantly check the available imagery to see when and where dry slot formation occurs, and how the storm is developing as a whole. This type of forecasting might require a sleepless night, but should improve the accuracy of snow totals. Also, there is something to be said for experience and intuition in this case. There is no substitute for having been through a situation before, and years of experience forecasting these storms can only help one recognize the patterns of dry slots.
1. Interview with Jim Nichols, Meteorologist NBC Weatherplus. April 2008.
2. Lutgens, Frederick K, and Tarbuck, Edward W. The Atmosphere Pearson Prentice Hall, 2004.