Snow is likely this weekend, coming in two waves — the first Saturday afternoon and evening, and a second on Sunday into Sunday night. But there are serious questions about the intensity of the snow and how much will ultimately fall.

Temperatures, which will go below freezing while snow falls, should allow the snow to stick. However, if the snow remains light and spread out over a long period, it will be slow to pile up and it will be easy to clear and far less disruptive. But moderate to heavy bursts are possible, which could result in snow-covered roads and hazardous driving conditions.

The two periods when driving conditions could turn most difficult are Saturday evening, and during the day Sunday, depending on how far north the storm comes.

While model forecasts continue to jump around, we’re holding on to a storm total best estimate of 2 to 4 inches within a one-county radius of Washington. Most likely amounts drop to 1 to 3 inches in northern Maryland and increase to 3 to 6 inches in Washington’s southern areas, including Southern Maryland.

Around Washington, we believe there’s about a 70 percent chance of at least one inch, and about a 50-50 chance of at least 3 inches.

Over the past two days, computer models generally agreed that around 2 to 5 inches of snow would fall across the region from the two waves combined, but there were outliers that showed the potential for less or more. Some of the latest model forecasts have shifted lower and higher. It is not clear if these new forecasts are just a blip or a real trend toward a smaller or bigger snow event.

We are still two to three days from the onset of the event and models are still getting a handle on what Wes Junker describes as a “convoluted mess.”

“Part of what makes the forecast so difficult is there are two periods when we could see snow and Washington will sit close to the edge of where meaningful accumulation occurs both times,” said Junker, Capital Weather Gang’s winter weather expert. “The first wave of precipitation appears to be aimed to the west and north of the city. By contrast, the heavier snow with the second batch of precipitation is likely to be centered to the south of the city.”

Washington could “end up between the two areas of more significant snow on the short end of the proverbial snow measuring stick,” Junker added. But he also said we could get hit by one of the two or both of the batches. Just small shifts in where the snow sets up in either wave could be enormously consequential for total amounts.

Storm timing and evolution

First wave

It seems likely that snow will break out from the first wave between mid- and late Saturday afternoon. The snow could become steady and accumulate between a dusting and a couple inches Saturday evening before midnight. We may then see the snow ease or even stop for a time through early Sunday morning.

Some models suggest the heaviest snow amounts from this first wave may focus just north and northwest of Washington, with just a dusting close to town. But other models do indicate at least an inch or so could fall in the immediate Washington area.

Second wave

After a possible lull in the snow during the predawn hours Sunday, it may pick up again between early and late morning hours (models vary on timing). Depending on how far north the moisture comes, snow could become moderate to heavy for a time during the day Sunday, continuing well into the afternoon. New accumulation of one to four inches are possible, with the heaviest amounts likely in our southern areas. Areas from Washington north could just see a dusting from this second batch if the moisture feed from the south gets cut off from dry air streaming in from the north.

There’s also the possibility of some snow continuing Sunday night or even the early-morning hours Monday. Its intensity would probably be light but a little additional accumulation cannot be ruled out, especially south and southeast of Washington.

Storm scenarios

Given the forecast uncertainty, we see two equally likely scenarios with the outside chance of a third. We aren’t yet in a position to narrow the forecast down to one most likely scenario, but we hope to by tomorrow.

Here are the three scenarios:

Scenario 1: Moderate snow event scenario — 2 to 5 inches (50 percent chance)

The European model, which forecasts 3 or 4 inches, support this scenario. Its forecast has remained very steady for the past 48 hours, somewhat increasing confidence in its prediction. The UK model, which is the second most accurate model on average, also supports this scenario.

Scenario 2: Light snow event scenario — Dusting to 2 inches (40 percent chance)

The NAM and American (GFS) models, which forecast a dusting and 2 inches, respectively, support this scenario.

One might argue the American model forecast might better fit into Scenario 1. However, its snowfall forecast has been cut in half in the past 24 hours. You can see the remarkable reduction in its larger family of 20 simulations. On Wednesday, 8 of its simulations predicted at least 6 inches. Now, only one predicts this much.

Scenario 3: Heavy snow event scenario — 5 to 10 inches (10 percent chance)

The Canadian model, which forecast 6 to 8 inches, supports this scenario. We don’t particularly trust the Canadian model, which has the lowest performance scores.

Areas most likely to witness this heavy scenario are in Washington’s southern suburbs.

Correction: The original version of this story incorrectly indicated the UKMet model projected over 6 inches of snow. It actually forecasts about 3 to 4 inches. This story has been revised and we have slightly adjusted the likelihood of different scenarios accordingly.