It goes without saying that New Yorkers are unhappy with the way their city — and their mayor, Michael R. Bloomberg — handled the winter storm that hit the city on Sunday and Monday. Given the number of unplowed streets and unheeded 911 calls, they have ample reason to be annoyed.
Still, skeptic that I am, I wonder if Mr. Bloomberg has some reasonable line of defense.
There are essentially two kinds of arguments that a lawyer, taking on the mayor’s side in the case of Bloomberg v. Blizzard, might make. First, he might argue that the city’s response to the blizzard was actually quite ordinary — even though it might not have seemed that way. Second, he might argue that the snowstorm was exceptional in some way.
The mayor’s first line of defense, however, seems to be relatively feeble. While we don’t have any particularly objective metrics to gauge a city’s effectiveness in responding to a snowstorm (something like the percentage of streets that were plowed in the first 36 hours might work, but nobody has bothered to keep statistics on that), I instead searched for contemporaneous reporting by The New York Times on the seven most recent comparable storms to have hit New York before this one, dating back to 1969.
The two most recent large storms to have hit before this one — which occurred in February 2010 and in February 2006, respectively — both happened to coincide with the Winter Olympics. In both cases, the term “snow” appeared at least as often in The Times in conjunction with the Games as it did with New York’s weather, which would suggest that city was getting on with its business. Indeed, The Times’s reporting was heavy on slice-of-life pieces — the Broadway shows must go on!; look at the cute kids sledding in Fort Greene Park! — and seemed to reflect a citizenry that perceived its city as having responded competently to the emergency.
The blizzard of 1996, meanwhile, was reported (although with some degree of sarcasm) as a triumph for Mayor Rudolph W. Giuliani, who wore a crew-neck sweater as he toured all five boroughs in his four-wheel-drive vehicle, giving interviews and mugging for the cameras every step of the way.
“Snow is to a mayor as foreign policy is to a president,” Mitchell L. Moss, the director of the Taub Urban Research Center at New York University, told The Times in 1996, suggesting that the storm had been a boon for Mr. Giuliani. “You have a monopoly on it, so you have to go out and handle it. So you leave City Hall and go to all five boroughs, and you can look terrific fighting the snow.”
The one clear exception was in February 1969, a comparatively modest storm — it left about 15 inches of snow on Central Park, compared with the 20 inches that the blizzard left earlier this week– that nevertheless hobbled the city for several days. Not just “tertiary streets,” but several major parkways were still closed to traffic 48 hours after the snow stopped, The Times reported. Fingers from all five boroughs were pointed at Mayor John Lindsay, whose public comments — somewhat anticipating Mr. Bloomberg’s — were acerbic and aloof; Mr. Lindsay blamed poor weather forecasts and budget cuts for the debacle.
Forty years later, that blizzard is still referred to as the Lindsay Storm, and although it did not end Mr. Lindsay’s political career — he was narrowly re-elected later that year, although with just 41 percent of the vote in a three-way race — it almost certainly diminished his standing.
So, although every winter storm elicits its share of complaints, in six of the last eight cases the city seems to have responded satisfactorily, judging by the news reports, the exceptions being the Lindsay Storm and this most recent one, which I’ll refer to as the Bloomberg Blizzard.
But perhaps, instead, there was something exceptional about these two storms? Here is Mr. Bloomberg’s other chance at a defense.
One can think of about six dimensions that might plausibly affect the city’s ability to respond to a winter storm:
1. How much snow fell.
2. At what time of day, and on which day of the week, the snowfall began.
3. How long the snowfall lasted, and how intense it was at its peak.
4. What type of snow fell. (Hold your laughter, please.)
5. To what extent the snowfall was accompanied by other types of severe weather — like high winds, ice or extremes of temperature — that might have hampered remediation efforts.
6. How much snow was forecast.
Let’s look at each of these in the context of the Bloomberg Blizzard.
Amount of snowfall. New York City — although it gets its fair share of snow — is not Buffalo, Fargo, N.D., or Vladivostok, Russia, and so any blizzard is in some ways exceptional. By this most basic measure, however, the Bloomberg Blizzard was quite comparable to other large snowstorms. Accumulation was about 20 inches in Central Park, which is identical to the average for the other seven storms in our study, which ranged from 15 inches (the Lindsay Storm) to almost 27 (the Blizzard of 2006).
Timing of snowfall. Central Park started to accumulate snowfall as of about 11 a.m. on Sunday, although the peak came somewhat later — between about 6 p.m. and midnight. There does not appear to be anything particularly unusual about this timing. Five of the last eight blizzards, oddly, also began on Sunday (and the Blizzard of 1978, which began just after midnight on Monday morning, nearly did as well). Mr. Giuliani’s Blizzard of 1996, in fact, began not just on the same day of the week but also at about the same time, at about 11 a.m. on a Sunday.
Duration and intensity of snowfall. Snow fell continuously for about 21 or 22 hours from Sunday to Monday. This is somewhat on the short side: the snowfall lasted for about 34 hours in the February 2010 snowstorm, and 36 hours in the February 1978 one, although with some periods of inactivity.
That the snowfall was shorter, however, is not necessarily a boon for cleanup efforts. In fact, the opposite might be true: if it takes less time for a given amount of snow to fall, that means more snow is falling at once.
Below is a chart comparing the accumulation of snow during the Bloomberg Blizzard to that of the previous snowstorms, using historical data from the Web site Weather Underground. I’ve had to take a couple of liberties here. First, because the historical records measure precipitation in water content rather than snowfall, I’ve converted the amounts to snow, assuming that the snow-to-water ratio was constant over the course of each storm. Second, because in a couple of cases, it was just above freezing when the snowstorm began, I’ve assumed that half the snow melted immediately so long as this was the case. Third, detail is not available from Central Park for the February 1983 blizzard, so I’ve used data from La Guardia Airport instead.
Those technicalities notwithstanding, the Bloomberg Blizzard — which is highlighted in red on the chart — does seem to have been associated with a more rapid accumulation of snowfall than most other storms. Although snowfall was slow for the first few hours, it began to accelerate rapidly as evening came, and 15 hours into the storm, about 18 inches had fallen, compared with an average of 12 inches for the seven prior storms.
So here is a possible piece of exculpatory evidence for Hizzoner — the snow came down very fast! — although it rests on the assumption that a relatively brief but intense snowfall is more cumbersome to deal with than a more prolonged, but slower-paced one. Perhaps that is so, although it doesn’t help Mr. Bloomberg’s case that the other storm the city had a great amount of trouble dealing with, the Lindsay Storm, was fairly slow-paced, having initially begun as a wintry, sleetlike mix as temperatures hovered around freezing.
Type of snowfall. There is a running joke in the British tabloid media that turns on the phrase “the wrong type of snow”, which refers to a statement made in 1991 by a British Rail spokesman named Terry Worrall who blamed a disruption in rail services on the snow having been unusually soft and fluffy.
There are in fact, however, several kinds of snow, which could conceivably affect the ease of cleanup. One governing factor is the density of the snow; you may have noticed, for instance, that the snow you were trying to shovel off your front stoop yesterday was fairly heavy — or that it formed nice, icy snowballs.
One way to measure this is with the snow-to-water ratio: essentially, how much water the snow would form if it were melted down. This ratio differs somewhat from storm to storm. Higher snow-to-water ratios indicate a lighter, fluffier snow, which might be easier on plows; lower ratios indicate a denser, icier and more compact mass.
I estimate the snow-to-water ratio for the Bloomberg Blizzard as having been 10.1, which means a heavier snow than in most past storms. It had averaged 12.0 for the seven other storms in our study, the largest figures being 14.5 in 2006 and 15.5 in 1978, both of which were quite fluffy. Still, the snow-to-water ratio for several other storms, including February 2010, January 1996 and February 1983, were similar to that of the Bloomberg Blizzard.
Other types of severe weather. The definition of a blizzard is slightly fuzzy, but usually requires very strong winds (above 35 miles per hour) along with very low visibilities for a sustained period, in addition to large snowfalls. This week’s snowstorm appears to have met this definition, as did the blizzards of 2006, 1996 and 1978.
All of our winter storms, however, except to some extent February 2010, were nevertheless associated with at least intermittent periods of very heavy winds. Others came with a variety of hazards, like fog, ice pellets and even thunder (in 2006, 1983 and 1978). The Bloomberg Blizzard, then, does not appear to have been exceptional in this respect.
Its temperatures, which ranged from about 23 degrees to 26 degrees while the snow was falling, were also about average. Temperatures were about 14 degrees when the 2003 and 1996 storms began, but above freezing initially for the 2006, 1969 and February 2010 storms.
The weather forecast. I haven’t looked up the forecast for all of the prior storms — but the available data would seem to provide more of an excuse to Mr. Lindsay than Mr. Bloomberg. In 1969, the weather service said it doubted that more than one inch of snow would accumulate in New York, according to an account by The Times; the precipitation was expected to manifest itself as rain or sleet instead. Instead, 15 inches worth of snowflakes gathered at Central Park.
This week’s storm came in a little heavier than expected — but there was not the same order-of-magnitude mistake that there was in 1969. By Christmas, the National Weather Service had issued a Blizzard Warning, saying it expected 11 to 16 inches to accumulate in New York. Given the uncertainties inherent in weather prediction — and winter storms are a particularly vexing problem for meteorologists — a 20-inch snowfall could hardly be considered out-of-bounds given that forecast.
Mr. Bloomberg’s defense, then, would appear to be quite thin. If you put this week’s blizzard in a lineup with other recent New York snowstorms, it would not particularly stand out. It was worse than most of the others in some ways — the snowfall accumulated quite rapidly, and the snow was fairly dense. But other storms, like in 2006, have been associated with somewhat higher accumulations, or with other types of severe weather, like thunder and ice, that this one did not have. The forecast had been a little to the low side, but not to the extent — like in 1969 — that anyone can really claim to have been taken by surprise.
Nor does there appear to be any sort of signature for which types of storms the city has the most trouble responding to: the Lindsay Storm, sleety and sloppy, bore little resemblance to this one.
So what exactly did happen, then? Why did the city’s response seem so inadequate, particularly when Mayor Bloomberg had seemed to handle several other storms — this is at least the fourth major winter storm under his tenure — relatively well?
One thing that seems quite likely to me is that how the city responds in the first several hours of a major snowstorm is tremendously important. If the largest thoroughfares aren’t cleared of snow almost as soon as it falls, then the plows almost certainly have no hope of reaching secondary and tertiary streets, and may even be more or less immobilized until the snowstorm ends.
In other words, you have to do enough plowing to ensure that you can continue plowing.
In the Lindsay Storm, the city clearly got off to a slow start, but it had a moderately good excuse: up until almost the very last minute, the weather service was insisting that there would be little accumulation of snow. But the snow did accumulate, and the city found itself hopelessly behind.
This week’s storm wasn’t as much of a surprise. But — and this is just a guess — perhaps the fact that the forecast worsened significantly on Christmas, the very last day that most people want to think about disaster planning, contributed to a sluggish response. Then, when the snowfall began falling late on Sunday morning, it was fairly light at first — perhaps providing for some false sense of security — but it became severe quite quickly, and the accumulation was fairly rapid even by the standards of a blizzard.
No, this is not an excuse: mistakes were made during the Bloomberg Blizzard, and they had serious consequences.
But the nature of a winter storm is such that small mistakes can compound into large ones rather easily; being slightly tardy in the first few hours might eventually translate into an extra day or two before the city digs itself out. And it may be much longer than that before Mr. Bloomberg entirely repairs his reputation.