
RodneyS
Members-
Posts
1,336 -
Joined
-
Last visited
Content Type
Profiles
Blogs
Forums
American Weather
Media Demo
Store
Gallery
Everything posted by RodneyS
-
A far more exciting contest this season than the last two, with six lead changes and anticipation in mid-February of a blockbuster storm at the four airports on February 19-20. Unfortunately for snow lovers, the Euro model's Valentine Day's tease of 16.1 inches in DC on those two days was correct only after the decimal point -- the other 16 inches went south. Still, that storm played a crucial role in determining this season's winner. To recap this season's contest, @LongRanger's zero across the board forecast maintained the lead until some early January snow propelled @RickinBaltimore into a brief lead on the 6th of that month. It appeared that @psuhoffmanhad wrested the lead away on the 7th, only to find that downward revisions to RIC's total had instead put @StormyClearweather in front. However, additional snow did eventually put @psuhoffmanin control on January 11th, and he held onto the lead for more than a month. A February 11-12 storm put @mappyinto the lead on the 12th. It was about that time that all of us began hearing about a potentially historic storm in the Mid-Atlantic about a week down the road. But by the time Valentine's Day weekend concluded on February 16th, it was clear that would not happen at BWI, DCA, or IAD, although there was still a possibility at RIC. Even when it became certain on February 18th that the major snow would fall mainly in southern Virginia, there was still a realistic chance of at least 4.3 inches at RIC, which would have put @southmdwatcherin the lead on a tiebreaker. After an early report of 2 inches there on February 19th put @mattie gin the lead, an additional 2 inches at RIC gave the lead to @HighStakes on a tiebreaker with @mattie gat the end of the day. The question then became, would February 20th produce at least another 0.3 inches at RIC? In a word, no, as only a trace fell there on that day. The final question was, would the rest of the 2024-25 winter season produce at least a little more snow? The answer was a ringing no, as the Mid Atlantic turned into Florida North in March. And so, the time has come to award the 2024-25 Championship Trophy to @HighStakeson a tiebreaker over @mattie g. This is only the second tiebreaker in the history of this 11-year-old contest, with the previous one having occurred six years ago, when @olafminesaw edged out @Stormpc. See you in November!
-
March 2025 was the 4th warmest March in DC history, with an average temperature of 54.3 at DCA, exceeded only by 2012 (56.8), 1945 (56.2), and 1921 (55.5). During last year's recordbreaking warm year in DC, March averaged 51.9 at DCA. However, January-March 2025 at DCA was well below last year, at 43.2 vs 45.4. The warmest January-March in DC history was 47.3 at DCA in 2012.
-
I'm still going to wait until next week to officially close out the contest, but it would take a minor miracle for more snow to fall at any of the four airports, and so your 5th place finish looks assured. You did better at SBY than our top two finishers, and I'm sure that co-leader @mattie gis wishing he had gone with your forecast of 13.8 inches there, instead of his forecast of 8.6 (and our presumptive winner @HighStakesforecast of 10.8). However@IronTyhad his 15 minutes of fame with a nearly dead-on forecast at SBY of 21.5 inches, compared with the actual of 21.3. FYI, @DanTheManhad the best forecast at LYH of 11.0 inches, compared with an actual of 12.4.
-
The 2024-25 astronomical and composite winters were each much cooler than last season at DCA. The former averaged 41.1 degrees, tied for the 29th warmest in DC history and down 4.1 from last year; while the latter averaged 41.3 degrees, tied for the 27th warmest in DC history and down 3.8 from last year.
-
Are you saying that because you can recall April 1938, Roger? Actually, there was only a trace of snow in DC then, after an average March temperature of 49.8 degrees. However, I can recall a stunning Mid-Atlantic turnaround that occurred much more recently. March 1990 at DCA averaged 50.2 degrees, and featured these maximums during March 12-16: 89/87/86/81/81. The Mid-Atlantic winter was over, right? Not so fast. During March 20-April 7, 1990, BWI received 2.8 inches of snow, DCA received 2.6 inches, RIC received 0.2 inches, and IAD received -- would you believe -- 10.4 inches? Moreover, each airport received measurable snow on either April 6th or 7th, with IAD checking in at 2.7 inches on the 6th and 1.1 inches on the 7th.
-
The Capital Weather Gang has declared winter over in DC, and they have a good track record of not doing so until the last measurable snow has fallen in these parts; see https://www.washingtonpost.com/weather/2025/03/10/winter-over-spring-starts-dc. While I will not award the 2024-25 trophy until at least April 1st, things are looking good for @HighStakes. Even if a little more snow falls at any of the four airports, it will likely be at BWI or IAD, where he can still improve. @southmdwatcherhas some hope, but only if RIC can eke out at least 0.3 inches, with little or no snow at the other three airports. Under a 0.3 at RIC and zero at the other three airports scenario, he, HighStakes and @mattie g would each finish with a departure of 10.6, and he would win a three-way tiebreaker. However, the far more likely scenario is a two-way tie between HighStakes and mattie g, with HighStakes winning the tiebreaker with his closer forecast at SBY.
-
February 2025 at DCA ended warm, bringing the average temperature to 41.0 degrees -- tied for the 34th warmest February in DC history. That brought the 2024-25 meteorological winter (December, January, and February) average to 39.0 degrees -- tied for the 43rd warmest in DC history. Nonetheless, the 2024-25 DCA meteorological winter was the coldest in 10 years -- surpassing every winter for cold since the 2014-15 winter averaged 36.7 degrees. The historical average for all 154 DC winters since recordkeeping began in 1871-72 is 37.1 degrees.
-
FYI, each of the four airports recorded only a trace of snow today. However, I am updating the table to show that -- barring corrections to the snow totals -- two more contestants have been eliminated from contention, to bring that total to 50 (in red), leaving only 19 who still have a chance to win. Most stunningly, I have determined that co-leader @mattie g has bitten the dust. Why is that? Because while he could still finish in a first-place tie with @HighStakes, he would lose the tiebreaker based on his forecast of 8.6 inches of snow at Salisbury (SBY) vs High Stake's forecast of 10.8 inches there. Because SBY has already recorded 17.8 inches this season, mattie g is history -- just as the other 49 of us are.
-
-
We have a new leader, as RIC has already recorded 2.0 inches today. That puts @mattie gin first place, with @HighStakesa big threat to move into a tie before day's end and @southmdwatcherlurking if RIC records several more inches.
-
Yes, and this season may make it 15 times. By the way, the largest amount by which DCA snow exceeded BWI snow was 5.5 inches in 1979-80 -- 20.1 inches at DCA, 14.6 at BWI. On the other hand, the largest amount by which BWI snow exceeded DCA snow was 20.9 inches in 2009-10 -- 77.0 inches at BWI and 56.1 inches at DCA. Both totals were record-breakers for any year in Baltimore and DC dating back to the 1880s, and the BWI total even exceeded IAD's record-breaking total of 73.2 inches that memorable snow season 15 years ago.
-
DCA became the official DC weather location in July 1945, and BWI began operations in June 1950. Thus, winter season comparisons between the two airports begin in 1950-51, and so there are now 74 winter season comparisons in the books. During those 74 seasons, DCA has averaged 15.5 inches of snow, and BWI has averaged 19.8 inches, with BWI snow exceeding DCA snow 58 times, DCA snow exceeding BWI snow 14 times, and two seasons ending in a tie.
-
Thanks. What I was going by was Wednesday's daily report for DCA issued at 12:35 AM this morning, which may be found at "Observed Weather" "Daily Climate Report" for "Washington-National" at https://www.weather.gov/wrh/Climate?wfo=lwx. The newest version is currently for Thursday, but you can find the 12:35 AM version for Wednesday by clicking "2" at the top (in blue). It's head-scratching how these contradictions occur, but you are correct that the snow total for yesterday is now back up to 0.5 inches. The below table reflects that change for DCA:
-
The DCA measuring gurus have helped you and a number of other folks by revising downward yesterday's snow total there from 0.5 to 0.3 inches. It would be interesting to know how these revisions come about -- perhaps some of the gurus don't realize that you are not supposed to take snow away from the measuring location to make a snowman? In any event, here is the slightly revised leaderboard, subject to further revisions without warning. P.S. I have listed in red all of the unfortunates who have been eliminated . . . whoops, that includes me.
-
Don't shortchange yourself, @snowfan -- you are currently 12th. Unfortunately, because you have all negative departures, your chance of winning are the same as mine and 38 others -- zero.
-
You aren't out yet, @nw baltimore wx. Your only negative departure is DCA, but the entrants ahead of you aren't doing well there either, and so if you nailed the other three airports you would win.
-
Apparently RIC did receive only 1.0 inch in this storm. However, I was 0.1 over the total at IAD, and so this revision corrects that. Quite a shakeup on the leaderboard, with some unexpected contenders this season. Yes, @Roger Smith, I say that with you in mind. Now, if @George BM gets into contention, @Jebmanwill be wondering why he did not enter.
-
The latest snow totals that I have for the February 11-12th storm are: BWI 3.8, DCA 6.4, IAD 4.9, and RIC 1.0. However, the latter total is likely incomplete, as for some reason the current Wakefield, VA report gives totals for every tiny town, but not RIC. Maybe @RIC Airportknows? Our new leader is @mappy
-
-
Recent January DCA Temperatures Appear Inconsistent with 1981-2010 Normals
RodneyS replied to RodneyS's topic in Mid Atlantic
An illustration of how anomalous the period January 12th-22nd at DCA has been in recent years can be seen by comparing the 1991-2020 daily temperature normals for each day of the year (minus February 29th) calculated by the National Centers for Environmental Information (NCEI), with the actual daily averages during that 30-year period. By far the two largest daily negative anomalies (averages are below the normals) for the calendar year are for January 21st-22nd, and by far the two largest positive anomalies (averages are above the normals) for the calendar year are for January 12th-13th. The below list shows the Top Ten daily temperature anomalies (daily average minus daily normal) at DCA during 1991-2020. It can be seen immediately that there is nothing comparable to the period January 12th-22nd. December 8th-9th has consecutive negative anomalies of -3.28 and -2.97 degrees at DCA, but the largest consecutive December positive anomalies there are not listed below because they are only 1.13 and 2.15 degrees, respectively, on December 22nd-23rd. 1) January 21st -4.43 degrees 2) January 13th +4.37 degrees 3) January 12th +4.32 degrees 4) January 22nd -4.20 degrees 5) February 13th -3.78 degrees 6) December 8th -3.28 degrees 7) March 4th -3.20 degrees 8) April 29th -3.02 degrees 9) December 9th -2.97 degrees 10) April 5th -2.92 degrees I believe that the most plausible explanation here is that the DCA 1991-2020 temperature "normals" calculated by the NCEI for January 12th-13th and January 21st-22nd are not truly normals in any meaningful sense, but are simply the end product of a mathematical process based upon there being no known reason for temperatures on those four calendar days to deviate so significantly from temperatures on nearby calendar days. However, if a reason were to be found, the NCEI would likely have to increase the DCA January 12th-13th normals by about four degrees each and decrease the January 21st-22nd DCA normals by about four degrees each. -
Quite a contrast to two years ago at DCA, even if January did finish on the warm side this year. This year's January average temperature was only 34.2 degrees -- 11 degrees below January 2023. That put DCA not only 3.3 degrees below the 1991-2020 January normal of 37.5, but also 1.2 degrees below the 1871-2025 historical DC average of 35.4. January 2025 tied for only the 92nd warmest January in DC history and was the coldest January at DCA since 32.2 in 2014.
-
Recent January DCA Temperatures Appear Inconsistent with 1981-2010 Normals
RodneyS replied to RodneyS's topic in Mid Atlantic
Thanks for this analysis, Tom. Regarding February in DC in recent years, your memory is spot-on. If you compare DC temperature averages during 1872-1983 and 1984-2024, the average monthly increase from the earlier period to the latter period is 2.7 degrees. But as I think you suspected, February has experienced the largest increase of any month, at 3.6 degrees. December is second largest at 3.5 degrees. So, in keeping with the idea that the winter is being pushed back a couple of months, January must be third largest, right? Wrong! January has the second smallest increase, at only 2.2 degrees. Why? Because while a number of January days have had large daily increases, including the largest of the year on January 13th (7.4 degrees), a number of other January days have experienced temperature decreases, including the largest of the year on January 21st and 22nd (-4.6 degrees on each day). So anyone attempting to calculate recent daily January Temperature Normals in DC is in for a head-spinning experience. Regarding "Winter conditions more common in Mar/Apr, Summer conditions in Sept/Oct" -- the numbers do not support that in DC. Between 1872-1983 and 1984-2024, March has increased 3.1 degrees (3rd largest), and April has increased 3.0 degrees (tied for 4th largest), while September has increased only 2.4 degrees (4th smallest), and October has increased only 2.5 degrees (5th smallest). So, the picture is quite muddled, but January is in a class all its own. FYI, here are each of the average monthly temperature increases in DC from 1872-1983 to 1984-2024 in ascending order: May 1.7, January 2.2, June and September 2.4, October 2.5, November 2.7, July 2.9, April and August 3.0, March 3.1, December 3.5, February 3.6.