Tag Archives: Snowpocalypse

Forecasting uncertainty in practice: Snowperbole

Example of snow forecast communicating levels of undertainty; image from the Capital Weather Gang

Example of snow forecast communicating levels of uncertainty; image from the Capital Weather Gang

Because making accurate predictions is extremely difficult, we can dramatically improve both the accuracy of forecasts and enable effective communication about the forecast by embracing the uncertainty involved in the forecast. This allows decision-makers to both use the information available while understanding the limits of those predictions.

Following forecasts for a “potentially historic” storm set to hit New York and New England, public officials in New York City went to great lengths to emphasize the dangers of the storm. The Governor closed down New York’s subways in anticipation of the storm (showing one of the quirks of New York’s transit governance, local transit is under state control).

There was just one problem: the storm mostly missed NYC.

In their forecast post-mortem, the Washington Post’s Capital Weather Gang highlighted the key shortcomings of the forecast – a failure to present the level of uncertainty in the forecast.

Why were the forecasts so bad?

It’s simple: Many forecasters failed to adequately communicate the uncertainty in what was an extremely complicated forecast. Instead of presenting the forecast as a range of possibilities, many outlets simply presented the worst-case scenario.

Especially for New York City, some computer model forecasts were extremely dire, predicting upwards of 30 inches of snow – shattering all-time snowfall records. The models producing these forecasts (the NAM model and European model) had a sufficiently good enough track record to take them seriously.

However, some model forecasts (e.g. the GFS model) signaled reason for caution. They predicted closer to a foot of snow.

Part of the challenge here is that most of the forecast was accurate. This was a historic storm; the storm simply tracked a bit further to the east. Areas like New York City were right on the margins, where a small change to the inputs can mean a large change in the outcome – and the forecast did not adequately convey that uncertainty. Add in the fact that the forecast miss happened to be the largest city in the United States, and you have a very public error.

When a forecast is so sensitive to small changes (eastern Long Island, not far away, received 30-plus inches), it is imperative to loudly convey the reality that small changes could have profound effects on what actually happens.

It’s easy to second-guess public officials making key decisions like closing transit systems after the fact (and after the forecast bust), but they can only act on the information that they have in front of them. It’s easy to argue that it is better to be safe than sorry (and this is certainly true – it is better safe than sorry) but there is a real risk of eroding public confidence in these kinds of decisions when the forecast doesn’t pan out. (It doesn’t help that despite closing the subways, the MTA’s snow plan called for trains to remain in operation without passengers to keep the tracks clear of snow)

As some meteorologists suggest, conveying the uncertainty in their forecasts should be a larger element of both the forecast and communication. It’s not just a matter of using the best information available, but also understanding the uncertainty involved.

Snowpocalypse III – Linkage

Image from Wayan Vota on Flickr

Some more pleas for realistic expectations: Ryan Avent chimes in on the economics of it all, and Jon Chait notes the basic, physical problem with dealing with so much snow in such a short period of time:

In my neighborhood, like much of Washington, people park along the street. When it snows, plows go down and shove the snow away from the middle of the street and toward the sides. When it snows large amounts, the plows create massive snow barriers between the cars and the street. Digging out one’s car becomes a huge task. You have to scoop all the snow off the car itself, around the perimeter of the car, and this is just a tiny warm-up to the major task, when you have to breach the snow wall so that your car can get out to the street. This is even harder than it sounds. Every shovelful has to be carried back form the middle of the street and deposited on the front lawn.

Before the latest snowfall, the barriers in my street stood at around three or four feet. When the plow comes, they’re just going to get bigger. The nearly-intractable problem here is that there’s simply no place to put the snow. All the spare space along the side of the street is taken up by parked cars. The snow has nowhere to go.

One part of the solution is to truck the snow away to a remote location. Washington is already beginning to do this. With enough money to hire enough trucks and equipment, the government could probably remove all the snow. But this is a massive project that would take an unthinkable commitment to finish. I’m wondering if I’ll see my office again until spring, or spring-like weather.

This is the crux of the issue. Where is it going to go? Who’s going to move it? Any solution requires participation from the city’s residents.

Gabe Klein agrees – where is it going to go? Dr. Gridlock’s blog has some good notes on the challenges:

“This is no longer just a plow operation,” said Gabe Klein, director of the District Department of Transportation. “There is too much snow accumulation on some streets for the plows to adequately move the snow, the snow has to be physically removed and hauled away. This will add some time to our cleanup efforts but we have crews working around-the-clock to minimize how long and to assist us in being as efficient as possible.”

In addition to 250 to 270 pieces of equipment for plowing and treating roadways, the city has deployed specialized equipment such as backhoes, frontloaders, dump trucks, and dumpsters.

Dumpsters, eh? Sounds unorthodox. Speaking of unorthodox…

Snow removal tools, realistic and not: The City Paper has some great suggestions for makeshift snow removal tools – but missed one of the obvious ones I’ve seen out there – the dustpan. In previous snow storms, I’ve seen cutting boards and spatulas making their way from the kitchen to the front yard. Still, that’s not as sweet as the flamethrower option (appropriately tagged under ‘cool shit‘).

Don’t bring the grill inside: Just don’t do it. Seriously.

And then there’s this:

Snow perspective, in graph form.

Since we’ve now eclipsed the seasonal record, it’s worth noting how unusual it is for DC to get lots of snow in a season, to say nothing about snow storms coming back to back. Let’s look at the history:

We’re now well above that blue star for the 1898-99 season record. As Gabe Klein noted on Kojo yesterday, you can’t budget for those intervening years and then expect to deal with the extraordinary snowfalls.

Likewise, there are physical limits to how fast you can remove snow, and when the snow is accumulating faster than that rate, you’ve got a problem.

City Paper has a survey of ANC commissioners, asking about their street conditions. Gary Thompson has some good perspective on snow removal:

Gary Thompson, of ANC 3G, who lives in the 2800 block of Northampton Street NW: “My own street’s been plowed very well,” he says. “They’ve done an excellent job plowing, especially under the circumstances. There have been a few forgotten streets here or there, like a dead end culdesac,” but the city is catching up. “People are very quick to complain,” he says. “But Mother Nature is what it is. It’s a pretty powerful storm.”

Ben Thomas, not so much:

Ben Thomas, of ANC 7E, who lives in the 1100 block of Chaplin Street SE: A plow came by sometime last night, he reports (the second of two times a plow has made an appearance during D.C.’s three snowstorms this season). “They don’t really do anything but pile the snow against peoples’ cars,” he says. But the street is at least passable. “I saw a car go by a little while ago,” he says.

That’s what plowing does – it moves snow around. And that works pretty well when you get a little snow here and a little snow there. When you get 40 inches in a matter of a few days, there’s only so much you can do within the realm of what’s physically possible.

Perspective, please.

Snowpocalypse III – the removal

Photo from InspirationDC on Flickr.

Well, it’s official. The winter of 2009-10 is now the snowiest winter on record in DC, eclipsing the snow season of 1898-99 That’s saying something, since DC’s current official weather station is at National Airport, which has abnormal weather conditions compared to the rest of the District, thanks to being surrounded by water on all sides and at low elevation. That 1898-99 record wasn’t taken at DCA of course, since the airport then was nothing but Potomac River mud flats. Hell, heavier than air human aviation wasn’t even around yet.

So, we’ve got a lot of snow. Now we have to deal with it.

While grades for this current snowfall are obviously yet to be determined, DC did an excellent job for the first round back in December. The response to round two this past weekend was also, all things considered, quite effective.

I bring this up because I hear a lot of complaints about snow removal, and almost all of them strike me as a product of unrealistic expectations and a lack of experience dealing with snow.

Snow doesn’t magically disappear. Having grown up and lived in several cities in the Midwest frostbelt, I’ve dealt with plenty of snow. It seems to me that many DC residents equate snow plowing with making the snow disappear – this simply isn’t the case. Plowing moves snow around. In the Midwest, you get used to this – snow sticks around, and you learn to deal with it. Snow Emergency routes will get plowed down to the pavement, but most residential streets will have snow on them until winter ends – this is considered ‘plowed’ because those streets are more than passable.

Still, snow takes up a lot of space – even Minneapolis has run out of places to put it this winter, forcing them to implement permanent one-side parking restrictions as the ever-growing snowbanks are encroaching on street space.

Shoveling is your responsibility. Plenty of DC bloggers have noted this (as is the case in just about every snow city in the US), but the responsibility for sidewalk snow removal falls on residents/tenants/occupants/owners.

“It shall be the duty of every person, partnership, corporation, joint-stock company, or syndicate in charge or control of any building or lot of land within the fire limits of the District of Columbia, fronting or abutting on a paved sidewalk, whether as owner, tenant, occupant, lessee, or otherwise, within the first 8 hours of daylight after the ceasing to fall of any snow or sleet, to remove and clear away, or cause to be removed and cleared away, such snow or sleet from so much of said sidewalk as is in front of or abuts on said building or lot of land.” (D.C. Code § 9-601)

There are good reasons for this – the city is struggling to deal with the current snowfall – adding miles and miles of sidewalks to their duty list would make the task impossible. There’s absolutely no way such removal could be handled in a timely fashion. It’s your civic duty, it’s good public policy, and (when people are engaged with the snow culture), it works far better than any other option.

Perspective matters. Other cities might be better at dealing with snow, and they might have a stronger snow culture to deal with shoveling and whatnot – but let’s not forget that this particular snowfall is exceptional. These kinds of storms, dumping several feet of snow on an area, will cripple even the best-prepared cities. As noted above, this is the most snow DC’s ever had on record. Sure, it’s a ballpark total similar to the average snowfall in Minneapolis, but the snow in Minnesota comes in small increments that are much easier to deal with than the huge storms we’ve seen.

All of that record snowfall has basically come in three storms. Even the best prepared snow cities will be slowed down significantly by storms of that magnitude, especially when they hit back-to-back.

Snowpocalypse III – time to assess blame

Blame this guy:

Snowpocalypse II – the aftermath

Some more pics from walking around Capitol Hill today:

Different shoveling philosophies

Those who’ve lived in snowy places know this as a matter of fact – it’s a lot easier to shovel 5 inches of snow 5 different times throughout a storm than it is to try to shovel 25 inches all at once. End result? A narrow little path. Better than not shoveling at all, however.

Eastern Market Metro canopy

The snowplow brings mixed blessings for those trying to dig out

Lincoln Park

Makeshift sledding in the alley next to Mott's Market

Salt and infrastructure beneath the city

Ever wonder where all that road salt comes from? A question that’s quite topical today. Mammoth has a post up on an operating salt mine beneath the city of Detroit.

Detroit Salt Mine

John Nystuen has a discussion of the legal implications, acquiring mineral rights for salt 1,000 feet below the surface of an active city. His map of the area shows the approximate extent of the mine in Southwest Detroit.

Approximate extent of the Detroit Salt Mine. Image from John Nystuen

Nystuen notes that the shape of the mine lends itself to the economies of scale in negotiating mineral rights contracts with the larger, industrial landowners. The main east-west axis that connects these areas lies beneath a rail yard. Much of this area of Detroit is extremely industrial. The middle branch of the mine above extends right up to the edge of Ford’s massive Rouge complex. This above-ground landscape has some fascinating visuals, particularly as it ages but remains in use.

The layers of underground infrastructure are fascinating – everything from storm and sanitary sewers, subways, aqueducts, and other utilities – to active industry such as this. DC doesn’t have the same kind of active resource extraction, but it does have some massive water supply infrastructure that feeds the city’s reservoirs. Not all of it is active, either – but the vestiges of these underground operations on the surface of the city is quite interesting.

McMillan Sand Filtration site. Image from M.V. Jantzen on flickr.

This isn’t new ground for Mammoth. Mammoth’s interest in the forms of infrastructure and the design of spaces “looking for an architect” is fascinating, I always look forward to reading their thoughts on the matter. Of particular interest is the disconnect between designed, architectural spaces and networked, infrastructural ones. For some reason, there’s enough of a disconnect where the infrastructural frameworks lack the design gravitas – not everything can be a Calatrava-designed bridge, nor does that bridge alone show the true nature of the network’s design.

Metro snow operations

Given the heavy (and ongoing) snowfall, Metro is only operating rail service in select underground locations, in order to prevent trains from getting stranded as accumulating snow makes it difficult to maintain contact with the third rail, and also to use existing tunnels to keep rail cars dry and operable, rather than buried in snow and exposed to the elements in Metro’s rail yards.

The adjusted service map looks like this:

Metro's snow map. Image from WMATA.

Riding the trains today, the service is essentially single-tracking in the underground portions of the system. The segments of each open line in the middle have both tracks open, with each line essentially having two trains to cover the entire length of a line – they shuttle back and forth on a single track, passing each other on the double-tracked stretch in the middle.

Waits for the trains are long, and as is usual during single tracking operations, the PIDS aren’t all that reliable for train arrival times:

PID at U St, during snowpocalypse

In the single-tracked areas, the extra track is being used for train storage so that there are rail cars ready to enter service as soon as tracks are cleared:

Storage train riding out the storm at U St.

Metro did a good job of getting the system up and running again after the December 19th storm, opting to prepare the entire system for Monday’s rush rather than restore service immediately. They’ll likely do the same this time around.