Flashback: October 2006 and other early-season lake effect snowfalls

By: sullivanweather , 11:09 PM GMT on October 03, 2009

Share this Blog

Over the last several days global forecast models have been insisting on building an anomalously cold airmass over central Canada and dropping it into the eastern US next week following the passage of a significant storm system. Several model runs have shown 850mb temperatures dropping into the negative 10's Celsius along the US-Canadian border with -6°C air @850mb moving across the Great Lakes. An airmass this cold has shown in years past to produce early-season lake-effect snowfall. Other model runs have shown a more modified solution, with air still cold enough for wet snow to mix in with lake-effect rain, but nothing substantial in terms of accumulating snow. Models have centered on the 12-13th as the coldest days of the upcoming 'arctic' blast, which is a popular date for early-season snowfall, it seems.

Three years ago today numerical models were giving meteorologists fits. First to appear on the always questionable-in-the-long-term GFS model near the end of September was an anomalously deep and cold trough diving into central Canada and across the Great Lakes delivering a winter-like blast of arctic air capable of producing a lake effect snow event. Initially shrugged off as a couple of anomalous runs, support slowly gathered. Once within the forecast range of the operational Global GEM (Canadian) model and the operational ECMWF (European) model the same deep and cold trough was there, daring climatology. 850mb temps were progged to be nearly 5-sigma below normal as the core of this cold airmass descended into the US but would this actually happen?

As the event grew near and it became apparent that there would be an outbreak of arctic air new questions arose. It had just been the second warmest summer on record, nearly beating out the Dust Bowl year of 1936, and despite a cooler than average September across the Great Lakes region, water temperature anomalies in the Great Lakes ranged 5-8°F above normal, running at all-time highs for early October. Could such anomalously warm lake temperatures yield any lake effect snow or would they moderate the airmass enough to produce lake effect rain?

The front moved through the western lakes on the 11th with the lake effect beginning quickly in its wake. First, rain showers moved off Lake Superior but those slowly changed over to snow showers by the evening. As the night progressed snowfall picked up in intensity off Lake Superior as even colder air flowed over the lake in a lightly sheared surface-700mb environment with several communities in favored areas picking up 6-8 inches of snow by morning. However, lake-effect snow off Lake Superior, the coldest of the five Great Lakes, isn’t all too uncommon in mid-October. For the communities downwind of Lakes Michigan, Erie and Ontario, mid-October lake-effect snow isn’t just uncommon, it’s rare. And climatology was hard to overcome at first. Through the early morning of the 12th most of the lake-effect precipitation moving off Lake Michigan was in the form of rain, save a few localized areas in northern-lower Michigan. But the airmass had other plans. Temperatures aloft continued to cool and by mid-morning, the lake-effect rain had changed over to snow in most areas across the State of Michigan. Even communities close to the lake shore started to see a mixing of precipitation types which eventually changed over to snow.

So the airmass was, in fact, cold enough to deliver the goods in terms of snowfall. Not only was there snow falling but there was snow accumulating. Widespread accumulations of 2-4 inches across southern Michigan were common and in some localized areas, such as Cheboygan, Michigan, near the Straits of Mackinac, up to a foot of snow fell! But would this cold air have enough drive to complete its journey across the Great Lakes to deliver snow to the eastern Great Lakes region, where water temperatures were warmer still? The average temperature of Lake Erie on the morning of the 12th was 65°F, about 5 degrees above normal. And due to the position of the low to the northwest of Lake Erie, the wind flow would have to come down the entire length of the lake, allowing plenty of time for the air to moderate as opposed to the flow over Lake Michigan, which cut widthwise across the lake, leaving less time for a moderation of temperatures.

At first, it seemed the warmer waters of Lake Erie would win out. The front crossed the Niagara Frontier mid-morning of the 12th and several hours after its passage only lake-effect rain fell. But nature always has tricks up her sleeve. The airmass coming over Lake Erie was anomalously cold, but yet not yet cold enough to produce snowfall after moving down the length of warm Lake Erie through early afternoon. Models had projected 850mb temperatures to drop to a marginal -4 to -6°C and upon reaching those temperatures rain continued; some other factors were needed to conjure up a snowstorm. They would soon be realized.

Models are a great tool for guidance but they don’t hold a candle to real-world observation. 1800Z soundings from around the Great Lakes showed an underestimation of how truly cold this airmass was. 850mb temps were a couple degrees Celsius cooler than models had projected them to be. Soon, areas of western New York away from the lake began to mix with graupel and change over to all snow by 2pm. Buffalo airport began to change over to snow around 3pm and snow eventually started falling right to the shores of Lake Erie. A very odd confluence of events started taking shape.

Special upper air soundings from Buffalo showed a dramatically cooling 3,000-6,000’ layer indicating a dynamically cooling atmosphere typical of ‘top down’ snow events. Additionally, inversion heights were off the scale, over 20,000’ and nearly double all previous known events. By 3pm the National Lightning Detection Network (NLDN) noted several cloud to ground lightning strikes within the developing lake band, indicative of frozen precipitation. Due to the unidirectional surface-700mb wind flow, extreme lake-induced instability and incredible inversion heights, as the band developed and intensified the extreme upward vertical motion over the lake required an equally strong inflow of air in the low-levels of the atmosphere. This was the wild card. The air at low-levels was abnormally dry, which aided in evaporative cooling of the column and likely the final factor leading to a complete changeover to snow, despite a lake temperature more than 30°F above freezing!

The snowfall in western New York came in two stages. The first stage resulted in a very heavy wet snow that only accumulated 2-6” during the afternoon and early evening hours. In most locales this initial snowfall melted on contact with the warm ground early on, then later in the afternoon started to coat trees and elevated surfaces. It wasn’t until the sun set that the wet snow began to add up and due to its high moisture content, just two inches was enough to start tree damage. What would follow during the overnight hours was truly one of the more remarkable weather events the Buffalo-metro area would ever see.

As night fell and the atmosphere cooled further, the lake-effect band intensified. Snow rates picked up and snow:water ratios rose from 6:1 to 12:1. For about 12 hours the band remained stationary, producing heavy lake-effect thundersnow over the Buffalo-metro area and points towards the northeast extending to the southwest suburbs of Rochester. During the time 12-18 additional inches of snow fall, resulting in a crippling snowstorm. Most trees were still in full leaf and simply couldn’t handle the weight of 1-2 feet of snow. Tens of thousands of trees fell during this storm, knocking out power to over 400,000 residents. The tree damage was exacerbated by abnormally wet weather during the month of September, amounting to 6.95” at the Buffalo airport. Some locations remained without power for up to two weeks due to the widespread damage inflicted by the heavy wet snow. The Buffalo Airport reported a 24-hour snowfall total of 22.6”, which would end up being the city’s 6th largest 24-hour snowfall total, no matter what time of year.

This event will go down in the record books as the earliest lake-effect snowfall of such intensity on record. Looking back through history no other event on record produces as much snow as early as this event. There have been instances of snow falling earlier in the season but those events don’t compare in regard to the amount of snow that actually fell. There is one event where snowfall amounts were similar, but this occurred a week later in the year, on October 19-20th, 1930.

Other notable early-season lake-effect snow events are as follows:

September 29th, 1895 – An unusually strong low pressure moving from Montana to the Great Lakes region tapped into an unseasonably cold airmass over the Canadian Prairies. This cold air was then transported over the lakes resulting in an early season snowfall for much of the region. Buffalo saw up to 6” of snow with snow being reported as far south as Philadelphia! The arrival of this system was anticipated up to 24 hours in advance by telegraph from the Chicago Weather Bureau with a forecast for gales on the Great Lakes, which ended up being a fairly accurate forecast. What took folks by surprise was the snow that accompanied the high winds.

October 10th, 1906 – Once again, a very strong cold front connected to a gale center over Ontario swept over the Great Lakes and Northeast, delivering an unseasonably cold airmass on strong west winds. Snowfall from this storm piled up over a foot in the southern suburbs of Buffalo, causing similar damage to that of the October 12-13, 2006 event. However, the bulk of the snowfall remained south of the city, unlike the 2006 event where the entire metro area was pummeled. Snow was reported as far south as eastern Kentucky, resulting in the earliest snowfall on record for several weather bureaus located in the state. Additionally, killing frosts covered a wide area of the Ohio and Tennessee Valleys.

October 12-13th, 1909 – An early season snowstorm moved from the Northern Plains to the Upper Midwest, depositing 10-22 inches of heavy wet snow to the northwest of the low track across Minnesota, northern Wisconsin and Upper Michigan. As the low pressure pulled into Canada cold west to northwest winds behind the low pressure activated an early-season lake-effect snow event. Generally 2-4” of snow fell downwind of the lakes with several areas picking up as much as a foot of snow. Buffalo, NY recorded 6” of snow by the time it tapered off and killing frosts covered much of the Northeast in the days following.

October 18-19th, 1930 – Similar to the 1906 and 2006 events, one to two feet of snow fell on the Buffalo-metro area downing trees and power lines. Now in the age of the automobile and fast-paced commerce, the snowstorm had devastating effects on the local infrastructure and transport of goods. Many vehicles were stranded by the heavy wet snow, which made roads impassable.

References and further reading.

Bigelow, F.H., 1895: AREAS OF HIGH AND LOW PRESSURE. Mon. Wea. Rev., 23, 325–329. - http://docs.lib.noaa.gov/rescue/mwr/023/mwr-023-09-0325d.pdf

Hamilton, Robert S., Zaff, David and Niziol, Thomas, 2006: “A catastrophic lake-effect snowstorm over Buffalo, NY October 12-13, 2006” – http://ams.confex.com/ams/pdfpapers/124750.pdf

GARRIOTT, E.B., 1906: FORECASTS AND WARNINGS. Mon. Wea. Rev., 34, 478–481. - http://docs.lib.noaa.gov/rescue/mwr/034/mwr-034-10-0478.pdf

COX, H.J., 1909: DISTRICT No. 4, LAKE REGION. Mon. Wea. Rev., 37, 734–741. - http://docs.lib.noaa.gov/rescue/mwr/037/mwr-037-10-0734.pdf

BENNETT, M.C., 1930: THE WEATHER ELEMENTS. Mon. Wea. Rev., 58, 427–428. - http://docs.lib.noaa.gov/rescue/mwr/058/mwr-058-10-0427.pdf

Current watches, warnings and advisories.

Eastern US current watches/warnings
Fig.1 - Current watches, warning and advisories issued by the National Weather Service. Courtesy of NOAA.


Forecast Discussion

Coming later...

Radar: Northeast Region Loop

NE radar
Fig.2 - Radar loop of the Northeast region. Courtesy of Weather Underground.


Local SST's

Northeast SST's
Fig.3 Sea-surface temperatures off the Northeast Coast. Courtesy of NOAA.


All hits.
hit-counter-download.com .

Unique hits.
hit-counter-download.com .

Adopt one today!
My weather dragon!

Stone bridge (sullivanweather)
Stone bridge about 5 miles from my house.
Stone bridge
Stone bridge (2) (sullivanweather)
Stone bridge about 5 miles from my house.
Stone bridge (2)
Fall foliage '09 (sullivanweather)
Fall foliage 2009.
Fall foliage '09
Scablands (sullivanweather)
Behind me there's a dam. In front of me are massive boulders carried down into a ravine over several 20-30' cliffs. Must have been a sight when the floodgates are open.
Floodgates (sullivanweather)
There they are!
September Sunset (sullivanweather)
Sunset on 9/10/09
September Sunset

Reader Comments

Comments will take a few seconds to appear.

Post Your Comments

Please sign in to post comments.

Sign In or Register Sign In or Register

Not only will you be able to leave comments on this blog, but you'll also have the ability to upload and share your photos in our Wunder Photos section.

Display: 0, 50, 100, 200 Sort: Newest First - Order Posted

Viewing: 23 - 1

Page: 1 — Blog Index

23. weathergeek5
1:39 AM GMT on October 05, 2009
Here in Delaware we got close to 30 inches from that storm in 1996.
Member Since: December 25, 2005 Posts: 0 Comments: 1744
22. TheDawnAwakening
1:34 AM GMT on October 05, 2009
Oh wow. I got 10-20" from that Blizzard. A great storm indeed. The March 1914 Blizzard is the deepest low pressure system I believe for a Northeast Snowstorm with 952mb.
Member Since: October 21, 2008 Posts: 248 Comments: 3970
20. TheDawnAwakening
1:07 AM GMT on October 05, 2009
Quoting StSimonsIslandGAGuy:
Doubt something like this will happen--earliest ever snow seen in Manhattan, October 9, 1804

The great "snow hurricane" of 1804 Link

I've seen thundersnow twice--once in Atlanta when about 2" fell as the rain changed to snow just after midnight on Jan 7 as the great blizzard began to take shape.

And the other time was during the December 23 1989 snowstorm here, we had thunder between 2:30 and 4 a.m.

What Blizzard began to take shape?
Member Since: October 21, 2008 Posts: 248 Comments: 3970
18. TheDawnAwakening
9:41 PM GMT on October 04, 2009
Quoting sullivanweather:

I did see that. Combination tropical system in the Gulf and blue norther from Canada all merging on the East Coast. Yikes! LOL

Do you think it's possible? Well it is possible given that we had stronger storms then the Superstorm of 93 in terms of pressure.
Member Since: October 21, 2008 Posts: 248 Comments: 3970
17. upweatherdog
6:52 PM GMT on October 04, 2009
Quoting Blizzard92:
sullivanweather- Did you get to see the 0z GGEM? Lol, it is up to its usually antics with some wild coastal storm it cooked up dropping the H85 0C down to Florida towards the very end of the run. Ha, that model is always cooking up something.

It seems the models want to bring some kind of tropical disturbace into the trough, and phase it with the northern stream shortwave energy. I do not know where the models are getting the idea of a tropical disturbance developing.
Member Since: October 14, 2007 Posts: 173 Comments: 1372
16. sullivanweather
6:19 PM GMT on October 04, 2009

I did see that. Combination tropical system in the Gulf and blue norther from Canada all merging on the East Coast. Yikes! LOL

Member Since: March 8, 2007 Posts: 273 Comments: 12612
15. Zachary Labe
2:14 PM GMT on October 04, 2009
sullivanweather- Did you get to see the 0z GGEM? Lol, it is up to its usually antics with some wild coastal storm it cooked up dropping the H85 0C down to Florida towards the very end of the run. Ha, that model is always cooking up something.
Member Since: December 14, 2007 Posts: 286 Comments: 15158
14. TheDawnAwakening
1:44 PM GMT on October 04, 2009
If I didn't mention the date it was February 12th, 2006. The Blizzard developed a mean eye like feature and a dry slot also developed over my region which cut the snowfall down to 8" tops. I don't remember if it was windy or not.
Member Since: October 21, 2008 Posts: 248 Comments: 3970
13. originalLT
5:12 AM GMT on October 04, 2009
Nice post Sulli, great detail about all those early lake effect snows. In response to TDA's post #8, those of us in Stamford CT. missed out on the huge totals that NYC got and the totals that Fairfield CT. received. On that day I did several measurements around my yard and neighborhood, and could find no total higher than 18".The mesoscale band that affected NYC. must have arced northeastward and just missed our area, but covered Fairfield Ct.When I heard the totals on the news I could not believe the differences in snow totals in such a relativily small area. Needless to say I felt cheated!
Member Since: January 31, 2009 Posts: 0 Comments: 8011
11. LakeShadow
4:09 AM GMT on October 04, 2009
oh, and we lost over 40,000 trees... I still have topless trees in my yard.
Member Since: August 10, 2007 Posts: 3 Comments: 2134
10. LakeShadow
4:06 AM GMT on October 04, 2009
Great post there, sulli! That was one INCREDIBLE storm...The coolest part about that snowstorm were the flashes of pink lightning and the green flashes from blowing transformers...the sounds of cracking branches..and then walking around after the sun came up the next day was just incredible. I saw a street light pole snapped right in half and the top half was dangling upside down... the trees were all fallen into the roads, no vehicle could pass anywhere. We hunkered in, spent some time with the neighbors and listened to the Sabres, drank and played cards by flashlight...
Whatta storm... there will never be another like it. And how badly did the mets blow the forecast that night!!! It was Friday, Oct. 13..They called it the October Surprise Storm (in homage to the cluelessness of the local mets) ..turns out it was a named storm... Lake Storm Aphid. (that year storms were named after insects..I didn't even know they named lake storms...)
Member Since: August 10, 2007 Posts: 3 Comments: 2134
9. sullivanweather
1:22 AM GMT on October 04, 2009
To further illustrate the similarities between the October 2006 LES event and the upcoming pattern at the beginning of week two check out these 500mb charts.

This is the 500mb analysis @8pm on 10/12/2006.

This is the forecast 500mb chart given by the 1200Z run of the operational ECMWF for 8am on 10/12/2009.

Member Since: March 8, 2007 Posts: 273 Comments: 12612
8. TheDawnAwakening
1:17 AM GMT on October 04, 2009
My Dad said that we had thundersnow in the Blizzard of 2005 in the early morning hours of Sunday, January 23rd, 2005. This storm was a classic Northeast Coastal Snowstorm Intensity Scale study. Given that we had a +PNA, +NAO and a +4C GSO anomaly. This storm exhibited a tremendous pressure drop of 42.7mb/24hr period exhibiting roughly 1.82mb/hr drop for 24 hours. This storm moved over this GSO +4C anomaly indicating that these water temps aided in potential rapid intensification of the Blizzard's surface low. THe primary low developed a newly forming surface low pressure where the old low fully filled in and weakened while the new center developed and moved eastward passing over the SE of Nantucket, MA buoy 44008 having the pressure bottom out at 980.1mb. This storm had the +PNA which would indicate a ridge in the west followed by a large negatively titled 500mb chart trough. The +NAO indicated the progressive nature of the low pressure system which the Blizzard had lasting roughly 30 hours for Cape Cod, MA with snowfall rates roughly 1" an hour for 30 hours, although this average may be right, the snow was much more intense in spurts and for a few hours at a time followed by less intense snowfall rates in other periods. THis was especially true earlier on in the storm's duration, however this was not true for the hours of Sunday from 12am to 9pm Sunday where Blizzard conditions lasted for most of the day inhibiting travel for the area. Mesoscale banding existed with some locations seeing roughly 35" like my frontyard, compared to a few towns westward receiving just 20". A large mesoscale band existed in the Plyouth County, MA region where snowfall rates reached an incredible 8" in 75 minutes. That Sunday morning I had to shovel the walk way in the frontyard with already 12" of snow on the walk. I shoveled and in the time it took me to shovel the walk, probably half an hour, there was a new coat of snow roughly four inches deep already where I first started. This same morning I saw that the Blizzard's surface low developed a clear eye like feature which was pretty exciting considering this was the most snow I saw in a single snowstorm. The next winter another monster snowstorm hit the NYC metropolitan area with 26" of snow falling in the less than 24 hour period. The famous intense snowstorm developed a very rare single tremendous mesoscale band which dumped the excess amount of snow the city plows were unable to keep up with. This large band produced snowfall near 32" in Fairfield, CT, in a single region of SW CT and the NYC metro region with the 25 - 32" of snowfall when region wide less snow fell on average near 10". 8" had only fallen on Cape Cod, MA where there was roughly at least 12" forecasted. The Blizzard of 2005 while carrying similar single band characterisitics, the snow continued to fall to the east and west of the band. The band did not develop singular characteristics like we saw with the February 12th, 2006 snowstorm. These two storms will be tremendous case studies for this new scale I will try to create to have a better understanding with the public and a way to communicate with them.
Member Since: October 21, 2008 Posts: 248 Comments: 3970
1:08 AM GMT on October 04, 2009
My only thundersnow (that I remember) was Groundhog Day 2003
Member Since: January 17, 2006 Posts: 18 Comments: 3382
6. sullivanweather
12:31 AM GMT on October 04, 2009
Quoting StSimonsIslandGAGuy:
Great writeup of the Arborgeddon storm!


I've only seen thundersnow on 4 occasions. Once in the 1993 Superstorm. Once in the December 30th, 2000 'Turn of Millennium Storm' in 2007 while living in Bethel and last year in December's snowstorm.

I couldn't imagine 12 straight hours of thundersnow! That's something else. I couldn't even imagine 12 straight hours of good ol' regular thunderstorms...lol
Member Since: March 8, 2007 Posts: 273 Comments: 12612
5. sullivanweather
12:28 AM GMT on October 04, 2009
Strong storm going about 5 miles to my east...

Special Weather Statement

821 PM EDT SAT OCT 3 2009

821 PM EDT SAT OCT 3 2009




Member Since: March 8, 2007 Posts: 273 Comments: 12612
3. TheDawnAwakening
11:54 PM GMT on October 03, 2009
I am now calling the region of the NW Atlantic the Gulf Stream Oscillation, 60-80W 35-40N region south of New England, east of the Delmarva peninsula. Buoy 44008 SE of Nantucket, MA is in this region which sees most of the tracks of winter storms in a given winter season. If there is indeed a correlation with the PNA, NAO and GSO positive anomalies, then we will know it with the 30 case study storms shown by Paul Kocin's book Northeast Snowstorms.
Member Since: October 21, 2008 Posts: 248 Comments: 3970
2. sullivanweather
11:26 PM GMT on October 03, 2009
Hiya Art!

Everything is going good here. Finally had time today to sit and write out a lengthy blog for once. Feels nice to have a free day.

It does seem as if we in for a pretty bad winter but I'm not 100% sure where this is going just yet. Still some time remaining for sub-seasonal shifts in the overall pattern to occur which may become seasonal patterns for the wintertime.

Hope all is well with you too!
Member Since: March 8, 2007 Posts: 273 Comments: 12612
1. bigtrucker
11:17 PM GMT on October 03, 2009
Hi Sully, Nice update!!
There has not been a pattern change since last spring. Upper level lows marching from the Great lakes to New England. I think there will be a lot of lake effect snow this year.
Hope all is well with you
Member Since: January 9, 2006 Posts: 80 Comments: 6119

Viewing: 23 - 1

Page: 1 — Blog Index

Top of Page

About sullivanweather

Thomas is an avid weather enthusiast, landscaper and organic gardener. This blog is dedicated to Northeast and tropical weather forecasting. Enjoy!

Local Weather

63 °F

sullivanweather's Recent Photos

Mid-October snowfall
Mid-October snowfall

Personal Weather Stations

About Personal Weather Stations