Weather

Dave Epstein reflects on the significant snowstorm forecast that never materialized for many

As forecasting accuracy has increased, so too has the pressure for meteorologists to be right all the time.

By David Epstein

updated on February 14, 2024

• Facebook
• Twitter
• Email
• Email
• 36

Perhaps you worked from home Tuesday because your kids didn’t have school. Maybe your doctor’s appointment got canceled, or the trash, which normally would be picked up, is still sitting at the end of the driveway. All these disruptions in your schedule because of a forecast of a significant snowstorm that never materialize for much of the Boston area.

Why was everything canceled when ultimately a coating of snow to the north to a few inches around Greater Boston is all that would fall? You were expecting to use the snowblower, but the driveway is just wet. Why did Sunday’s forecast of heavy snow change so drastically by Monday afternoon?

You might be asking some or more of these questions about our latest brush with winter weather.

Let’s look at two scientific camps to evaluate this storm and the fallout from it.

The first would be the meteorology of it; the second would be the psychology.

Let’s delve into meteorology first. Forecasting has come a long way in the past 15 years. I often see criticism around how bad a particular forecast is and how someone like Don Kent, the Dean of New England Weather Forecasting, did it before the models?

The reality is that forecasting back in the 1970s and ‘80s, prior to things like the European and American weather forecasting models, was statistically much worse than it is today. As forecasting accuracy has increased, so too has the pressure for meteorologists to be right all the time.

And although it’s pretty unusual for the models we use to be so different just a day before a storm is about to start, the reality is that the weather is still ever-changing.

I spoke with my colleague, Chief Meteorologist Eric Fisher over at WBZ, about Tuesday’s storm, and although he understands the agitation people feel when a forecast doesn’t pan out, he says there’s always room for “uncertainty,” even with the most sophisticated technology.

“I’m both pleased and concerned with the level of confidence people show in weather forecasting,” Fisher said Tuesday. “While most of the time things go according to plan, it’s worthwhile to remember nature is in charge and no one forecasting anything about the future is right all the time. We don’t have the ‘Grays Sports Almanac’ for snowstorms.

“There is and always will be uncertainty, and I think we have gone a bit ahead of our skis canceling things so far in advance in the past few years. It sometimes leads to very frustrated workers, parents, and forecasters. Many of us are feeling that today, and it was avoidable if we all just waited a few more hours on Monday to make those calls.”

Monday morning I mentioned that if the European model ended up being correct we were looking at a light to moderate snowfall not the heavy one. I knew my day was going to be filled with evaluating the data and seeing what the trends were going to evolve.

Then by mid-afternoon, my fellow forecasters and I had lowered expected snowfall totals dramatically.

Monday evening’s prediction changed to one forecasting the heaviest snow to fall across Connecticut, southern Rhode Island, and southeastern Massachusetts. And indeed by late morning Tuesday, parts of those areas had received 6 inches to a foot of snow, most of it falling in just a few hours. The heavy snow bands I was concerned with over the weekend did materialize; they just were about a hundred miles farther south.

Jeremy Reiner, chief meteorologist for 7News (WHDH-TV) in Boston, said it’s been quite awhile since he’s experienced such an abrupt change in a winter storm forecast.

“Tricky storm to say the least…for me, alarming how most/all the modeling was in agreement as late as Sunday afternoon that 6-12″ was highly likely (ensembles over 70% confident of those amounts occurring)….but to see such a dramatic downward trend in amounts with less than 24 hours before the storm’s arrival is something I haven’t seen in a decade.”

You might ask why we all didn’t just go with the European model sooner. Isn’t it the most accurate?

It is a more accurate model than the others, however it is not perfect and to discount other models that still have a high level of accuracy would have been foolish.

Organizational and individual psychology is another major player in all of this. Media outlets are businesses, and weather gets attention. Most forecasters I know would never purposely hype a forecast. As a matter of fact, I am particularly cautious when it comes to storms because I understand how powerful storm coverage can be. But it is the rare storm that leaves a lasting impact — think Blizzard of ‘78.

When you have everyone talking about a storm and then you add in all the social media coverage, much of it from untrained sources, you have a recipe for a lot of misunderstanding of what is actually going to happen.

Some even took to social media to express their “views”:

Even in well-forecast situations, I often hear people parroting bad information in the supermarket or at the gym. The people making decisions to close schools, not pick up our trash, and shut down businesses are making these decisions with the best information available at that time. They also have their own biases and their own fears. Someone who runs a particular business doesn’t want to be responsible for keeping it open if the roads are really bad and one of their employees ends up getting in an accident trying to get to work.

It’s the same thing with superintendents. School gets canceled because people worry about kids waiting for buses and parents driving to school in snow. Any of us, including myself, can lament about the good old days when school was not canceled for just a few inches of snow. But those days are gone. That’s not what we do anymore. Risk of lawsuits to an organization add pressure to just close a business rather than take a chance being open.

When a small pocket of districts adjacent to one another close school it begins a domino effect as more and more surrounding districts follow suit. For example, if Medfield cancels school it puts pressure on Westwood because Westwood might have teachers who live in Medfield who need to stay home to take care of their own children. The more districts that close, the more it becomes a “snowball” effect. Before you know it, it’s 6 p.m. and the lead story is all the schools that have already closed for the following day.

I definitely didn’t like having to change the forecast Monday for a storm that was going to begin in less than 24 hours. But perhaps cities and towns also need to hold back a little longer before making some of these decisions that can impact a lot of people.

Growing up we would listen to the radio or watch TV to see the school closings scrolling across our screens in the morning, wondering if school would be canceled. Society has changed a lot since those days, but here in a cold climate where at least some snow and cold is a part of every winter, how low is our collective tolerance to venture outside in bad weather and have we tipped the cautionary scale too far?