Friday Map Discussion: 9 November 2007

Part I of Friday map discussion for 10 Nov'07 focused on the European wind storm of 8-9 Nov'07. This storm featured widespread gale-to storm-force winds over the North Sea, the extreme eastern part of the UK and over portions of northern France, Belgium, the Netherlands and Germany. The storm deepened to ~975 hPa beneath strong northwesterly flow as it moved E/ESE from the vicinity of Iceland just before 0000 UTC 8 Nov to just north of Denmark by 0000 UTC 9 Nov. The strongest winds occurred to the SW of the track of the storm where the sea level pressure gradient was strongest between the deepening cyclone and a 1036+ hPa anticyclone situated to the west of the UK.

A comparison of NCEP/OPC analyzed North Atlantic surface analyses and GFS-initialized automated analyses revealed that the storm formed shortly after 0000 UTC 5 Nov. It was manifest as a weak closed circulation with an estimated central pressure of ~1010 hPa near 40 N and 60 W at 1200 UTC 5 Nov in advance of a moderate cold-core trough that was crossing the mid-Atlantic coast. Storm formation occurred along the warm side of a moderate 1000-500 hPa thickness gradient left behind in the wake of TC/EC Noel. Ex-TC and now EC Noel, still showing signs of being warm core, was located near 60 N and 58 W at 1200 UTC 5 Nov. Subsequently, the storm moved NE and weakened slightly. By 0000

UTC 7 Nov it was manifest as a weak 1014 hPa surface trough near 57 N and 42 W. Noteworthy, however, was that the weak surface trough was becoming situated beneath the equatorward entrance region of an anticyclonically curved 250 hPa jet stream at this time.

By 1200 UTC 7, deepening was under way as evidenced by the appearance of a closed 1008 hPa cyclone near 62 N and 28 W to the SW of Iceland. At this time the now-deepening surface cyclone was passing poleward of a 1036+ hPa anticyclone situated near 48 N and 13 W. At 0000 8 Nov, 1200 UTC 8 Nov and 0000 UTC 9 Nov the storm was located near 62 N and 11 W (~985 hPa), 61 N and 3 E (~975 hPa), and 59 N and 10 E (~975 hPa), respectively. During the rapid deepening phase the storm graduate became associated with the poleward exit region of the next upstream jet.

Loops of potential temperature and winds on the dynamic tropopause (DT) and 925-850 hPa relative vorticity (selected images are attached) obtained from the McTaggart-Cowan GFS animation builder (link given below) established that the storm was influenced by two DT disturbances (PV anomalies). The first was the aforementioned disturbance that crossed the mid-Atlantic coast. This disturbance remained weak

and induced no deepening as the storm moved NE toward Iceland by 0000 UTC 7 Nov. Deepening began in earnest, however, after 1200 UTC 7 Nov as the second DT disturbance, this one of arctic origin that had broken away from an arctic vortex over extreme northeastern Canada, began to interact with the surface storm as it was moving eastward just south of Iceland.

The second DT disturbance is best seen at 0000 UTC 8 Nov as it is breaking away to the ESE of the core arctic vortex over extreme northeast Canada. At issue is to what extent secondary ridging on the DT in the Davis Strait and southern Greenland area at this time may have distorted the arctic vortex over extreme northeastern Canada sufficiently to enable a piece of the vortex to break away to the E/ESE and

induce rapid storm deepening downstream as a result. The observed ridging over the Davis Strait and southern Greenland was occurring ahead of a moderately intense storm (sub 996 hPa central pressure) located over northeastern Labrador at 0000 UTC 8 Nov.

In effect, discontinuous retrogression of the long-wave upper-level ridge situated over the North Atlantic was in progress during this period. The first ridge eruption occurred over northeastern Canada and the northwestern Atlantic on 4-5 Oct in conjunction with the ET of Noel as the storm approached the Davis Strait. The second ridge eruption occurred in roughly the same location on 7-8 Oct in conjunction

with a trough passage across eastern North America. A related scientific question is how the various transient synoptic-scale disturbances interacted with one another and the larger scale ridge during the discontinuous retrogression of the long wave ridge.

Part II of Friday map discussion on the recent flooding rains in the Mexican state of Tabasco will follow later.

Lance

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

Part II:

Hi Folks,

The second part of Friday map discussion for 9 Nov'07 focused on the very heavy and destructive rains in the Mexican state of Tabasco (adjacent to the southern end of the Bay of Campeche) at the very end of October 2007. These rains triggered massive flooding in many parts of Tabasco. Details on the flooding (in Spanish) can be found at the following link:

http://smn.cna.gob.mx/ (click on link at left entitled: "Especial: Resena del frente frio No 4")

The heavy rains followed a few days after the passage of a strong cold front across Mexico on 23-24 Oct. The NCEP/OPC surface maps (attached) for 1200 UTC 24, 26 and 29 Oct show the cold front passing the Yucatan peninsular on the 24th, stalling south of western Cuba on the 26th, and then retreating northwestward into the southern Gulf of Mexico on the 29th. The heaviest rains in Tabasco state fell on 28-29 Oct in the strong N/NE onshore flow that developed to the west of the westward-retreating frontal boundary as sea level pressures (SLPs) fell to the east and rose over the extreme northwest Gulf of Mexico ahead of a second cool surge. Morning and afternoon QuikScat images on 28 Oct and a morning QuikScat image on 29 Oct (link below) show the development of the strong northerly flow over

the Bay of Campeche. Of interest, is that the strongest winds (> 15 m/s) develop west of southern Florida and then move SW toward the northwest corner of the Yucatan peninsula. By the morning on 29 Oct the winds in the southern Bay of Campeche reach 20 m/s (and are locally higher) and spill into the Gulf of Tehunatepec as a part of a strengthening Tehuantepecer.

Individual integrated precipitable water (PW) mosaics (attached) for the Gulf/Caribbean/Atlantic at 1200 UTC 27-29 Oct show that a corridor of high PW values became established from Tabasco state NNE/NE to northern Florida. The attached PW images were extracted from individual loops at the following link:

http://cimss.ssec.wisc.edu/tropic/real-time/tpw2/global/main.html

http://www.ssmi.com/qscat/scatterometer_data_daily.html

Select the North Atlantic link and then click on the archive button below the map to access older loops. The take home message from the PW loops encompassing the 28-29 Oct period is the development of a corridor of high PW values that move SSW/SW toward Tabasco state as deep tropical moisture over the

Caribbean associated with the percolating Noel is transported westward and southward.

Supporting evidence for the aforementioned high PW corridor and the westward displacement of the frontal boundary by 29 Oct is found in the attached 1200 UTC 26 and 28 soundings (from U-Wyoming; link below) for 76644 (Merida, Mexico). Weak cold advection below 700 hPa on the 26th is replaced by weak warm advection below 700 hPa on the 28th as the PW value increases from ~29 mm to ~47 mm. Backing winds above 450 hPa between the 26th and the 28th are indicative of trough development over the northwestern Gulf of Mexico and ridge building aloft to the east in conjunction with the development of TD/TS Noel near Hispaniola. A visible image for 1745 UTC 28 Oct'07 (source: GIBBS; link below) shows a solid band of cloudiness from Tabasco state NNE/NE to northern FL in conjunction with the corridor

of high PW values. Coarse resolution satellite-derived rain-rain measurements (mm per 3 h) obtained from the Navy/NRL (link below) show widespread 20-40 mm rainfall totals in the 3 h periods ending 0000 and 1200 UTC 29 Oct that coincide nicely with the aforementioned corridor of high PW values that extends into

Tabasco state. There is also evidence of significant orographic rainfall enhancement in the mountains to the south of Tabasco ending 0000 UTC 29 Oct. Rivers with headwaters in these mountains empty into the Bay of Campeche and were a source of the widespread flooding.

http://weather.uwyo.edu/upperair/sounding.html

http://www.ncdc.noaa.gov/oa/rsad/gibbs/gibbs.html

http://www.nrlmry.navy.mil/sat-bin/rain.cgi

Selected maps of potential temperature and winds on the DT with layer-averaged 925-850 hPa relative vorticity, and 850/300 hPa streamfunction and nondivergent winds for selected times between 23-29 Oct round out the attached images. All of these images were obtained from the McTaggart-Cowan GFS animation builder (link below). At 0000 UTC 23-24 Oct the NNE-SSW oriented band of 925-850 hPa layer mean relative vorticity that marks the initial cold front can be seen advancing eastward across the Gulf of Mexico. An E-W oriented PV tail from a trough over the central Atlantic is evident from north of Nicaragua eastward to Puerto Rico and beyond. TD/TS Noel would eventually form under remnants of this PV tail a few days later. Weak disturbances (PV anomalies) that fractured from the western end of the PV tail and continued moving westward helped to maintain deep tropical moisture over the western Caribbean. One of these weak fractured PV anomalies is evident near the eastern side of the Yucatan peninsular at 0000 UTC 24 Oct. By 0000 UTC 27 Oct, remnant patches of 925-850 hPa layer-mean frontal relative

vorticity are evident from just west of Florida southward to the east of Yucatan. Comparison with the attached NCEP/OPC surface frontal analysis for 1200 UTC 26 Oct suggests that these vorticity patches lie near and to the west of the analyzed surface frontal position.

At 0000 UTC 28 Oct a concentrated area of 925-850 hPa layer-mean relative vorticity is centered near 25 N and 87 W. This concentrated cyclonic vorticity maximum developed between 0000 and 0600 UTC 27 Oct in the patchy frontal vorticity region noted previously to the west of Florida. This cyclonic relative vorticity maximum can also be seen at 850 hPa to the northwest of western Cuba at 0000 UTC 28 Oct. Over the next 24 h this cyclonic vorticity maximum moves southwestward to the Bay of Campeche (seen also in the 925-850 hPa layer-mean relative vorticity) and is accompanied by a northerly surge into Tabasco state in conjunction with rapidly increasing PW values and the onset of warm advection. To the east, a broad 850 hPa cyclonic circulation strengthens by the 29th as TD/TS Noel begins to percolate. At 300 hPa, a trough over the northwestern Gulf of Mexico at 0000 UTC 28 Oct is associated with confluent flow over the western Gulf and the Yucatan peninsula. Broadly speaking, the corridor of high PW values that develops from northern Florida to the Bay of Campeche and Tabasco state on 28-29 Oct is associated with weak warm-air advection near (but not under) the right-front entrance region of a modest 300 hPa jet that resides over the northwest Gulf of Mexico.

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

Loops of: 1) 1000-500 hPa thickness, 700 hPa vorticity and winds, 2) 925/300 hPa streamfunction and nondivergent winds, and 3) 925/300 hPa velocity potential and divergent winds for this case can be found at the below links constructed and maintained by UAlbany Ph. D. student Heather Archambault. In conjunction with the attached images for specific times, these loops make it evident that the low-level cyclonic vorticity maximum that developed west of Florida on the 27 Oct and reached the southern Bay of Campeche and Tabasco state on 29 Oct was the main player in focusing the excessive rainfall that led to the widespread and destructive flooding in the low-lying areas of Tabasco state.

· 1000–500-hPa thickness, SLP, 700-hPa absolute vorticity, 700-hPa wind

· 500–200-hPa thickness, SLP, 300-hPa absolute vorticity, 300-hPa wind

· Streamfunction, nondivergent wind, full wind vorticity: 925 hPa 300 hPa

· Velocity potential, irrotational wind, full wind divergence: 925 hPa 300 hPa

· 700-hPa geopotential height, temperature, wind, upward motion

· 700-hPa geopotential height, wind, total column precipitable water

· 1000–500-hPa thickness, SLP, 24-h SLP change

The events that conspired to enable the widespread and destructive flooding in Tabasco state strike me as relatively rare (one or two sigma events?). Most cold fronts that reach the Gulf of Tehuantepec are fast-moving and are usually devoid of significant deep moisture and associated heavy precipitation. Such cold fronts are driven by a strong polar anticyclone (usually 1030+ hPa) that surges equatorward along the eastern slopes of the Rockies. As the anticyclone moves equatorward east of the Rockies it often splits into two separate cells. One cell, terrain-tied, tracks equatorward through eastern Mexico while the second cell, dynamically driven, typically tracks eastward along the Gulf coast (polar anticyclones don't like warm water) and then up along the Atlantic coast.

The initial cold front passage across the western Gulf of Mexico on 23-24 Oct mostly followed the classic Tehuantepecer "script" whereas the second event on 28-29 Oct did not. Instead, the second Tehuantepecer event was driven by an increasing SLP gradient across the Gulf of Mexico that was as much a product of weak SLP pressure falls over the eastern Gulf of Mexico and western Caribbean as it was by weak SLP rises over the northwestern Gulf of Mexico. The weak SLP falls to the east occurred in the western envelope of the cyclonic circulation associated with the percolating TD/TS Noel. Low-level warm-air advection in easterly flow centered along a corridor located near ~20 N from the Gulf of Mexico to near Cuba/Hispaniola appeared to play an important role in the observed SLP falls over the eastern Gulf of Mexico and the western Caribbean. The dynamical role of the W-E oriented PV tail described above in facilitating SLP falls associated with the westward-moving frontal boundary and, in effect, inducing a Tehuantepecer and heavy rains beneath a corridor of SSW/SW-moving deep tropical moisture in Tabasco

state needs to be ascertained. Ditto for the organization of low-level vorticity maxima near and close to the frontal boundary in the eastern Gulf of Mexico.

Lance

Part III:

Hi Folks,

Appended below is a post from Jim Steenburgh I received

earlier this evening about a short-lived intense storm in the Black

Sea yesterday (Sat) that sunk a number of ships and drowned an as yet

unknown number of sailors. Included is my response and Jim's response

to me. The bottom line is that the same arctic PV anomaly that

contributed to the rapid intensification of European storm #1 to ~975

hPa over NW Europe likely played a similar role in European storm #2

(I included a brief synopsis in my response to Jim appended below).

There is also the suggestion that the arctic PV anomaly received a

small reinforcement of arctic air from the NNW in the 12 h ending

00Z/10.

Attached are selected images of SLP/1000-500 hPa thickness

/250 hPa isotachs, and DT pressure/850 hPa-DT shear/925-850

layer-mean relative vorticity at 12 h intervals for 9-11 Nov'07 (all

images were obtained from the McTaggart-Cowan GFS animation builder).

The salient points are as follows:

1. The arctic PV anomaly that helped to intensify European storm #1

outran this storm as it dived SE toward the Alps while the storm

itself moved ESE and filled.

2. Cold air moving SE in the wake of European storm #1 was clearly

blocked by the Alps as evidenced by the 1000-500 hPa thickness

packing on the north/east side of the Alps at 12Z/9.

3. Because the Alps have a finite east-west width, the advancing

cold air surged around the eastern end of the barrier and cascaded SE

across Italy and the Adriatic Sea.

4. After the arctic PV anomaly crossed the Alps and encountered the

strengthening baroclinic zone from the Adriatic Sea eastward to the

Ionian Sea it triggered new cyclogenesis (note the closed 1004 hPa

isobar over the southern Adriatic Sea at 00Z/10).

5. Rapid deepening occurred in two stages: 1004 to <992 hPa in the 12

h ending 12Z/10 and <992 hPa to <980 hPa in the 12 h ending 06Z/18

when the storm was over the Black Sea. Little change in central SLP

is indicated in the GFS initialization between 12Z/10 and 18Z/18

while the storm was over land.

6. As Jim rightfully pointed out, European storm #2 could have been

considerably more intense than shown on the mesoscale over the Black

Sea in response to significant surface heat and moisture fluxes.

7. Between 00Z/10 and 12Z/10 the coupling index (CI; not shown;

defined as theta on the DT minus theta-e at 850 hPa) was < -5,

indicative of tropospheric-deep instability over the southern Aegean

and Ionian Seas where European storm #2 first intensified rapidly.

The storm's transit across the Black Sea during the second phase of

rapid deepening was accompanied by CI values <0.

Interesting series of storms overall.....

Lance

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

Date: Sun, 11 Nov 2007 18:10:59 -0700

From: Jim Steenburgh <Jim.Steenburgh@utah.edu>

To: meteo-inscc@lists.utah.edu, adam.varble@utah.edu,

Jon Zawislak <jon.zawislak@utah.edu>,

Lance Bosart <bosart@atmos.albany.edu>

Subject: Loss of ships on the Black and Azov Sea

Folks:

If you don't believe that shipwrecks still happen, check out the

book "Extreme Waves" by Craig B. Smith, or this latest posting on CNN

about the disaster on the Black and Azov seas:

http://www.cnn.com/2007/WORLD/europe/11/11/russia.spill/index.html.

What happened over the Black Sea? GFS analyses

(http://www.met.utah.edu/cgi-bin/jimsteen/gfs/displayGFS.cgi?region=SY_EU&fileHour=anal07111118F000)show that the incipient cyclogenesis occurs in classic fashion to the lee of the Alps as an upper-level shortwave trough impinges on that barrier in large-scale northwesterly flow. The cyclone then drifts

SEward to the boot-heel of Italy before swinging eastward and then

NEward across the Black Sea.

It's relatively unremarkable case of cyclogenesis until until the

low moves over the Black Sea. At around 10/18Z the analyzed central

pressure is only about 988 mb, but after this time the cyclone undergoes

incredible mesoscale intensification, dropping about 12 mb in 12 h (in

the analyses, which undoubtably do not resolve the subsynoptic scale

structure of this event). An incredible pressure gradient develops in

the wake of the storm over the Black sea. The analyzed cyclone fills

rapidly as soon as it moves over mainland Russia.

This suggests that this event may be similar to an "arctic

hurricane". Rapid mesoscale development occurs as nearly vertically

stacked surface low and upper-trough move over the warm waters of the

Black Sea. Similar mid-latitude events have been documented over the

Mediterranean by Dick Reed and others. It would be interesting to know

if SSTs over the Black Sea are unusually high and if this may have

contributed to the extreme nature of this event. I'd also like to know

if this has been an unusually warm fall in this part of Eurasia.

Jim

Hi Jim,

The arctic PV anomaly that got involved with the rapid

deepening phase of the NW Europe storm (European Storm #1) of 8-9 Oct

that was the subject of my part I Friday map discussion post from 9

Oct was also responsible for triggering new cyclogenesis to the lee

of the Alps (reminds me of the Bleck and Mattochs and Mattochs and

Bleck papers from the mid-1980s). This arctic PV anomaly outran its

original progeny but then latched on to the surface baroclinic zone

on the east side of the Alps (it looks to me that the cold air in the

wake of European storm #1 was initially blocked by the Alps but then

did ye olde end-around to the east of the mountains. From there, the

cold air raced down the Adiratic Sea and then established a

relatively strong baroclinic zone from the Ionian Sea eastward to the

Aegean Sea. With the arrival of the arctic PV anomaly it was then

Katie bar the door time and European Storm #2 was born.

Go ahead and send a post to map?

Lance

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

Date: Sun, 11 Nov 2007 19:24:32 -0700

From: Jim Steenburgh <Jim.Steenburgh@utah.edu>

To: Lance Bosart <bosart@atmos.albany.edu>

Subject: Re: Loss of ships on the Black and Azov Sea

Lance:

Yes, go ahead and sent to Map. The cyclogenesis in the lee

of the Alps is of the Uberstromungs type (Pichler and Steinacker

1987), which I note only because I like the name. Based only on a

quick and dirty analysis based on the GFS, it reminded me alot of the

case described in the classic paper by Buzzi and Tibaldi 1978. I

wonder how good the forecasts were. You can't get the mesoscale

right if you can't get the large scale right. A small error in the

track of the PV anomaly and low center would have probably yielded

nothing like was observed. Perhaps this contributed to so many

ships being caught off guard.

Jim