1/20/18

"A GREAT STORM IS UPON MICHIGAN" THE GREAT BLIZZARD OF 1978 - 40TH ANNIVERSARY!

"A GREAT STORM IS UPON MICHIGAN"
THE GREAT BLIZZARD OF 1978!

Written by: William R. Deedler, Weather Historian - Southeast Lower MI

While I worked countless snowstorms in my long career with the NWS; the two most impressive storms were in order: the January 26-27, 1978 Blizzard and the December 1st, 1974 snowstorm. The huge snowstorm of December 1974 was quite noteworthy in the amount of snow that fell continuously for hours over the Detroit Metro area and portions of Southeast Lower Michigan.

A more powerful (in terms of intensity/extent) storm and remains of strong interest to all meteorologists who have studied winter storms in the Great Lakes is the January 1978 Blizzard. This storm is also of interest and remembrance to many longtime residents of the Great Lakes, the Upper Ohio Valley and Ontario, Canada who had to deal with winter's full fury late in January of 1978. In addition, the storm certainly casts many memories for those of us (author included) who were on duty and worked during the storm while observing in tremendous awe the development and subsequent immense strength of this great monster. With the 40th anniversary of this Great Blizzard at hand, it is worth taking a step back in time to re-live this monumental example of nature's fury.

While there are several contenders for the worst blizzard ever to hit the Great Lakes in relatively modern times (since 1870 when records began in Detroit), the immense and intense Blizzard of January 26-27th 1978 must rank at or near the top along with the Great White Hurricane of 1913 with its similar track and powerfulness.

The incredible Blizzard of January 26-27th, 1978 evolved out of a winter that was infamous for cold and storms. The Winter of 1977-78 thus far had been one the coldest, since records began, in many areas from the Rockies eastward to the Appalachians. Mammoth blizzards occurred late in January and early February from the Midwest to the East Coast as strong Arctic plunges dove south into the country and met up with the warmer winds from the deep south. The winter of 1977-78 was similar to its predecessor (1976-77) in terms of cold. The main difference between the two winters, however, came in February. In 1977, temperatures moderated rapidly during February, while in 1978, the cold actually worsened - with several locations reporting their coldest recorded February to date. The Winter of 1977-78 is written down in the record books as Detroit's seventh coldest winter, Flint's fifth coldest and Saginaw's sixth. West of the Rockies, it was a different story as a dominant upper ridge of high pressure provided a relatively mild winter, with some stations even reporting one of their warmest winters on record.

The Great Storm

Since there were some forecasting variances of the intensity and track of the storm, and considering the primitive model of the day (LFM - Limited Fine Mesh), forecasters did an admirable job in forecasting one of the most severe winter storms ever to hit the Great Lakes Region.

A Winter Storm Watch was posted as early as Tuesday night, the 24th, for the southern half of the Lower Peninsula for Wednesday Night into Thursday. Gale Warnings for the Great Lakes were hoisted the following Wednesday morning, along with the Watch. A weaker system had moved through the region earlier during the day on Tuesday and already dropped some snow on the region (a Winter Storm Watch had been issued for this system as well, earlier on Monday, the 23rd). After Tuesday's snow, the headline on the Special Weather Statement that was issued by the NWS Tuesday evening read as follows: "Another Winter Storm Threatens Lower Michigan" and thus, a second Winter Storm Watch was officially posted.

Meanwhile, the ingredients of what would later prove to be a truly fascinating yet vicious winter storm were coming together from different parts of the country. As with the "White Hurricane of 1913," the massive storm actually began as two smaller but distinct storms. A strong low pressure with an attending arctic air mass was entering the Northern Plains by way of Northern Minnesota on Tuesday evening (24th). At the same time, another developing low pressure system was taking shape over the eastern Texas/Louisiana area.

The phasing of two distinct jet streams aloft proved to be the key as to the subsequent strength and massive extent of the storm. A very strong and energetic Arctic impulse surged almost due south and plowed the Arctic front through the Northern Plains late on the 24th. At the same time, another very strong upper wind impulse surged south through southern Arizona. These two jet streaks made up the larger North American jet stream as a huge upper ridge of high pressure along the West Coast of the U.S. diverted the powerful Pacific Jet north into Northern Canada. This northern jet (containing a wind max of 110 knots) then dove due south, like on a giant roller coaster, across the western U.S. as the second, subtropical jet (with an even stronger wind max of 130 knots) surged across southwestern states. On Wednesday (25th), a deepening area of low pressure made its way east across the Gulf States into Georgia by evening (surface | 500mb). Meanwhile, across the north, the Arctic front barreled east across the Upper Midwest into the Western Lakes by Wednesday evening.

Earlier that Wednesday morning, the Winter Storm Watch for Southeast Lower Michigan was changed to a Heavy Snow Warning, while a Travelers Advisory was issued for Western and Northern Lower Peninsula. Later, at the issuance of the evening forecast, the entire Lower Peninsula was upgraded to a Heavy Snow Warning. Meanwhile, a rapid deepening of the surface low over the southeast portion of the country also commenced on Wednesday evening. As the low intensified over Alabama and Georgia, Atlanta registered its lowest barometric pressure ever late on the 25th. At the same time, further north in Michigan, snow was falling over much of the Lower Peninsula. In and around the Ann Arbor and Metro Detroit, the snow mixed with or changed to light rain Wednesday night as slightly warmer air surged northwest into that area ahead of the deepening storm.

While the storm was organizing in the lower levels of the atmosphere over Georgia, the Subtropical and Arctic jet aloft began to merge and phase over the Southeast part of the country. This merging of jet streaks contained a wind max of 150 knots which helped induce a rapid intensification of the Georgia Low as it surged northward into West Virginia early on the 26th. Record low barometric pressures were set all along its path as an ominous track (trough) began to materialize toward the Eastern Great Lakes.

Bands of heavier snow spread north into much of Southern Lower Michigan during the very early morning hours of the 26th. Rain continued to fall, however, over the extreme southeast corner of Lower Michigan. At 1 AM EST, rain was observed at Detroit Metro Airport with the temperature comfortably above freezing at 36 degrees. Further north at Flint, however, sleet and freezing rain were falling as the temperature hovered around freezing. Air pressure tendencies were noted as falling rapidly /PRESFR/ and continued that way for several hours (in fact, several stations in this storms path had to re-adjust their barographs for station pressures traces that were BELOW initial chart scale).

The aforementioned Arctic cold front that was across the Western Great Lakes advanced steadily east into Lower Michigan as the main southern low underwent explosive deepening (this low's central pressure fell 40 millibars in 24 hours)! The central pressure was recorded at 28.28 inches as it tracked north across eastern Ohio, just west of Cleveland, at 7AM EST. As the low moved out over Lake Erie, the Arctic cold front over Southeast Lower Michigan was pulled sharply east into it's mammoth cyclonic circulation. Any residual rain over Southeast Lower Michigan quickly changed to heavy snow and blowing snow during the pre-dawn hours of the 26th. As the Arctic front plowed through the Cleveland area, the wind gusted to an incredible 82 mph! As the Arctic air flooded the Cleveland area, the temperature dropped from a relatively balmy 44 degrees at 4AM EST to a bitterly cold 7 degrees by 1000 AM EST.

Blizzard Warnings were hoisted across much of the Great Lakes and Upper Ohio Valley Region by daybreak Thursday. The center of the huge storm (surface | 500mb) continued to trek north northwest across Southwest Ontario (roughly between Chatham and London) while Detroit measured its lowest pressure reading at 28.34 inches at 650 AM EST. The incredibly deep center made its way north along the St. Clair River with Sarnia ON reporting the lowest pressure on land at 28.21 inches. Not only was the depth of this mammoth storm's center very impressive, so too was the extent of low pressure from its center. Even locations that were far removed from the storm's center also reported record low pressures. Stations such as Cincinnati OH, Rochester NY and Toronto ON and even as far east as Wilmington N.C., all recorded record low pressure readings from this monster. In fact, at Toronto, where records go back as far back as 1840, the lowest pressure reading of 28.40 inches broke the old record of 28.57 inches by 0.17 inches. In addition, dozens of other cities, with records going back a century, also recorded their lowest pressure reading of all time or, for at least the month of January. This massively intense storm was responsible for strong wind gusts as far away from the center as Boston /72 MPH/ and Chesapeake Bay Bridge /90 MPH/ with even damaging winds reported as far south as Tallahassee FL.

As the Arctic air circulated throughout the storm while it made its way over Lake Huron, the lowest pressure was reached around 950 millibars or a hurricane-like 28.05 inches! "A Great Storm is Upon Michigan" read the headline of the 800 AM EST Special Weather Statement issued by the National Weather Service Forecast Office in Ann Arbor that Thursday /26th/ morning. Heavy snow and blizzard conditions were extensive as wind gusts in excess of 35 mph whipped the snow into huge drifts across much of Southeast Lower Michigan. Other areas of Eastern Michigan, Indiana and Ohio reported near hurricane-force winds, heavy snow and temperatures hovering between zero and 10 above, resulting in extreme blizzard conditions. These conditions later expanded further east into Pennsylvania and West Virginia and prevailed into the night (26-27th) across much of the Eastern Great Lakes, Southern Ontario and the Upper Ohio Valley. With the storm generating copious amounts of snow and very strong winds, whiteout conditions were widespread. All land and air traffic came to a stand still in the affected regions. Several major roads were closed for at least two to three days, if not longer, while clean up got underway. Numerous NWS employees were stranded at work, home, or on the road somewhere between the two. Several employees worked double shifts into at least Friday (some longer) because of the impassable roads with others simply unable to get to work.

The Blizzard Warnings were allowed to die across Michigan during the forenoon hours of Friday, the 27th. Record 24 hour snowfall totals from the storm included, 16.1 inches at Grand Rapids, 15.4 inches at Houghton Lake and 12.2 at Dayton, OH. Snowfalls for the entire storm (25-27th) included a whopping 30.0 inches at Muskegon (some of which was Lake Michigan enhanced), 19.3 inches at Lansing and 19.2 at Grand Rapids. Snowfalls were less over Southeast Lower Michigan (mainly because of the rain that fell for a period) and included 9.9 inches at Flint and 8.2 inches at Detroit.

The following is a quote from the summary written about the storm by Meteorologist in Charge, C.R. Snider on January 30th, 1978 at the National Weather Service Ann Arbor:
"The most extensive and very nearly the most severe blizzard in Michigan history raged throughout Thursday January 26, 1978 and into part of Friday January 27. About 20 people died as a direct or indirect result of the storm, most due to heart attacks or traffic accidents. At least one person died of exposure in a stranded automobile. Many were hospitalized for exposure, mostly from homes that lost power and heat. About 100,000 cars were abandoned on Michigan highways, most of them in the southeast part of the state."
The employees of the National Weather Service Forecast Office in Ann Arbor had just set up shop at the new quarters at the Ann Arbor Federal Building a WEEK before the storm hit. The forecast staff had transferred from the Detroit Metropolitan Airport Office while the observing and radar staff remained at the airport. The majority of employees still lived in and around the metro Detroit area and all major roads between Detroit and Ann Arbor were blocked for approximately 18 hours due to the storm. Several employees put forth efforts beyond the call of duty, stated Mr. Snider in his storm report.

Yet, as mentioned earlier, the Winter was not yet over by any means as the month of February (after the storm) was brutally cold across much of country. The below normal temperature departures of February 1978 were strikingly similar to that of January 1978 (and in some places, February was actually colder). The average temperature for Detroit that winter came in at just 20.5 degrees /normal 27.1/ which again, made it the seventh coldest winter on record. Snowfall totaled a hefty 61.7 inches which made the winter of 1977-78 the ninth snowiest winter on record at Detroit. Flint's average temperature of 19.1 degrees made it the fifth coldest winter on record, but Flint received less snow than Detroit with 50.6 inches. Saginaw's winter average temperature of 17.9 degrees made it the sixth coldest winter on record and was accompanied by 55.6 inches of snow. The 1977-78 snow season at both Flint and Saginaw has since dropped off the top 20 snowiest winters list.

More on the storm in surrounding areas can be found here.

https://www.ncdc.noaa.gov/news/using-regional-snowfall-index-evaluate-great-blizzard-1978
http://www.mlive.com/news/grand-rapids/index.ssf/2017/01/surreal_snow_scenes_from_the_b.html

https://www.youtube.com/watch?v=YuAMWRtT4NY
http://nbc4i.com/2017/01/25/remembering-the-blizzard-of-1978/


Making weather fun while we all learn,
Bill Deedler -SEMI_WeatherHistorian




1/14/18

A Brief Comparison of Model Performance for Storm Projected Sunday Jan 21st 2018

After my last post on model projections and some wild posts on the internet; I thought it would be interesting to post three model solutions - the /American/GFS, European/ECMWF and the Canadian/GEM each day of the 00Z model projections for the same time period; 00z 01/22/18  (Jan 21st, 2018 Sunday 7PM). Obviously then; model projections will likely change (less hopefully as the time period grows closer) at each successive 00z model run as I post the 500MB position along with the surface for 00z 01/22/18.

All models are projecting some storm with a variety of solutions for the exact time period. Again; the main concern here will be the map projections of the 500 MB jet and associated storm at each day interval for the next week for the time period; 00z 01/22/18. The maps will be from the 00z map projections of all three models with brief comments will be made throughout.


DAY 1; 8 days out (00z - 01/14/18 model run; Sat Jan 13 7PM)

As you note; the postings will be GFS, ECMWF & CMC

1-The GFS has the main low over Northern Indiana, The Euro over Arkansas while the Canadian has placed it over extreme Northern Lower Michigan.

 DAY 2; 7 days out (00z - 01/15/18 model run; Sun Jan 14 7PM)

General trend is slower (this continues on today's 12z also) with storm while high to the NE becoming stronger & more prevalent on most model trends. The Euro has joined the general trend with main low now further north while the GEM is the fastest with the low up in the NW Lakes.
DAY 3; 6 days out (00z - 01/16/18 model run; Mon Jan 15 7PM)

The GFS has basically held the low over E-SE Iowa while weakening it from 993 to 998 MB as the Euro actually backtracked the low to SE Nebraska/NE Kansas and also weakened the low from 990 to 996 MB. GEM has also backtracked the most dropping the low from Northern Wisconsin down to NE Missouri as pressure weakened from 995 to 1002 MB. In actual fairness, it actually elongated the low from Wisconsin down into Missouri with the center in NE Missouri and is similar in position to the GFS.
 DAY4; 5 days out (00z - 01/16/18 model run; Tue Jan 16 7PM)

Interesting how the models subtly shift the low, sometimes similar to the former time-period of another model the day before. All focus has come around to the from the Central Plains/Western Midwest. While the GFS focuses on the Iowa/Missouri area, the Euro remains the furthest sw over E Kansas and the GEM is furthest NW in NW Iowa/NE Nebraska. The GFS hangs on to a 998 MB low intensity while the Euro weakens the low from 996 MB to 1001 MB and the GEM reverses that: deepens the low back to 997 from 1002 MB.

DAY 5; 4 days out (00z - 01/16/18 model run; Wed Jan 16 7PM) and GIF for the the 5 days

Since all models have converged on a Central Plains/West Midwest position and not affect SE Michigan with warm air advection snow; I've decided to end the comparison project with a GIF for each model for the 5 Days running up until 00z Thu, Wed Jan 16, 7PM.


It's very interesting to note the GFS/American model appears to have the least variance for position of the low for the same forecast hour (22/00z Jan 21, 7PM Sun) especially after Day-1. The Euro/ECMWF comes in second while the GEM was all over the place and most variant. However; it must be mentioned that the GIF cycle of the GFS contains all runs (00z,06z,12z,18z) and not just the 00z, while the Euro and Gem contain the 00z and 12z. Therefore, the GFS has the most GIF positions which can be confusing. Comparing just the 00z daily runs of all three models is the best and fairest way to look at it....which was done above for 5 days. Of course, the final analysis won't be available until the Mon, 00z when the initial run for Jan 22, 2018 is done and posted.







1/10/18

Winter Storm Brewing - But for Who - Now THAT is the question?

Latest plausible indications continue to push a strong cold front through Southeast Lower Michigan slowly overnight Thursday into Friday morning. Rain with become mixed with/and changes to sleet and then over to snow as the atmosphere column thickens with the colder, snow supporting air. At this time; the front will slide east and southeast as a developing low pressure and storm rides north northeast along the front pushing heavy snow and mixed precipitation through the Ohio Valley and on into the Northeast. In the colder air behind the front over Southeast Lower Michigan; up to a couple of inches are possible on Friday.

Beside the upper air support for the low pressure and storm, it must be noted the High pressure pushing southeast into the northern plains will exude a strong push of Arctic air, once again, southeast into the country. This too will aid in pushing the storm out to the east-northeast. I drew a dotted line as of this writing the snow still could push along that "buffer-zone" if the storm develops faster, high weaker or upper air isn't sampled well enough but the above scenario seems most likely.

Now here's the caveat: model sampling of the atmosphere from upper air, satellite and radar is still not complete as the energy and subsequent phasing is still ongoing. Yes, things can change but again at this time the above seems the most plausible. If dramatic changes evolve from better model sampling and data, I'll surely update.


Finally:
Let me just say; again with some models predicting an Armageddon snowstorm for the Ohio Valley and at times extending into the Southern Lakes, I've stayed out of it with nothing written on it. The crap put on various web-sites pumping an almost certainty fictitious amount of snow for some areas was ridiculous! Those posting 2 or even near 3 foot snowfalls across portions of Southern Michigan into the northern Ohio Valley; especially when many records for 100+ years would be shattered is complete disregard to the general public. I'm not talking about various weather forums that are a mix of all interests...these extreme model solutions hopefully stay with those members, who generally know the "problems" with model output.

Every time there is a hint of a "big" storm, several of these snowstorm enthusiast (spelled N.U.T) start posting the most extreme solutions. Several have questionable educational backgrounds - if any in meteorology - and never even worked in the weather field. And, this happens EVERY winter with the advent of the internet. I can't tell you how many times before and after I retired from the NWS, I was asked: "I heard we are getting this "huge amount of snow" is it true, what do you guys say - and it really has blossomed the past 10 years.

Model forecasting has improved phenomenally in the past 20 years, especially the past 10 but they still get it wrong as all forecasters do. Sometimes during a season, one or two models will have the inside track for accurate forecasting that particular winter - and days out - only to lose it off at times.

I've certainly gotten it wrong on individual events and my Outlooks, nobody is infallible but I try to be reasonable and write various articles and blogs the most objective and limit, to use the in-vogue vernacular -  "Fake News".  Many times when something I forecast is wrong or out of bounds with what happen, I do talk about what went wrong and I wish more would do that because I try to...

Make weather fun while we ALL learn,
Bill Deedler - SEMI_WeatherHistorian

1/9/18

Early Winter Unfolding Along With Prevalent Analogue Pattern Seen

Officially, we've only been into winter a little better than a month and the cold season /Nov -Mar/ two months. There still is, by historical terms, plenty of winter left - and looking keeping on eye on the hemispheric meteorological patterns and the other on the Winter 2017-18 Outlook issued back in November, can attest to that. After a brutal introduction to 2018 with a continuation of bitter cold temperatures and wind chills; we managed to squeak out a record number of days with temperatures below 20 - twelve days at Detroit and Flint. The old record for Detroit was eleven and was set twice back in 1936 /Jan 23-Feb 2/ and 1979 /Feb 8-Jan 18/. The below 20 streak was tied at Flint with 12 which tied with Feb 8-19 1979. Saginaw did not hold below 20 the entire streak, hitting 21 on the 2nd and thus messed up the record there. More on the cold streak can be found here.

The dominant analogue winters for December were a good predictor for this December in all temperature projections and two out of three snowfalls calling for cold and snowy. Saginaw;'s snowfall came up lower than the higher December snowfall winters but still appreciably above the average.

What is most striking is how close the temperature averages from the analogues for December came  to the actual December 2017 numbers, at all three sites using the colder winter averages added to all the winter averages and dividing by two. I used both numbers; the average of all the analogue Decembers and add that to DOMINANT trend (below normal or coldest Decembers) to get a fair projection for the actual month (see below) which is reflective of the Winter Outlook.

Detroit's temperature analogue data contained a whopping 10 out of 14 below normal Decembers. The average for those 10 winter's came in at 25.6. Even the December temperature average for all 14 winter's came in notably below normal at 27.9 /norm 30.1/. Adding the averages of the coldest December winters and then all the Decembers gives an average of  the two /25.6 + 27.9/ of 26.7. The actual temperature average for December 2017 ironically fell between the two average data projections with the actual 26.6 - just one tenth lower than the average of the two. I'd say the December projection for temperatures was very telling and fits nicely into the other dominant 10 winters (71% of the analogue sample)!

Snowfall projections for December at Detroit were also telling with the dominant analogue December snowfalls having an average of the 7 winters at 18.7". The extremes for December snowfall due to the storm track establishing itself was problematic in the outlook from the start with the range being from 4.3" to 25.1"! However, the pattern of dominant snowfalls (7 out of 14; 50% and 1 December being in the normal range and 6 below normal) did revel itself in the study as likely a snowy December. Actually out of the 7 snowy Decembers, there were 4 where that snowfall averaged around 20" or more...a fairly strong showing in itself and right up there with the 22.5" recorded this past December. That 4 out 8 very snowy winters showed that when there was a snowy December; there was a 50% chance that it would be at least more than double the normal amount /9.7/.

All in all; stepping back to all the dominant; prevailing Decembers in the study for Detroit showed an average temperature of 25.6 degrees with 18.7" of snow...and this past December verified the prevailing pattern.


                                 Detroit Analogue Breakdown For December
  
 

Likewise; Flint's average for the all the sample analogue December temperatures came in at 25.4; 1.6 degree below the 27.4 normal. Using the dominant 7 (out of 12) coldest Decembers at Flint gave a projection of 21.6; this compares to Flint's actual December 2017 average of 22.6. Then again; adding both the 25.4 for all of Decembers analogues to the coldest December analogues at 21.6; gives an average of 23.5 Flint's actual temperature came in between the two at 22.6.

Flint's actual heavy snowfall in December followed right along with the heaviest snowfalls in the study. Using the snowiest but not dominant snowfall trends in the analogues finds: the largest snowfalls projected 28.5" out of three very snowy Decembers and well above the normal of 11.8". Flints projected snowfalls in the analogues were much more variable than even Detroit's in December; partly due to the variable placement of the storm track that sets up. It was noted on the Winter Outlook analogue chart that Flint's snowfall for 1910 -11 was highly suspect since both Detroit and Saginaw measured considerable higher amounts (20.1" and 17.4" respectively) as compared to Flints 11.5" so it was not used. Flint's December 2017 snowfall totaled at 23.5" more than double the norm /11.8"/ and almost double the analogue average at 12.8". There is a huge gap between the analogue average and analogue snowiest Decembers in Flint; ranging from 12.8" to 28.5". Looking at the average between the two (12.8" to 28.5") gives 20.6" and not far off the actual total. Even wider were the analogue December snowfalls all together: from 4.4" to 35.3"! Predicting the December 2017 snowfall for Flint was difficult to say the least but still landed in the heaviest snowfall categories like Detroit.

Up at Saginaw gives the following results; the average December temperature for the analogues came in at 25.5; while the average for the below normal or coldest Decembers came in at 21.6. Using the average of the two numbers leads us to the exact number Saginaw recorded in December 2017 - 23.5! Snowfall at Saginaw was the only lacker in the group. The actual snowfall for December did come in above normal with 13.2" /norm 9.0"/ and close to all the December analogue snowfall average at 14.7". But the snowiest Decembers average number was quite a bit higher at 22.1" This was highly skewed by two Decembers with around 40" not unlike Flint.


The overall precipitation thus far this winter has been lacking however with about half of the normal (1.49", -1.33") falling through January 6th. In Detroit's study there were 3 drier than normal winters, 3 wetter than normal and a dominant 8 normal precipitation winters. If we are to have a normal precipitation winter in alignment with the prevailing winters in the study, we're gonna have to step up the precipitation - and recent models indicate as much. While on the subject of future weather; latest indications are that January will make up some of that below normal departure with more moderate to even milder weather at times between long cold spells. This short thaw ongoing is more than welcome for most after temperatures averaged 20 degrees below normal at around 6 at Detroit through the 6th. The first week or so is reflected in the departure map below...

One of the other notable pattern trends of the analogues was the more than average roller-coaster temperature pattern, reflecting the amplified and variable jet. This pattern occasionally locked into the prevailing pattern - cold - but with relief from the cold breaks throughout the winter and early spring. Thus far; there are several analogue years being tagged as similar in our winter pattern thus far with a few very similar.

Next Up: Winter Storm Brewing - But for Who?

After Mid Month - Part Two: A Mid Winter Update - What Do the Best Performing Analogue Winters Thus Far Suggest About the Rest of the Winter and Cold Season?

Making weather fun while we all learn,
Bill Deedler - SEMI_WeatherHistorian