Friday Map Discussion: 16 February 2007

 

Friday Map Discussion Archive 

home

 

Hi Folks,

 

        Friday map discussion for 16 Feb'07 focused on two topics: 1) the upper-level conditions associated with the big lake-effect

snows on the Tug Hill Plateau in the first two weeks of February, and 2) the origin and evolution of the Valentine's Day snowstorm, the

latter of which went in as serendipitous direction.

 

        Because of the size of some of the images, the map discussion synopsis will be posted in two parts. Part I summarizes the Tug

Hill Plateau record snows while Part II, to follow separately, discusses the Valentine's Day storm.

 

I. Tug Hill Plateau snows:

      

        The daily snowfall amounts (24 h ending 1200 UTC) that contributed to the record-breaking storm-total 141" (3.58 m) of snow

that fell in Redfield, NY, on the Tug Hill Plateau are given below based on official statistics provided by NWS WFO BUF. The

record-breaking event was defined by the duration of the snowfall as much as the overall amount of the snow. The maximum 24 h reported snowfall ending 1200 UTC was "only" 28" (71.1 cm). An impressive event but I suspect considerably bigger 24 h snowfall amounts have been observed in the past. Note that until the last decade there were very few cooperative observer reports available to the NWS from the Tug Hill Plateau region. My instincts tell me that amounts greater than what buried Redfield have occurred in the past but that they likely went unobserved.

 

...........................................................

 

NATIONAL WEATHER SERVICE BUFFALO NY

122 PM EST MON FEB 12 2007

 

*** THESE ARE FINAL STORM TOTALS FOR THE EVENT FROM SATURDAY FEB 3

THROUGH THIS MORNING FEB 12 WHEN THE EVENT ENDED. ***

 

** REVISED REDFIELD FINAL TOTAL 141 INCHES....DAILY TOTALS WERE AS

FOLLOWS...

FEB 3...18 INCHES

FEB 4... 7 INCHES

FEB 5...20 INCHES

FEB 6...15 INCHES

FEB 7...28 INCHES

FEB 8...23 INCHES

FEB 9... 6 INCHES

FEB 10..14 INCHES

FEB 11.. 3 INCHES

FEB 12.. 7 INCHES

..............................................................

 

        Loops of: 1) 500 hPa streamfunction and non-divergent winds, 2) potential temperature/winds on the dynamic tropopause (DT) and

the 925-850 hPa layer-mean relative vorticity, and 3) pressure on the DT and the 850-DT vertical wind shear and 925-850 hPa layer-mean relative vorticity generated from the McTaggart-Cowan GFS diagnostic animation builder (link below) can be used to collectively show that the major snowfall amounts at Redfield given in the table above can be associated with the passage of subsynoptic-scale and

mesoscale DT disturbances rotating around the large cyclonic vortex that has been anchored over Hudson Bay for most of February to this point. Note also the narrow and elongated thermal vorticity maximum on the western and southern periphery of the Hudson Bay polar

vortex. It was along this corridor of maximum thermal vorticity that many of the critical subsynoptic-scale end mesoscale upper-level

disturbances propagated southward and eastward toward the Great Lakes.

 

        Selected images from these loops are attached to help illustrate the salient subsynoptic-scale and mesoscale disturbances

responsible for the major snowfall periods as follows:

 

a) 500 hPa streamfunction and nondivergent winds: 1200 UTC 5 Feb, 1800 UTC 6 Feb, 0600 7 Feb and 1200 UTC 8 Feb.

 

b) DT theta/winds and 925-850 layer-mean relative vorticity: 0600 UTC 7 Feb and 1200 UTC 8 Feb.

 

c) DT pressure/shear and 925-850 hPa layer-,mean relative vorticity: 0600 UTC 7 Feb and 1200 UTC 8 Feb.

To really see what is going on, however, you need to rebuild the loops for yourselves. Conventional wisdom for lake-effect

snowstorms as published in countless introductory textbooks is that cold air blowing over open water that is significantly warmer than

the overlying air is the "magic elixir" for a lake-effect snow dump. Less discussed in the literature, but known to experienced

forecasters, is the critical role played by mobile synoptic-scale disturbances in organizing and enhancing lake effect snowfall (e.g.

Lackmann 2001). The availability of higher resolution datasets in the last couple of years (the loops above are based upon 0.5 degree

GFS analyses) has made it possible to see the critical role played by transient subsynoptic-scale and mesoscale disturbances in

controlling the timing, duration and amount of lake-effect snowfall.

 

        These mobile disturbances are every bit as important as overwater fetch and air-water temperature differences in determining

the severity and extent of lake-effect snowfall events. Other factors that may have contributed to the severity of the recently

concluded lake-effect event likely include the relatively warm first half of the winter that allowed for the lake waters to be

characteristically mostly unfrozen by the end of January and the abruptness of the circulation reversal in mid January that enabled a

cold vortex to setup shop over Hudson Bay with repeated incursions of arctic air across the Great Lakes in the subsequent weeks until

roughly 15 Feb.

 

http://www.atmos.albany.edu/facstaff/rmctc/DTmaps/animSelect.php

 

http://www.nrlmry.navy.mil/sat-bin/epac_westcoast.cgi (choose Northwest vapor)

 

Lackmann, G. M., 2001: Analysis of a Surprise Western New York Snowstorm, Weather and Forecasting, 16, 99-116.

 

Lorenz, E. N., 1963: Deterministic Nonperiodic Flow, Journal of the Atmospheric Sciences, 20, 130-141.

 

                                                        Lance

.......................................................

 

Hi Folks,

 

        Part II (Valentine's Day storm) of the two-part Friday map discussion synopsis from 16 Feb'07 follows.

 

II. St Valentine's Day storm:

 

        Last week I wrote about an interesting mesoscale dry vortex over the eastern Pacific that did not seem to be much more than an

idle curiosity in the Friday map discussion synopsis for 9 Feb'07. The relevant excerpt from that discussion follows. As described

below, this apparent idle curiosity has perhaps morphed into one of Lorenz's butterflies. With that proffered bait, read on.....

 

..............................

 

Excerpt from the 9 Feb'07 map discussion post on a dry mesoscale vortex:

       

"A particularly interesting feature was a mesoscale dry vortex that formed along the eastern boundary of a narrow NNW-SSE oriented

streamer of relatively dry air in the mid- and upper-troposphere to the south of the Aleutians on the 7th (check your favorite source

of archived water vapor imagery for the eastern Pacific for the details). As the vortex formed the dry air became secluded within it

while moist air to the north and south wrapped around the vortex. In the WV imagery the vortex appeared as a "donut" with drier air

defining the hole and moister air surrounding the hole. The vortex, almost perfectly circular at times, had a diameter of 200-300 km (a

patch of slightly more moist air resided in the interior of the vortex). A shear line running NW toward the Aleutians from a larger

scale vortex centered near 46 N, 136 W in the attached DT image from 1200 UTC on the 7th coincided fairly well with the initial dry

streamer. The mesoscale dry vortex formed near the NW end of this shear line and can be found near 51 N, 150 W where it was moving

slowly NW at this time. By 0000 UTC on the 8th the mesoscale dry vortex had reached its most poleward position (53 N, 152 W; not

shown). It subsequently turned equatorward and is marked by a small patch of potentially cooler air and a cyclonic circulation near 50

N, 153 W at 1200 UTC on the 8th. The vortex remained located near the tip of the aforementioned shear line as it continued equatorward (the shear line began curving cyclonically within the outer envelope of the overall cyclonic circulation. By 0000 UTC 10 Feb the mesoscale vortex was situated near 38 N, 146 W where it could be identified with a quasi-circular patch of lower-level cyclonic

vorticity for the first time. As of 1200 UTC 10th the mesoscale vortex could still be identified near 35 N, 138 W as it was being

reabsorbed into the trough that had previously made it a free agent. What goes around comes around....."

 

..............................

 

  Missing from the previous week's map discussion post was selected satellite imagery to illustrate the birth of the dry

mesoscale vortex. This oversight (with the benefit of hindsight) has been corrected here with the inclusion of water vapor (WV) imagery

over the eastern Pacific for 1200 UTC 7 Feb, 0000 and 1200 UTC 8 Feb, and 0000 UTC 9 Feb (source: Navy/NRL link given below). At 1200 UTC 7 Feb a narrow swath of dry air runs from near the Aleutians (56 N, 157 W) south-southeastward  to near 40 N, 142 W. By 0000 UTC 8 Feb a mesoscale vortex has formed at the northwestern end of the aforementioned band of dry air near 52 N, 152 W. This mesoscale vortex can be found near 50 N, 153 W at 1200 UTC 8 Feb, and near 46 N, 153 W at 0000 UTC 9 Feb.  The mesoscale vortex appears as a "donut hole" (sans trans fats as it gets ready to transit the eastern Pacific) of dry air in the WV imagery.

 

        Also attached are a series of DT pressure/shear and 925-850 hPa layer-mean relative vorticity images for 1200 UTC 8 Feb, 0000

UTC 10 Feb, 0600 UTC 11 Feb, 0000 UTC 13 Feb, and 0000 UTC 15 Feb to help illustrate that the mesoscale vortex apparent in the WV imagery can also be seen and followed in the GFS analyses (again, build your own loops to see the details). The potential significance of the mesoscale vortex as one of the possible forerunners to the Valentine's Day storm was realized in one collective "oh my god" serendipitous moment by most of the participants in Friday map discussion while we were examining loops of the above imagery. At 1200 UTC 8 Feb the mesoscale vortex so readily apparent in the attached WV imagery is also clearly defined as a mesoscale DT disturbance (thermal vorticity maximum; pressure near 450 hPa) near 50 N, 153 W.

 

        By 0000 UTC 10 Feb the mesoscale DT disturbance has dropped southeastward to a position near 38 N, 146 W. The pressure on the DT is still near 450 hPa and a layer-mean 925-850 hPa relative vorticity maximum is situated immediately to the east of the vortex. By 0600 UTC 11 Feb the mesoscale DT disturbance, slightly weaker, is approaching the coast of California. It can be identified near 37 N, 125 W. It is still associated with a thermal vorticity maximum, and a layer-mean 925-850 hPa relative vorticity maximum out ahead of

it. Its position at this time also places it near the southeastern end of an extended shear line (thermal vorticity trough) to the

northwest. Although the mesoscale DT disturbance lies near the southern end of a larger-scale thermal trough, it has resisted

reabsorption and can still be identified as a separate feature.

 

        For later reference note the larger scale thermal vorticity maximum located just poleward of Hawaii and the associated

acceleration of the subtropical jet (STJ) to the south and east of it of this second trough at 0600 UTC 11 Feb. This second trough and

STJ will eventually catch up and interact with the mesoscale DT disturbance that was approaching the coast of California at this time.

At 0000 UTC 13 Feb a somewhat larger-scale DT disturbance situated near 35 N, 100 W marks the likely position of the mesoscale DT

disturbance that was located just west of California 48 h earlier. To make an independent assessment of this inference, please make

your own loop at the McTaggart-Cowan website. While my inference is not a 100% done deal the evidence is fairly suggestive and should be scrutinized further for errors of omission and commission.

  The DT disturbance near 35 N, 100 W grew upscale over the 48 h ending 0000 UTC 13 Feb as it crossed the southern Rockies and

began to interact with warm, moist and unstable air flowing northward from the Gulf of Mexico (tornadoes in the New Orleans area were

reported ~8 h later near 0800 UTC ahead of this advancing disturbance). Subsequently, the developing surface cyclone moved

northeastward across the Tennessee and Ohio Valleys. With a strong surface anticyclone locked in place north of New York and New

England behind a trough in the northern branch of the westerlies the stage was set for low-level arctic air to pour southward east of

the Appalachians. Accordingly, the stage was set for a warm-air advection and icing event across parts of the Middle Atlantic region as

the cyclone approached from the southwest. The primary cyclone began to weaken and die over Pennsylvania as a secondary cyclone

developed east of the Appalachians along the low-level baroclinic zone bordering the southern extent of the aforementioned arctic air

mass in advance of the DT disturbance.

 

        By 0000 UTC 15 Feb the original eastern Pacific DT disturbance had turned northeastward along the Atlantic coast and was now

associated with a prominent 925-850 hPa layer-mean relative vorticity maximum situated over extreme southeastern New England. This

vorticity maximum marked the now-dominant secondary cyclone (978 hPa) known as Valentine Day's storm. It also appeared from the DT loops that the coastal DT disturbance was also interacting with a second mesoscale DT disturbance located to the west near 39 N, 75 W and a stronger and larger-scale DT disturbance of arctic origin located over the northern Great Lakes.  Cyclogenesis near the Gulf

coast around 0000 UTC 13 Feb was also assisted by the eastward movement and development of the STJ behind the larger scale thermal vorticity trough noted previously just north of Hawaii at 0600 UTC 11 Feb. The developing cyclone at 0000 UTC 13 Feb was situated beneath the poleward exit region of the subtropical jet. Strong coastal redevelopment in the next 48 h occurred in conjunction with ridging over Atlantic Canada that was likely driven by a combination of diabatic heating ahead of the coastal storm and downstream

development ahead of the arctic DT disturbance crossing the northern Great Lakes. A key aspect of this ridging was the observed

collapse of the downstream half wavelength between the advancing and deepening upstream trough and the quasi-stationary and amplifying

downstream ridge.

 

        More than 40 years ago Ed Lorenz speculated (metaphorically) that the flap of a butterfly's wings in Brazil could eventually

lead to a big rainstorm in India. He was discussing how unavoidable errors in the initial conditions of the atmosphere would inevitably

grow and destroy the ability of a numerical forecast to represent the observed state of the atmosphere on a time scale of roughly two

weeks. He based his conclusions and speculations on the results from a highly idealized numerical model (Lorenz 1963). Here I speculate that the "donut hole" mesoscale vortex that formed over the northeastern Pacific ocean on 7-8 Feb may have played the role of one of Lorenz's famous flapping butterflies. Remnant vorticity associated with this mesoscale votex moved onshore in California and became associated with a growing disturbance as it crossed the southern Rockies. A headline might read:

 

   Big "flap" in the eastern Pacific triggers a cascading series of events culminating in "flapogenesis" in the lower Mississippi

Valley, tornadoes in the New Orleans area, and the St. Valentine's Day snowstorm massacre in the Northeast.

 

        A careful diagnostic analysis of events leading up to the Valentine's Day storm may reveal the above highly speculative

development scenario to be a leap off of a tall building without a parachute and without a net below (and without Superman around to

save me from myself). No matter. I am prepared to be wrong as usual. That said, there are some very interesting scientific issues

associated with the Valentine's Day storm that can be summarized as follows:

 

1. Predictability (based upon the GFS) was relatively poor until the southern-stream disturbance neared the California coast.  Up until

that point almost all of the GFS ensemble members were opting for a relatively weak fish storm that would track well out to sea and

away from the I-95 corridor. At issue is understanding what the ensemble members latched onto that first suggested a storm would track

closer to the coast and be stronger than initially forecast.

 

2. Multiple player cyclogenesis. Preliminary evidence discussed above suggests that both upstream (trailing trough and STJ in the

southern stream, weak separate midlatitude DT disturbance, and a stronger arctic disturbance) and downstream (ridge amplification and collapse of the downstream half wavelength) features contributed to the severity of the Valentine's Day storm of 2007. To what extent

these various features act independently and/or in concert with one another needs to be determined to understand the cyclogenesis

process and the associated predictability ramifications, and suggests that the application of "PV surgery" techniques might be good for

the atmospheric patient.

 

3. Assess the role of the "idle curiosity" mesoscale vortex first mentioned in the 9 Feb'07 map discussion post in the eventually

development of the Valentine's Day storm in the grand scheme of things. Was this disturbance merely a turbulent eddy that while

suggestive was the ultimate red herring and a figment of our collective imagination or did it contribute in some unknown way, perhaps

in a stochastic sense, to the cascading series of events that culminated in the Valentine's Day storm and left officials of Jet Blue

Airways feeling.....well....."blue" in the face and embarrassed? And if this disturbance did play a supporting role in the storm then

how do we define and quantify its contribution?

 

     Finally, there were also important societal and economic issues associated with the Valentine's Day storm. This was a

high-impact storm from the lower Mississippi Valley to the Northeast. For example, Jet Blue Airways suffered a public relations

disaster with grounded airplanes, delayed and cancelled flights, flights stranded on JFK tarmacs for hours, and thousands of irate and

enraged passengers. Recovery from this natural- and self-inflicted debacle is by no means assured. Likewise, the disaster on the

Pennsylvania interstates (e.g., the 50 mile traffic jam on I-78 that closed the freeway more more than 48 h and trapped hundreds of

people in their cars and trucks) was such an embarrassment for the state government that it could be a future campaign issue.

 

http://www.nrlmry.navy.mil/sat-bin/epac_westcoast.cgi (choose Northwest vapor)

 

Lorenz, E. N., 1963: Deterministic Nonperiodic Flow, Journal of the Atmospheric Sciences, 20, 130-141.

 

                                                Lance

 

Attachments:  http://www.atmos.albany.edu/student/heathera/mapdisc_02-16-07/