How warm has it been in the Washington area in the fall? Warm enough that most of us are still waiting for the first freeze. But our time is up; a hard freeze threat is coming Saturday thanks to a late week arctic blast.

In Washington’s western suburbs, we’re now three weeks behind schedule for the first freeze. In some places, that’s as late as it’s been in more than half a century. Even so, in the urban center of Washington, a freeze in the days ahead would be the earliest since 2006 and more than a week ahead of schedule.

Normally, the first autumn freeze occurs by mid-October in the city’s western suburbs. While a few spots saw a light frost in the back half of October, none of Washington’s major climate stations have officially hit 32 degrees. This is an increasing oddity as we head into the second week of November.

As evidence of the oddity, Dulles Airport is now in record territory. This year’s first freeze (Nov. 10) was the second-latest on record, behind only 2005. As of Nov. 2, Dulles already surpassed its longest freeze-free period on record. The airport went 214 consecutive days without a freeze — more than a week longer than its previous record (206 days) set in 2011.

Washington region freeze factor

What determines the timing of the first freeze in our region? And why is there such a large range over such a short area? Here are the three key factors:

  • Elevation —  Higher elevations freeze soonest, although this is most notable west of our area.
  • Water — Water retains heat better than land, so locations nearest the Potomac River and Chesapeake Bay freeze later.
  • Urban heat island — Concrete and city infrastructure keep it warmer than more rural surroundings.

Typical first freeze at Washington’s major airports

While airports aren’t necessarily the best location to measure temperature, they tend to have the longest and most continual records. That makes them a decent starting point for comparing the timing of the first freeze across the region.

Below is the average date of the first freeze at each of Washington’s major climate stations (based on 1981-2010 climate averages compiled by the National Weather Service). Note that the average happens to be nearly identical to the median date of the first freeze, referenced in the map above.

  • Washington, D.C.: Reagan National Airport (DCA) — Nov. 18
  • Dulles: Dulles Airport (IAD) — Oct. 15
  • Baltimore: Baltimore-Washington Airport (BWI) — Oct. 28

Of course, the average date masks a lot of variability from year to year. Below we’ve charted the date of the first freeze at all three airports. (Records at Dulles began in 1963, so we’ve limited the time period to this common start date, even though climate records for Washington and Baltimore date back to the 1870s.)

One takeaway from this graph is that the arrival of the first freeze has been trending later, though not uniformly so at each location. In the District and at Dulles, the average date of the first freeze has shifted roughly two days later per decade. In Baltimore, the timing of the first freeze has stayed fairly consistent, gaining only a few days over the entire period.

First freezes in September and December 

In the District, the earliest and latest first freeze on record are more than two months apart! Note that the earliest freeze in Washington’s entire climate record — which dates back to 1872 — was on Oct. 10, 1895, when the official weather station was located downtown.

The latest first freeze at the current location, National Airport, on Dec. 22, 2001, is also the latest in Washington’s entire climate record. As the table below shows, all of Washington’s latest freezes have occurred in December. Seven percent of Washington’s first freezes have happened in December, and among these about half have occurred since the year 2000.

These days, an early freeze (see map) in our western suburbs would mean about the first 10 days of October, with mid-to-late October the target in and around the I-95 corridor. A late freeze (see map) would be roughly Halloween to the first week of November out west and northwest of the city, and mid-November along I-95. A late freeze in Washington and the warmest spots near the bay could easily mean December.

A very big range of dates

Anyone who has spent some time here knows that National Airport (DCA), the District’s official climate station, is typically an outlier in the region for much of the cold season. To better quantify the “DCA problem,” we asked our friend Brian Brettschneider, an Alaska-based climatologist, for some help.

His map below illustrates how nearly all of the major climate stations in the Mid-Atlantic and South record their first freeze around the same time, if not before, National Airport. Another way to think about it is that Washington’s typical mid-November first freeze is on par with a place like Tallahassee, Fla.

You can see in the map that there are a few “after DCA” dots in other urban heat island locations across the Northeast, and then closer to water in the D.C. region. But for the most part, only near the Gulf of Mexico and other coastal areas is the median date of the first freeze consistently after the first freeze at National Airport.

National Airport aside, a majority of our area typically sees its first freeze during the second half of October. A good rule of thumb is if you live outside the D.C. and Baltimore beltways and aren’t near water, you’ll usually see a freeze before November. That’s especially true in parts of western Fairfax and Loudoun counties near Dulles Airport, as shown in the map at the top of this post.

At the other extreme, some of the area’s unofficial climate stations, like the Baltimore Science Center and Inner Harbor, typically don’t see their first freeze until late November or early December. At higher elevation, a place like Damascus, Md., might expect a first freeze with about 10 days left in October, while Salisbury, Md., shouldn’t expect one until early November.

This year, it appears most of the region will record its first freeze around the same date. An early freeze at National Airport this weekend would be ironic, considering that most of our area is running three weeks behind schedule. Case in point for how the timing of the first freeze at that location is hardly representative of the greater D.C. region.

This post has been updated to reflect freeze data through Nov. 11, 2017.