Roger Smith

Two recognized methods for snowfall contests are sum of errors or sum of errors squared (which tends to be closer to correlation co-efficient). It's often a similar but not exactly same order of finish using those two methods. As Don and perhaps a few others here will know, DonSutherland has a very good record in even larger snowfall contests that ran for many winters on another site, and disappeared (sadly) this winter. In storm forecasts DonS has often been either top or top three in a field of 20-25 entries. I have had a much more variable record in the same contests but I did nail the Blizzard of 1888. Well 2010, one of those snowmageddon ones. And Nemo, or Juno (get them confused). But I have finished dead last a few times as well. Don is rarely out of the top third, maybe even never. There are many pros who don't know 10% of what Don knows. We have an annual snowfall contest over at the temperature forecast contest thread (main forum, nobody knows it's there) ... Don does well in those too, so does wxallannj who is a member here I believe, also RJay, hudsonvalley21, BKViking, dmillz25 from here (apologies if I missed anyone). Other Am-wx people with good track records in contests would include RodneyS, Tom, StormchaserChuck and wxdude64, plus Scotty Lightning who was once SD. Last year in particular so_whats_happening was very close to top. Various others who used to be great like Mallow, Stebo, DIT, Isotherm ... have dropped out over the years, in some cases probably because the NYC crew kept kicking their butts. The Mid-Atlantic forum is about on a par with NYC for contest exploits over many years.

Probably deep into phase 9.

DCA reported 1.99" liquid and 6.9" snow. If I'm generous and say .09" of that was freezing drizzle (probably an overestimate) then their snow report equals real snow plus sleet at some possible-to-calculate ratio after an assumption is made on the snowfall to liquid conversion. I will assume 10:1 in my calculation that follows. So let's say 5" real snow, then we have 5.0 + 1.40*(ratio/10) = 6.9 That reduces to 1.40*(ratio/10) = 1.9 and makes ratio/10 = .21, in other words a 2.1 : 1.0 ratio. Sleet ratios are often quoted at 3:1. For 3:1 to be correct, then snow at 10:1 is 2.7" .. that is clearly a low estimate based on regional reports. So either this was very watery sleet or very underestimated sleet. I would guess what really happened is that 3" of sleet pounded 6" of snow down to 4 inches and sat on top giving 7 inches.

12z GEM has a 960 low south of Long Island. I can guarantee you if that verifies the Boston train will not be pulling into Grand Central on time, or perhaps ever.

Some early maps look like the blizzard of 1888. But why stop there?

Patriots vs Hawks ... seems like a Trump cabinet meeting gone horribly wrong.

Potential storm is near full moon, would be big storm surge issues in e MA on some depictions. I have not formed an opinion on actual probability of this happening anywhere near land yet ... my usual method is to let smart people make that decision and then double down. Worked like a charm on the last one.

A hit at RIC (my residual is 3.0") combined with smaller amounts than I have left at the other three will knock out half to two-thirds of the remaining contenders no matter what happens later. I am the only one left with RIC snow to spare until you get way down into the big snow forecasts that will knock me out later I hope. -- except that Mappy also has 1.5" left and similar residuals elsewhere, so I guess together we could knock out all those others (residuals achieved = 2.0 x differential achieved). Our details on the other three are a bit different so mappy could survive a RIC 1.5 to 3.0 addition depending on how those break down. There are probably ways each of us could eliminate the other even if RIC goes past 3.0" additional. But that might be within a scenario where a larger set of forecasts eliminates both of us too. Another well-placed RIC residual belongs to wxdude64 with 3.1" left ... if my other three numbers get overtaken by very much, wxdude64 is waiting to finish me off a few inches into the overdraft. The RIC component is worth 0.2" alone to wxdude64 (by the 2 to 1 rule). Basically I could be done halfway through any big storm that includes RIC, and before that if RIC is excluded. My best chance is a bunch of small events that nickel and dime towards my residuals and are generous to RIC. (once it goes past my residual, the 2:1 rule changes to 1:1 no advantage for further amounts against lower forecasts) Exciting times, a contest with some suspense for the first time in several winters. With the pattern we have in place, some of the bigger numbers stand a chance but I like wxdude64's current situation and think that could be the favorite for bets right now.

BOS was past 19 inches last time they reported so I am hopeful of a W for BOS and that's the big highlight for my participation. I think final numbers will look something like NYC 11.5 BOS 20 BDL 16.5 ALB 12 PHL 9.3 DCA 6.9

Turns out that NYC tied the daily precip record of 1.80" for Jan 25, the earlier mentioned 1.82" includes .02" after midnight apparently. So it's a tie with the 1978 rainstorm component of the Great lakes blizzard/superstorm low that gave some places 15-30 inches of snow overnight into 26th Jan (1978). I will adjust any tables I have posted at end of month giving NYC a chance to settle their accounts, sometimes what you see one day is not there forever. Don, here's a challenge, I could work on this too, but what is the largest differential between a record snowfall and record precip or any precip on the 10:1 basis? I realize yesterday's storm was bulked up for precip by sleet that was also in the snowfall total. In Canadian climate records this might appear something like 0.20" rain and 12" snow for 1.80" precip. Canadian climate records show separate rain and snow amounts for every day and they seem to handle sleet a bit differently than U.S. practice. I am aware of one case where 40 mm of precip was really 10 cm of sleet but they give a 10 cm snow total for that case, and 30 mm rainfall. Part of the problem (if this can be described as a problem) is that Canadian observing practice was strictly 10:1 snow to precip untiil 1962 then changed to melt-it-down whatever-it-actually-was conversions after 1962. Thus if I look at some old storm record and it says 10.0" snow, 1.15" precip and 0.15" rain, I cannot say for sure whether that was rain, freezing rain, sleet melted down or whatever. Anyway, yesterday's differential was about 0.66" from 10:1, some values in Feb 1920 are definitely larger on a daily basis.

Toronto City as it is now known used to be a first-order station from its founding in 1840 through to about the 1980s, it gradually became more of a well-maintained climate station but the data have been reported more or less daily all along and I took it upon myself to maintain the data base even though Environment Canada nowadays seem to ignore that data base and only give out press releases about YYZ (Toronto airport, located about 12 miles west of downtown Toronto) where the period of record is 1938 to present. So there are a lot of very significant climate records like the 1936 heat wave that aren't in the YYZ data base. Toronto City has always been located somewhere within 0.5 miles of its longest situated location at the old headquarters building of the former Dominion of Canada weather service at 315 Bloor Street West. This building, next to Varsity Stadium, is now part of the U of T campus. Before 1908 the observations were made about 0.5 miles southeast of there near what is now Hart House and was then Kings College, basically a block west of Queens Park (the Ontario legislature building). Some time around 2003 they stopped making observations at the old h.q. building (which had been replaced by a much larger modern version in suburban Toronto), and moved the instruments to a suitable well-exposed location near Trinity College on the university campus. I assume it is read once daily around 0800h and also the instruments seem to be capable of sending some data hourly because the station reports are hourly as well as daily even nowadays. There was a problem with missing data, I worked through all missing data cases and found that 90% of them could easily be estimated from the hourly observations because those "missing" days were only missing an hour or two and not usually ones critical for estimating max and min temps. Whether by luck or design, almost all the missing days (I would say maybe a total of 50 from 2003 to 2025) are evidently dry days from regional zero reports. I also found that 2020 rainfall data had been compromised to some extent by daily small values that appeared to be the result of lawn sprinkler water getting into rain gauges (Toronto City was showing rain almost every day for several months when other stations were dry). So it's a bit of a challenge to maintain this data base but I have done it best as I can -- as they close in on 200 years of data, I feel they should duplicate my work (I know from experience they won't accept anything I say as in Canada I am basically a non-person) and bring the records back from the dead. Up until around 1980 there was a fully maintained data base with daily records etc. So I have basically extended that to the present day. I was going off memory quoting some of the all-time records, now that I have my file open I can be more precise. The existing one-day records and top ten calendar day amounts (any before 1978 were originally recorded in inches and those after 1978 have been converted from metric) for Toronto City are 1. 19.0" Dec 11 1944 t2. 18.0" Feb 14, 1850, Feb 22, 1846 t4. 16.0" Feb 5, 1863 and Mar 27, 1870 6. 15.7" Jan 23 1966 t8. 15.0" Jan 20 1867 and Mar 21, 1867 and Dec 25 1872, Dec 29 1855 12. 14.9" Jan 2 1999 (38.0 cm) Two-day record values (probably overnight storms) are 23.0" Dec 25-26 1872 (15.0 + 8.0) 22.5" Dec 11-12 1944 (19.0 + 3.5) 22.0" Feb 22-23 1846 (18.0 + 4.0) 20.0" Mar 15-16 1870 (10.0 + 10.0) 19.5" Jan 23-24 1873 (14.3 + 5.2). 18.0" Jan 20-21 1867, 16.6" 18.0" Feb 5-6 1863 (16.0 + 2.0) and also 24-25 1868 (12.0 + 6.0) (18.0" Feb 1850 daily value was not increased by any amounts either side) 16.6" Jan 22-23 1966 16.5" Mar 21-22 1867 15.5" Mar 8-9 1931 (12.5 + 3.0) also Mar 20-21 1876 (9.0 + 6.5) _______________________ The Nov 2d record is 12.5" from Nov 24-25 1950 and the April record is 9.4" from Apr 2-3 1975. As you can see almost all the top ten snowstorms were before YYZ data began. The average snowfall in 19th century winters was about 25% greater than any 30-year averages in the 20th century. 1869-1870 was the snowiest winter back in that era.

May have been two calendar days like an overnight storm? Toronto has never had a 20" calendar day storm (I have not had time to check the latest on yesterday) but it has three or four overnight 20 to 23 inch storm totals. I recall doing a study on all 8.0" two day totals for Toronto (hundreds of data points) and about half of them had a significant amount on a second consecutive day. So only about half of all snowstorms there manage to occur all within one calendar day, or mostly if we eliminate the 0.1 and 0.2 add-ons. A few storms require three calendar days to play out to their totals but I never did a census on those because in a fast-moving pattern there can easily be two separate events contained in three days of snowfall data.

Preliminary and needing to be checked from later data, but Toronto (downtown) may have had its heaviest January one-day snowfall ever (since 1843 when snowfall records begin). Reports are in the 18-20 inch range. This was 80% lake enhancement boosting a wider 5" synoptic scale snowfall. The previous record was 44 cm (meas 16.6" at the time) set Jan 23, 1966 (not the later coastal blizzard on 1-29). A very close second was 40 cm Jan 2, 1999. After that one and two later moderate falls, Toronto had its greatest snow depth of 65 cm (26 in). The current one-day record for any month was 19.5" (about 47 cm) Dec 11, 1944. February and March also have higher one-day records than January but that may not be the case after this new record is verified.

I took a blend of the Eduggs and the DonS models and assumed the NAM might be right about the sleet. But I think my broken abacus was a problem.

How is he getting out?

For downtown Toronto the old January daily record was 16.6" set on Jan 23, 1966. The one-day record for all months is 19.5" from Dec 11, 1944. Two-day record I believe is 23.0" Dec 25-26 1872. So yes this looks like an all-timer for January 1d for Toronto City (records began 1840, snowfall 1843). I have a thread on Toronto and New York City climate records in the climate change forum. Will wait for further data before updating and will report back on exact details, there is still a reporting station in downtown Toronto but they only give precip and snow depth obs. I have been going by those and using best estimates on recent snowfall amounts since actual daily snowfall reports stopped happening in 2017. Looks like this was 80% lake enhancement and 20% synoptic scale which is similar to the 1872 storm which took place during a frigid spell of record low temperatures with NYC also seeing 18" of snow on Dec 26th (1872). Last February Toronto was flirting with Feb 1846 snowfall records. March 1870 is the snowiest month of all time with four big falls that added up to over 60 inches. Also the record snow depth for Toronto downtown is 65 cm (around 26 inches) set in January 1999 after a number of snowstorms, including another 15-16 incher on Jan 2 but the record was not reached until nearly mid-January. Famously the Canadian army was called in to shovel the snow on that occasion. I have seen 30 inches of snow on the ground north of Toronto as recently as April 3, 1975 after a big storm in that region.

Offshore low still deepening, Nantucket 54SE was gusting to ESE 46 knots for a while, backed off a bit last hour, temp there is low 40s. Looks to me like 10-12" more could easily fall in many parts of MA and ne CT, 5-10" more further west. I was surprised at the low ratios in early snowfall reports from airports, contrasting with very high ratios in NY and NJ early reports. Is that something to do with ocean effect atmospheric chemistry?

Making a run at the daily precip record of 1.80" in 1978 (for Jan 25), the warm sector phase of the famous Great Lakes blizzard of Jan 25-26 1978. That may have included a bit of snow or sleet but was mostly if not all rain in NYC. Also the snowfall is now (I think) ahead of the old record value of 10.0" from 1905.

Looking at offshore buoys, temps mid-30s about 30-50 miles south of Long Island south shore with winds 25 to 40 mph (40 to 60 southeast of Nantucket). Pressure falling steadily towards predicted central pressures near 998 mbs. If this had deepened to 970 like some coastals have done, can you imagine the snow totals? 30-40 inches easily. I think what we're seeing is a modern climate version of the Blizzard of Feb 1899 which also followed severe cold and dropped 14 to 20 inches from DC to Boston. The one difference would be that storm evolved more like the March 1993 storm from the eastern Gulf up the Carolina coast, and had no inland primary, just the coastal bomb. The blizzard of 1888 also followed a similar track for the coastal portion but again had no inland primary. I wonder if Don has any way of finding any other storm that dropped 10" of snow in Columbus OH, NYC and BOS (if those numbers verify, CMH already has). As to storms that took place in frigid temps, blizzards of 1899 and 1857 come to mind, also Dec 26 1872, an 18" snowstorm at both New York City and Toronto (lake effect there mostly) with record cold in place over the region. I imagine the sleet bomb with that one was confined to the Delmarva and maybe far southern NJ.

Hard to imagine but Jan 24 1967 (68, 54) was fifty degrees warmer than yesterday was in NYC, and sixty degrees warmer than Jan 24, 1882 (6, -6). Jan 25, 1967 is still the warmest day in Toronto's now 186 years of January records (61 F).

In 1920 a strong arctic high settled over the region on Jan 31-Feb 1. Temperatures fell below zero overnight setting records, but both days had highs in the low 20s, probably midnight highs at both ends of the cold spell. Then a coastal moved slowly north and Feb 3-5 all had over 1" of liquid and over 5" of snow each day. I would imagine the mix was similar to the current storm except that it lasted longer. I've seen one old archived photo of horse-drawn carts and old-fashioned cars semi-stuck in a goopy looking mess but I wonder if there are other news reports from that series of events. The storm brought a very cold two weeks to a close and transitioned to a more average sort of pattern in Feb 1920. Another historical note, today's NYC record of 10.0" from 1905 was part of a two-day total of 11.0" and that storm was also a coastal low. Going back 48 years, residents of the Great Lakes region were told by forecasters to expect a big snowfall the next day, but the models of the time showed a deep low tracking from Alabama to Lake Ontario; nobody was quite prepared for the Cleveland superbomb result (955 mb at Port Huron-Sarnia with low moving due north into Lake Huron). At London ON, winds gusted to 80 mph from the south but it was an arctic wraparound! Temps had dropped from near 40F overnight to 20 F. The storm dropped 15-30" of wind-blown snow over most of Ohio, Michigan and southwest to northeast ON. Toronto had much less snow (about 4" on a SSW wind of 40-60 mph) and saw temperatures drop 15 F deg in two hours from 8 to 10 a.m. on the 26th. I was actually in a weather office plotting maps during this storm, to support air quality forecasts made by the company. My parking spot (in northwest Toronto) was occupied by the housing for the building's air conditioning unit, luckily for me it arrived a few minutes before I did that morning. Jan 26 is a date for huge Great Lakes storms, there was the Chicago blizzard of 1967 and a monster lake effect storm in 1971 that trapped hundreds of people in central Ontario for up to a week. During the 1967 storm, both Toronto and NYC had record highs on Jan 25 but in Ontario the warmth was followed by a sleet and snow storm lasting two days; eventually we went from 60 degrees F to 10 inches of frozen snow on the ground. It then stayed very cold for about six weeks.

Would expect it to get increasingly foggy this afternoon as temperatures rise to mid-high 20s with sleet transitioning to freezing drizzle, ice accretions should be fairly minimal, then later this evening and overnight a colder trend returns and snow grains will start to fall mixed with light sleet, before snow flurries in 10-15 F temps by morning. A further inch of snow could accumulate eventually on top of a frozen slush. This will not apply to parts of DE and se MD where temps will rise to near 40 F and straight rain will fall for a brief interval.

71/69 New Orleans and 17/10 Little Rock, 1006 mb low near Jackson MS. Coastal low looks to be in formative stages east of C Hat where it's 1019 mb and 48F. Below freezing inland as far south as central GA.

FLORIDA EXTREME COLD WARNING __ Temperatures may drop below 60 degrees. You may need a sweater or a light jacket. DON'T RISK YOUR LIFE pack a sweater or jacket.

__ Table of scoring for January 2026 __ FORECASTER ____________________DCA_NYC_BOS_ east _ORD_ATL_IAH _ cent _ c/e__ DEN_PHX_SEA__west___TOTAL Roger Smith ______________________ 66 _ 90 _ 80 _236 _ 64 _ 92 _ 78 _ 234 _470 _ 64 _ 84 _ 92 _240___710 wxdude64 ________________________ 58 _ 78 _ 92 _228 _ 70 _ 80 _ 94 _ 244 _ 472 _ 68 _ 56 _ 86 _210 ___682 Tom ______________________________ 54 _ 84 _ 90 _228 _ 86 _ 76 _ 96 _ 258 _ 486 _ 60 _ 38 _ 96 _194 ___680 StormchaserChuck1 ______________ 44 _ 38 _ 96 _ 178 _ 84 _ 88 _ 64 _ 236 _414 _ 76 _ 84 _ 98 _ 258 ___672 BKViking _________________________ 46 _ 62 _ 86 _ 194 _ 66 _ 90 _ 98 _254 _448 _ 56 _ 46 _ 80 _182 ___630 dmillz25 __________________________56 _ 84 _ 92 _232 _ 24 _ 74 _ 68 _ 166 _ 398 _ 94 _ 54 _ 82 _230 ___628 so_whats_happening _____________38 _ 54 _ 96 _ 188 _ 44 _ 76 _ 80 _ 200 _388 _ 84 _ 46 _ 92 _222 ___610 hudsonvalley21 __________________ 42 _ 38 _ 90 _ 170 _ 64 _ 94 _ 80 _ 238 _408 _ 40 _ 56 _ 92 _ 188 ___596 ___ consensus __________________ 28 _ 58 _ 90 _ 176 _ 40 _ 76 _ 86 _ 202 _ 378 _ 68 _ 46 _ 94 _208 ___586 DonSutherland1 __________________ 44 _ 52 _ 98 _ 194 _ 38 _ 64 _ 42 _ 144 _338 _ 86 _ 68 _ 72 _226 ___564 wxallannj _________________________ 06 _ 42 _ 76 _ 124 _ 42 _ 58 _ 74 _ 174 _ 298 _ 86 _ 46 _ 98 _230 ___528 (Persistence _ see below) _ 500 ____ Normal _______________________16 _ 34 _ 62 _ 112 _ 34 _ 92 _ 82 _ 208 _320 _ 46 _ 14 _ 88 _ 148 ___468 Scotty Lightning _________________ 00 _ 14 _ 72 _ 086 _ 24 _ 78 _ 88 _ 190 _276 _ 56 _ 34 _ 98 _ 188 ___464 RodneyS _________________________ 30 _ 46 _ 72 _ 148 _ 28 _ 54 _ 52 _ 134 _282 _ 56 _ 28 _ 96 _ 180 ___462 RJay ______________________________16 _ 34 _ 62 _ 112 _ 34 _ 48 _ 38 _ 120 _232 _ 66 _ 64 _ 88 _218 ___450 Rhino16 __________________________ 00 _ 00 _ 32 _032 _ 38 _ 38 _ 48 _ 124 _156 _ 38 _ 00 _ 82 _120 ___276 ============================== Persistence (Dec 2025) __________ 96 _ 60 _ 50 _ 206 _ 96 _ 74 _ 38 _ 208 _ 414 _ 00 _ 38 _ 48 _ 086 ___500 ____________________________________ EXTREME FORECAST REPORT DCA (-4.2), NYC (-3.3) are wins for Roger Smith with coldest forecasts. BOS (-1.9) does not qualify as an extreme forecast, fourth coldest forecast was high score. ORD (-3.3) is a win for Tom with coldest forecast. ATL (+0.4) does not qualify at 0.4, would be a win-loss situation below -0.3 IAH (+0.9) is a win for BKViking (+1.0) and a loss for wxdude64 (+0.6) and Normal. DEN (+2.7) is a win for dmillz25 (+3.0) and a loss for Roger Smith (+4.5). PHX (+4.3) is a shared win for StormchaserChuck and Roger Smith with highest forecasts. SEA (+0.6) does not qualify as an extreme forecast. =================