Synoptic-Scale Precursors to the Potential Upcoming Severe Weather Event over the Plains

21 February 2007

Map,

I thought I’d check out some synoptic-scale precursors to the potentially widespread severe weather event over the Plains anticipated for this Friday the 23rd (see SPC’s Day 3 outlook). The most obvious precursor to the deep upper-level trough forecast to dig over the western U.S. (see, e.g., the 1200 UTC 2/21 54-h GFS forecast obtained from the University of Utah webpage) is an impressive 19–21 February upstream ridge eruption over the central Pacific. The following images show the poleward eruption of high potential-temperature air on the dynamic tropopause (DT) between about 160°E and 170°W from a “northern polar” perspective (taken from the McTaggart-Cowan GFS diagnostic animation builder):

o 0000 UTC 2/19

This ridge eruption appears to have occurred in response to a relatively intense cyclone associated with high values of layer-mean 850–925-hPa cyclonic relative vorticity (black contours in the above figures) initially near 155°E on 0000 UTC 2/19.

The following images show the evolution of this cyclone between 0000 UTC 2/18 and 0000 UTC 2/20 (same maps as before, except from a West Pacific/Asia perspective):

Of particular interest is that this cyclone appears to have originated along the southern wave guide (indicated by the tight meridional potential temperature gradient) located near 30°N before being “handed off” to the northern wave guide located near 45°N by 0000 UTC 2/20.

The last set of images (from the same source; 300-hPa wind, 1000–500-hPa thickness, and SLP) shows that the evolution of the jet associated with the cyclone:

At the start of the period (1200 UTC 2/17 and 0000 UTC 2/18), the cyclone is tied to the exit region of an intensifying, zonally oriented subtropical jet streak over eastern China and the West Pacific Ocean. In the next 24 hours, 1200 UTC 2/18 and 1200 UTC 2/19, the cyclone is located beneath the left-exit region of the jet streak and the right-entrance region of a second, more poleward jet streak, both which appear to strength in response to the cyclone. By 0000 and 1200 UTC 2/20, the jet streak downstream of the cyclone has moved away and the cyclone is located beneath the left-exit region of a meridionally oriented jet streak.

Judging by the initial development of the cyclone in the exit region of an intensifying subtropical jet streak over eastern China and the West Pacific Ocean, it seems that the intensification of the subtropical potential-temperature gradient on the DT associated with jet streak formation over this region is an important precursor to the severe weather event over the Plains expected for this Friday.

Now, some questions:

o Is the development of the subtropical jet streak linked to diabatic outflow from tropical deep convection?

o How often do West Pacific cyclones originating in the subtropical jet reach high latitudes (this cyclone originated near 32°N and reached approximately 60°N)?

o Is the genesis of these types of cyclones inherently less predictable than the genesis of purely extratropical cyclones?

Comments/contributions welcome…

Heather

………………………..

Email responses (updated 2/21):

o 1. From Lance to me (2/21):

…The STJ influence is clear. The arctic PV anomaly is coming into

play only at the most recent time available (00Z/21).

I would add one more question: Is the arctic PV anomaly the

extra ingredient needed to pull the ridging even further poleward

into high latitudes so as to maximize the possibility for downstream

development to occur and, if so, is this another predictability issue?

………………………..

o 2. From me to Lance (2/21):

I do think the arctic PV anomaly is probably necessary for ridging at the

higher latitudes, and thus, downstream development. Without its

influence, I imagine that the cyclone would have had a more zonal track and the

bent-back ridging into high latitudes would not have occurred…If

we do get a great severe wx outbreak, it could be very interesting to look at

the event for a range of temporal and spatial scales.

………………………..

o 3. From Ryan Maue to me (2/21):

I spent a while compiling the ERA40 storm tracks and writing fancy code to

whip up any sort of filter one can dream up.

So, I would like to weigh in on your questions. Feel free to add whatever

to your discussion, or tell me I am completely wrong…

Is the development of the subtropical jet streak linked to diabatic

outflow from tropical deep convection?

>> I could envision enhancements caused by Tropical convection, however,

the mean large scale circulation during Jan - Feb has the jet max

located off the tip of Japan, hence it is the most active extratropical

genesis region in the Northern Hemisphere.

How often do West Pacific cyclones originating in the subtropical jet

reach high latitudes (this cyclone originated near 32°N and reached

approximately 60°N)?

>> Since the vast majority of W Pac cyclones originate off this jet, it is

interesting to see what is special about a cyclone that makes it to

north of say 55N. Obviously, if it has the ability to travel about 20

degrees of latitude, there is considerable upper-level

steering/shear/baroclinicity for sustained and especially explosive

growth, oftentimes.

So, in phishing through the ERA40 for Januarys and Februarys 1958-2002, I

filtered out cyclones that originated near the STJ location (see figure)

and required the cyclone to travel north of 55 degrees. I only consider

storms that last longer than 24 hours, undergo at least 2 mb of

intensification in 24 hours, and achieve a minimum central pressure of

lower than 1003 mb.

I got 175 such cyclones (~2 per year in each month) of which 102 are

"bombs" or undergo rapid deepening of 1 bergeron (geostrophically adjusted

to 45N) for a period of 24 hours. The average duration of these cyclones

is 6 days and on average have intense warm core structure at the surface

according to Hart (2003) phase space characteristics (-VTL = 129 mean

value for the 175 storms) --> Hence they are bona fide warm seclusions,

which are my dissertation topic (shameless plug).

More trivia: These cyclones on average reach their minimum SLP at 50N.

Is the genesis of these types of cyclones inherently less predictable than

the genesis of purely extratropical cyclones?

>> These are the archetype extratropical cyclones, I see nothing impure about

them. No clue about forecasting.

I also attached some frequency/density cyclone phase plots for the overall

tracks and genesis of these cyclones.

Figure 1 2 3 4 5 6

Captions:

(1) Cyclone phase space density plot: -VTL vs B for all track points of

the 175 cyclones that originiate in the "genesis" box and exceed 55N

during Jan and February. Binsizes for the density plot are 20 for -VTL

and 5 for B. Thus, the vast majority of the cyclones' lifecycle occurs

in the lower-level warm core/symmetric area of the cycphase.

(2) Same as (1) except -VTL vs -VTU. Binsizes are 20 by 20.

(3) Genesis density plot for the 175 cyclones. Thus, the -VTL and B for

the 175 cyclones -- showing strong cold-core asymmetric structure at the

beginning of their lifecycle.

(4) Same as (3) but for -VTL and -VTU. Full tropospheric cold-core

structure.

(5) The lat/lon positions of the 175 cyclones are binned and counted.

Their density is plotted with the maximum location scaled to 25.

(6) Storm tracks. Compare to (5).

………………………..

o 4. To Ryan from me (2/21):

Wow, really great stuff, Ryan…Hopefully I'll have some more comments for you once I take the time to look at your figures more carefully...

A point of clarification: in my last question ("Is the genesis of these types of cyclones inherently less predictable than the genesis of purely extratropical cyclones?") I meant the genesis of cyclones over the subtropics vs. over higher latitudes - I didn't mean to imply that these are not extratropical systems.

………………………..

o 5. To Ryan from me (2/21):

Hi Ryan,

To clarify, in regards to your comment about the cyclones having surface warm core characteristics, I assume you are referring to characteristics at the end of their life cycles (once they reach 55°N)?

Following up on your comment about the position of the climatological jet max off the tip of Japan during January and February, I created a composite map of the Jan– Feb 200-hPa wind direction and speed from CDC’s Monthly/Seasonal Climate Composites page using the NCEP/NCAR reanalysis. Sure enough, the composite map shows a subtropical jet max over the Pacific, putting the right-entrance of the climatological Pacific jet over the area that the cyclone formed in.

Perhaps surprisingly, this particular cyclogenesis event appeared to be forced by dynamics associated with the exit region of a jet streak over China rather than the entrance region of the Pacific jet. Of your subset of 175 cyclones that developed in this region and made it to 55°N (figure 6 from your post), any sense as to what fraction formed in the exit region (rather than entrance region) of a subtropical jet streak? I would guess that only some cyclones forming in the exit region of a jet streak in this area are able to reach the high latitudes, and that the ones that do receive considerable dynamical support from a disturbance of polar origin at some point along the way.

Heather

o 6. From Ryan to me (2/21):

Heather,

These cyclones only have cold-core structure in the lower-troposphere

during their most rapid development period, whether they qualify as a

"bomb" or not. Thus, the average bomb in the Pacific reaches its minimum

SLP (and max gale radius) in 60 hours, its maximum -VTL in 66 hours (6

hour lag from the MSLP), its maximum -VTU in 74 hours, and lysis at 115

hours. The long period between minimum SLP and lysis is largely spindown

of the equivalent barotropic structure, I would surmise. Hence, marine

"bombs" have a positive VTL for ~60% of their lifecycle. Calculation of

the VTU parameter between (600-300 hPa) is contaminated by the low

tropopause and intrusion of stratospheric air (hi PV), so it is not

necessarily as valuable for extratropical systems.

I don't know of a quick way to test whether a given cyclone is in the

entrance/exit region of a jet without manually eyeballing them. Any ideas

on how to automate?

I removed the requirement that the cyclones must reach 55N and gathered

new statistics.

This time: got 561 cyclones from the genesis box (~14 a year) and 255

bombs and a MSLP reached at 45N. Thus, there seems to be a significant

difference in terms of "explosive" capability for storms that exceed 55N,

since the percentage of bombs is 60% (exceed 55N) versus 45% (no latitude

restriction).

These cyclones that have a strong meridional component of motion seem a

lot like extratropical transition (ET) storms, where the configuration of

the jet (zonal index) determines a lot about intensification (i.e. Klein

et al. 2001)

Cheers

RYAN

o 7. From Ryan (2/22; follow-up on post 3):

The maximum mean -VTL of 129 represents simply

the MAX -VTL for each of the 175 cyclones and is only one point out of an

average of 19 (avg. duration from genesis to lysis). Thus, on the density

plot, their "weight" or frequency does not show up as a bullseye.