Network News

X My Profile
View More Activity
The new Washington
Post Weather website
Jump to CWG's
Latest Full Forecast
Outside now? Radar, temps
and more: Weather Wall
Follow us on Twitter (@capitalweather) and become a fan on Facebook
Posted at 11:15 AM ET, 01/25/2008

What's Happened to January?

By Matt Ross

Our coldest month -- an average of more than three degrees colder than February, and almost five degrees colder than December -- has been mostly a nonperformer of late. The DC area, as measured at Reagan National Airport, is currently facing its fourth consecutive January of above-average temperatures, and the 16th warmer-than-average January out of the last 20.

Why has our coldest month been a dud? Or has it? The devil is in the details.

Official monthly (and daily) averages are based on 30 years of data, which currently spans the period from 1971 to 2000. (Averages are updated every 10 years; starting in 2011 the 30-year period will shift forward to 1981 to 2010.) So, the average January temperature is simply calculated by adding up the average temperatures for all 30 Januarys and dividing by 30, right? Well, not exactly.

It turns out the averages are smoothed, such that data from unusually warm January days -- like Jan. 7 and 8 of the current month, which saw highs at Reagan National reach 69 and 73, respectively -- are adjusted so the daily average meshes more closely with surrounding daily averages. For example, let's say the average high for Jan. 15 over the 30-year period is 43 and the average low is 27. And that for the next day, Jan. 16, the average high is 48 and average low is 30 (a significant difference for two consecutive days), because it just so happened that unusual warmth struck on that date a couple times over the 30-year period. In this case, the data for Jan. 16 would be tweaked downward to fit better with the surrounding data.

Since January temperatures tend to be unusually high more often than they are unusually low, the official January average is a bit colder than the "actual" average. Officially, our 1971-2000 average temperature for January is 34.9 degrees. Yet if we calculated the true average for the 30 Januarys over that period, the result would be about a degree higher. Smoothing has made the official January average colder than it really is. What does all this mean for our current streak of above-normal Januarys?

Januarys that came in officially at slightly above average (e.g., 2000, 2001, 2005) are actually right around the true 30-year average, and perhaps slightly below. While there is no question that January has skewed warmer rather than colder in the last 20 years, the trend is not as significant as it looks. So when the weather pundits declare "another warm January," we need to assess how warm. Because calculating average temperatures isn't as straighforward as it may first seem.

Of course, this January will likely finish four to five degrees above average. Warm by anyone's standards.

By Matt Ross  | January 25, 2008; 11:15 AM ET
Categories:  Local Climate  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Forecast: A Cold Front True to its Name
Next: CommuteCast: Mid-Winter Chill Continues


I actually always wondered how the averages were calculated. Becuase if you use "actual" averaging, then if each consecutive January is warmer than the past, the "average" would rise. Thanks for the post.

Posted by: jeffc | January 25, 2008 11:36 AM | Report abuse

The 12z GFS is coming around to the DGEX solution for late next week. GFS now suppresses the track across nw South Carolina ne across se Va. to off the coast. This is at least 100 miles se of the 06 track and puts our western zones on the borderline of a significant snowstorm, which the DGEX already gives us.

Regarding Jan. temp. trends. The CPC often gives overwhelming weight to persistence when producing long range forecast packages, which may or may not result in accuracy.

Posted by: Augusta Jim | January 25, 2008 1:07 PM | Report abuse

Ok, so then, if this January tracks to be 4-5 above normal, does that have any meaning in the larger scale of things? For example, how does this relate to La Nina years, is it normal, above normal, etc? Were the last several years way above normal, just slightly? Is this a big departure?

I realize the article was about how these averages are made (and its a good one BTW), but whenever people start saying "it was a warm January", I always want to ask "what does that really mean? As compared to what?"

Posted by: jbroon | January 25, 2008 1:58 PM | Report abuse

Jbroon....This is not inconsistent with other mod/strong La Nina years..This January will probably finish colder than 2006 and 2007 but warmer than 2000, 2001, 2003, 2004, 2005....Warmer than normal, even for this decade, but consistent with the decadal trend of warmer mid winters

Posted by: Matt Ross, Capital Weather Gang | January 25, 2008 2:20 PM | Report abuse

Even more profound than the lack of cold is the lack of snow. The average annual snowfall for D.C. is 16.7", am I correct? So far this year it has snowed twice, with an accumulation of around 1 or 2 inches each time (I don't have the actual statistics). That means we must be running at least 10" behind in snowfall this year, with nothing on the horizon. According to most long range forecasts, temperatures for the next 15 days stay in the 40s to 50s with, at best, a chance for a rain/snow mix or flurries on some of the colder days. Pathetic for snow lovers!

As for me, I'll take the month of July when it's sunny and 90!

Posted by: Anonymous | January 25, 2008 3:09 PM | Report abuse

Ditto on the sunny and 90.

I'm counting the days til we can put away the SLCB for the year.

Posted by: Ivan | January 25, 2008 3:42 PM | Report abuse


Actually we are not THAT far behind snow wise. This is because, climotalogically speaking, the second half of January through late February are the snowiest parts of the year. That is, snowfall in this area is "back loaded" - the later months in winter have more of it than the early ones.

Posted by: Jim in Blacksburg | January 25, 2008 3:44 PM | Report abuse

From January 1949 till December 1954, every winter month was warmer than normal except December 1950 and January 1954 (and those were below normal by a degree or less).

Every winter from 1948-49 until 1956-57 was below normal in snowfall except 1953-54 (due to a 6.6-inch storm on November 6, the most ever for so early in the year).

Posted by: Anonymous | January 25, 2008 4:11 PM | Report abuse

hey dudes. Giants are gonna win da superbowl. Anyway, weather is lookin nice this weekend. finally warm weather.

Posted by: f | January 25, 2008 11:10 PM | Report abuse

Hey Matt or someone who does the weather section...If the NYT (in the winter) publishes daily the snow depth (cumulative) along with other liquid precipitation in their daily weather page, why not the WP?? There is certainly room in a small box for that to happen. Maybe the totals for both Dulles and National since levels vary around these parts.

Posted by: Arnie Daxe | January 26, 2008 4:13 PM | Report abuse

I want to see palm trees on the White House lawn.

Posted by: Anonymous | January 28, 2008 7:23 AM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2012 The Washington Post Company