I spoke with Colin Zarzycki, an associate professor of Meteorology and Climate Dynamics at Pennsylvania State University, on Monday morning about what we can expect from winter storms in a warming climate. Our conversation has been lightly edited for clarity, and the snow-weary should proceed with caution.
I've read both that blizzards will increase in a warming world because the atmosphere can hold more moisture to make more snow, and also that, because it’s warmer, a lot of the precipitation will fall as rain instead of snow, so the storms will decrease. What does the research actually say?
Let’s back up for one second. Blizzards like we have in the Northeast today are a subset of nor’easters. We also call them mid-latitude or extra-tropical cyclones — you hear people talk about “low pressure,” “bomb cyclones.” At the end of the day, these are synonyms for storms that track up the East Coast of the U.S. and dump a lot of snow, particularly along the major metro corridor.
A blizzard is a special subset, where you have strong winds that blow the snow around. And that’s really problematic, because — have you experienced a lot of snowstorms?
I went to college in Vermont and lived in New York City for 10 years, so I’m familiar with snow.
I ask because, every once in a while, you talk to someone from, like, Miami, and they’re like, “I don’t know what you’re talking about.” But during these strong wind events, blowing snow reduces the visibility. That’s very bad for transportation like aviation, but also just driving on highways and roads.
I want to be careful, because there’s been less work done on the wind side of things. The broad consensus is that if you measure nor'easters as a function of their low pressure — which is somewhat analogous to wind speed; they’re not exactly related, but they’re pretty close — there actually doesn’t seem to be a huge shift. For every storm that comes up the East Coast and turns into a bomb that’s blowing 80-mile-an-hour winds, the distribution of the wind looks pretty similar across different climates, whether cooler or warmer.
What you’re referring to about the precipitation: — this is the thing we’re most confident in the science [of]. If you make the very simple argument — which admittedly, our models indicate it is not a bad argument — that if the number of nor’easters that move up the coast stays relatively constant and the intensity of them doesn’t change a lot as measured by wind speed, but if the atmosphere is warmer and can hold more water vapor, then the rates of what’s coming out of the sky essentially increase.
Now if you’re thinking, “Okay, well, that’s snow,” then yes. If you could take this storm and put it in a time machine and move it 50 years from now, and if the atmosphere is 2 degrees [Celsius] warmer, then you’re going to have more precipitation coming out of the sky, all other things being equal.
But you mentioned the other tricky thing that complicates life. When climate scientists think about precipitation in, let’s say, Florida, where it doesn’t snow at all, it generally all just goes one way: It gets warmer, it rains harder. But in the Northeast, we have two things that compete with each other. On the one hand, precipitation increases, as we just discussed. But then obviously, if it warms, more of these storms are likely to produce rain rather than snow.
If you look at just the average number of snowstorms in a warmer world, whether you’re comparing today relative to 1850, or if you’re looking at today and trying to figure out what’s going to happen in 2100, in general, the warmer it gets, the less total snow and the less total number of snowstorms because more of them become rainstorms. The tricky thing is, the decrease really only happens with the weaker snowstorms, the nuisance types.
So if we still get periods in warmer climates where it’s cold enough to snow, and now we’ve turbocharged the atmosphere’s ability to hold moisture by warming, then what we’ve actually done is make it so that when it does snow, it snows harder. In general, we expect to see fewer overall snowstorms when it’s warming, which is very consistent with what we’ve seen in observations in the Northeast U.S. If you look at any major metro area and you plot snow since 1950 it’s generally been on a downslope. But these big blizzard-type storms aren’t going away.
The jury is out as to whether the most, most, most, most extreme snowstorms become a little more extreme. But the big take-home message is that the frequency of big nor’easters isn’t going away, even if the climate warms.
There has been a lot of talk about this being the first blizzard to hit New York City in nine years. I don’t think I can remember a storm quite like this from when I was living there. Is that because this is the most extreme version you’re referring to, that we haven’t seen as often?
If you were to ask someone who has lived in New York City since the 1950s, they would probably tell you that this is a bad snowstorm, but that they’ve seen similar ones. I’m not an expert on the history of New York City weather, but there were a couple of big storms, I think, in the 1970s that were analogous to this, if not a little worse.
What is unique about this storm is that we really haven’t seen one of these tight coastal blizzards this year. We had that storm that came through earlier this year, which also brought a decent amount of snow to New York, but it tracked across the country rather than forming right off the coast and moving up that direction. This one is dragging snow across New York City and Boston; it’s a very classic Northeastern U.S. blizzard.
I think the main aspect is that we have been in a period of luck. We haven’t had these storms as frequently in the past. Some of it goes to that kind of dice-rolling thing with the temperatures. But if you look over the last 10 years, I would assume it’s not that New York City has been nice and sunny and calm in the winter. It’s that you’ve had these wintertime cyclones, but it’s been a lot more rain, or wet, rainy, sleety snow. It hasn’t been cold enough air to really lock in the blizzard conditions.
My understanding is that blizzards are specific atmospheric events in which the wind speed must exceed 35 miles per hour and visibility is limited. How difficult is that to capture in the data? I know from my reporting on tornadoes that it can be really difficult to capture wind events. How do you study this?
The fancy word in climate science is “compound extremes,” and a blizzard is a form of a compound extreme where you have multiple hazards at the same time. Add one layer on top of another, and the more there are, the harder it is to get information out of the data.
Especially in densely populated areas like the Northeastern U.S., blizzards are fairly tricky to look at. When you read the National Weather Service’s definition of a blizzard, it’s like, “It has to be snowing, and you have to have sustained winds, and you have to have decreased visibility.” All of those mean you’re adding layers of complexity to the data.
Tornadoes are a little similar; they’re a discrete phenomenon, and you need specific ingredients to all line up, and there’s also an observation problem. It’s somewhat analogous to blizzards: I could be at JFK Airport in New York, which is right on the ocean. There’s not a lot in the way to slow down the winds. Especially if you have drier snow, it’s very easy for it all to blow around. If I’m a guy working at JFK, I’m saying, “This is really bad, it’s really windy, the snow is coming down, and we can’t see anything. We have to shut everything down.” But put yourself in Midtown or somewhere where you’re surrounded by buildings and a little further away from the ocean, then suddenly the winds might be reduced because you have more obstacles that can slow it down. You’re experiencing the exact same storm, but the impacts are very different.
You said at the beginning that the underlying assumption is that nor’easters will continue at the same rate they’re happening now. Is there anything I should know about the way climate change is impacting those events?
Precipitation is the main thing. There’s been some work on the frequency and track of the storms, and we’ve seen small changes. But we also have a sample-size problem. The more you want to focus on the intense storms, the less you have in your records, and the more challenging it is to tease out what’s going on. That’s one of the reasons I really like models.
So maybe, if you squint, you can see some small changes in the frequency or the track, but it’s on the order of 5% to 10% per year. But the number of nor’easters we actually get in a given winter is not small; depending on how you want to classify it, it’s something like 10 to 15 any given winter. They don’t all produce a lot of snow; some of them go offshore, and if you’re sailing a boat in the middle of the ocean, then you’d be like, yeah, this is a big problem. But generally, we have very high confidence in understanding the precipitation, and decent confidence in understanding how the rain-snow partitioning changes. The winds, I think, are kind of an open question. But we’re talking secondary effects relative to the precipitation for all of them.
Is there anything else I should know about blizzards and climate change?
I do interviews every winter about bomb cyclones and big storms. The fact that I do multiple interviews a winter implies that the storms themselves are not anomalous. If you actually count them, you end up with a decent number. You just need the dice to come up snake eyes — all the ingredients need to line up for it to be something impactful. And that’s what’s happening now.
What climate change does is change the underlying probabilities and distributions. But at the end of the day, the main thing that actually drives what’s going on with these storms is, can the atmosphere put the Lego pieces together for these impacts? Every cyclone that we get during the winter, if you go back and look at the historical record, there’s plenty of evidence for these types of storms.
Michael M. Santiago/Getty Images
A chart shows the expected new generating additions in the U.S. grid's pipeline. EIA