East-Coast Woes and Cut-Off Lows

By: 24hourprof , 10:10 PM GMT on October 10, 2013

Share this Blog
9
+

The current 500-mb pattern across the contiguous states provides a learning opportunity so that readers have a better sense for cut-off lows. Although the standard contour interval for 500-mb heights is 60 meters (see today's 18Z Rapid Refresh model analysis of 500-mb heights), I generated the 18Z Rapid Refresh model analysis of 500-mb heights (below) using 30 meters as a contour interval to better indicate the cut-off low over the Middle Atlantic States.


The 18Z Rapid Refresh model analysis of 500-mb heights on October 10, 2013. Contour interval is 30 meters instead of the standard 60 meters. This non-standard contour interval allows us to better identify the closed low over the Middle Atlantic States. Courtesy of Penn State.

I summarize, on the image below, the important distinction between a generic closed low (such as the closed low centered over northern Arizona on the 500-mb chart above) and a generic cut-off low like the one of the Middle Atlantic States today.


Courtesy of, and copyright by, Penn State's online certificate program in weather forecasting.

No, the 500-mb low over northern Arizona today (revisit the 18Z RR 500-mb analysis) is not cut-off from the westerlies. But it is a closed low. That's as far as you can go there.

True cut-off lows like the one over the Middle Atlantic States today are typically quite lethargic. One thing's for sure...there's very low confidence in the details of the quantitative precipitation forecasts (QPF) associated with this cut-off low. When I was a young forecaster (a long time ago), one of our mantra's was "Upper-level low, weatherman's woes."


The 1455Z mosaic of composite reflectivity on Friday, October 11, 2013. Note the mesoscale bands of convectively enhanced rain (mostly in yellow). Larger image. Courtesy of Penn State.

Update: Indeed, it's impossible to predict, except for very short-range forecasts, where and when mesoscale bands of convectively enhanced rain will form. To see what I mean, check out the 1455Z mosaic of composite reflectivity (above; larger image) on Friday, October 11, and note the mesoscale bands of convectively enhanced rain (mostly in yellow).

Lee

Reader Comments

Comments will take a few seconds to appear.

Post Your Comments

Please sign in to post comments.

or Join

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: 17 - 1

Page: 1 — Blog Index

17. WunderAlertBot (Admin)
1:34 PM GMT on October 14, 2013
24hourprof has created a new entry.
16. 1900hurricane
3:08 AM GMT on October 14, 2013
Hey Lee, I have a somewhat off-topic question to ask you if you don't mind, somewhat in the vein of your Colorado Flood blog. In that blog, you cited the media blaming the floods on EPac monsoonal moisture while you came to the conclusion that their presumption was very unlikely the main contribution (a conclusion that I very much agreed with). However, this got me thinking about Texas and the role the EPac, particularly their tropical cyclones, plays on the weather there.

Oftentimes EPac tropical cyclones are pointed to as the cause of heavy rainfall in my state, something I have accepted without question. However, my thinking back on your recent blog has made me give a second thought on these events. Like the plains and foothills of the Front Range, Texas lies on the other side of the continental divide from the Pacific Ocean and mountains in excess of a mile high lie between the two locations.



However, there are some notable differences. For one, Texas lies hundreds of miles closer to the Pacific than Colorado. Also, while the mountain ranges that traverse central Mexico are significant geographic features, they do not have the stature of the Rockies in Colorado.

Now to a few of the events themselves. Already this year one tropical cyclone has been pointed to as enhancing Texas rainfall (Hurricane Manuel), and a second event involving Tropical Storm Octave is currently in progress.



Many other Texas autumn rainfall events have occurred in concert with an EPac tropical cyclones, but perhaps the two largest in Texas history occurred in October 1994 with Hurricane Rosa and in October 1998 with Hurricane Madeline. Both of these systems are said to have been crucial to their respective rainfall events that dropped in the neighborhood of 30 inches of rain across parts of Texas. However, any EPac tropical cyclone that impacts Texas has to traverse land in excess of 2 kilometers high, so the lowest level moisture from the Pacific that can affect Texas would be somewhere between 850-700 mb.



Now if you have made it this far through my above rambling (I was quite distracted while typing this). I guess my ultimate question is as follows.

How much do Eastern Pacific Hurricanes really contribute to Texas rainfall in these types of events?
Member Since: August 2, 2006 Posts: 46 Comments: 11660
15. 24hourprof
6:34 PM GMT on October 13, 2013
Quoting 14. georgevandenberghe:


Thanks for reply. We're not going to handle 24 hour out placement of bands this decade. But the synoptic scale miss, more precip than forecast and 200km to the west of where forecast, is something we can analyze at present. One commenter (rpointwx) on Long Island on this blog thread mentioned he got almost nothing although much was predicted.

With this rain the heavy irrigation I had to do till Sunday 10/6 is over for this season. It was enormously beneficial for this area.


Agreed on all counts. For the first time in a while, my lawn looks green!
Member Since: October 24, 2012 Posts: 90 Comments: 798
14. georgevandenberghe
5:04 PM GMT on October 13, 2013
Quoting 13. 24hourprof:


Hi George,

Both. A few of of the forecasts I saw on television had New York City and southern New England getting hammered (as much as two inches of rain). Yet, look what happened:



Do we have any New York City readers (or southern New England readers) who can also discuss the busts of local forecasts for the 12th?

Southern New Jersey was predicted to receive the brunt of rain, but, as you can see, that didn't happen either (even though that area received a good rain).

So I'm saying both synoptic scale (the overall region affected by the upper-level low) and mesoscale bands of convectively (and orographically) enhanced rain were involved...upper-level lows, weatherman woes.

Best,

Lee


Thanks for reply. We're not going to handle 24 hour out placement of bands this decade. But the synoptic scale miss, more precip than forecast and 200km to the west of where forecast, is something we can analyze at present. One commenter (rpointwx) on Long Island on this blog thread mentioned he got almost nothing although much was predicted.

With this rain the heavy irrigation I had to do till Sunday 10/6 is over for this season. It was enormously beneficial for this area.
Member Since: February 1, 2012 Posts: 17 Comments: 1585
13. 24hourprof
3:53 PM GMT on October 13, 2013
Quoting 12. georgevandenberghe:


But the question I am now asking is what went wrong. Was it a synoptic scale miss or just enhanced banding that we missed? I'm not arguing that most of the precipitation wasn't in mesoscale bands. It was and it usually is in these systems. Forecasting where these will set up within a 200km square region is not currently possible as you say.
However this wasn't a situation where the areawide precipitation forecast verified but most of it was focused in a band that got much more while other locations got less. Most of the southeastern half of PA got much much more than forecast from several bands (an areal or basinwide bust by a factor of four rather than a point bust) and when there are a lot of bands and they move, precip amounts do tend to average out. And if it was a synoptic scale bust there is more hope for learning what happened and not repeating it because both theory and numerical modeling can handle synoptic space and time scales.

My soapbox is creaking so I'll get off. I am hindered because I still (being shut down) don't
have access to the data, particularly old forecasts, and analyses that I usually look at so my analysis of this event is not very good.
(modified above evening 10/12 to correct grammar and poor wording)


Hi George,

Both. A few of of the forecasts I saw on television had New York City and southern New England getting hammered (as much as two inches of rain). Yet, look what happened:



Do we have any New York City readers (or southern New England readers) who can also discuss the busts of local forecasts for the 12th?

Southern New Jersey was predicted to receive the brunt of rain, but, as you can see, that didn't happen either (even though that area received a good rain).

So I'm saying both synoptic scale (the overall region affected by the upper-level low) and mesoscale bands of convectively (and orographically) enhanced rain were involved...upper-level lows, weatherman woes.

Best,

Lee
Member Since: October 24, 2012 Posts: 90 Comments: 798
12. georgevandenberghe
4:47 PM GMT on October 12, 2013
Quoting 11. 24hourprof:


Here are the NAM grid interpolations at Harrisburg, Pa. from the 12Z run yesterday:



Looking at the METARS, the Harrisburg, Pa. area had roughly eight inches of rain yesterday and so far today. If that's not attributable to enhanced convective mesoscale bands, I don't know what is.

Just a terrible forecast, in my view.

The GFS was just as bad.

Lee


But the question I am now asking is what went wrong. Was it a synoptic scale miss or just enhanced banding that we missed? I'm not arguing that most of the precipitation wasn't in mesoscale bands. It was and it usually is in these systems. Forecasting where these will set up within a 200km square region is not currently possible as you say.
However this wasn't a situation where the areawide precipitation forecast verified but most of it was focused in a band that got much more while other locations got less. Most of the southeastern half of PA got much much more than forecast from several bands (an areal or basinwide bust by a factor of four rather than a point bust) and when there are a lot of bands and they move, precip amounts do tend to average out. And if it was a synoptic scale bust there is more hope for learning what happened and not repeating it because both theory and numerical modeling can handle synoptic space and time scales.

My soapbox is creaking so I'll get off. I am hindered because I still (being shut down) don't
have access to the data, particularly old forecasts, and analyses that I usually look at so my analysis of this event is not very good.
(modified above evening 10/12 to correct grammar and poor wording)
Member Since: February 1, 2012 Posts: 17 Comments: 1585
11. 24hourprof
10:20 PM GMT on October 11, 2013
Quoting 9. georgevandenberghe:


Thanks for the radar image.

I'm going to retract this newly italicized text pending analysis of older forecasts which I don't have access to right now. The 18Z GFS last night did have a significant rain signal to Western PA.


******* READ WITH CAUTION.. I'm maybe wrong **
However I'm not ready to just wash this away as missing a band location. The area where the majority of these bands set up was 200km to the west of forecast and that's a synoptic scale miss. The miss may be due to convective feedback which some forecast discussions a few days ago mentioned as "spurious" but has turned out real.

** END CAUTION **

Sorry to be skating on thin ice here but I am busier than usual and can't do a proper job of tracking and the above conclusion was not due to careful analysis. I'ts pouring here right now 2"/hour (which will last for maybe ten minutes). The rain is still welcome.




Here are the NAM grid interpolations at Harrisburg, Pa. from the 12Z run yesterday:



Looking at the METARS, the Harrisburg, Pa. area had roughly eight inches of rain yesterday and so far today. If that's not attributable to enhanced convective mesoscale bands, I don't know what is.

Just a terrible forecast, in my view.

The GFS was just as bad.

Lee
Member Since: October 24, 2012 Posts: 90 Comments: 798
10. rpointwx
7:17 PM GMT on October 11, 2013
Quoting 9. georgevandenberghe:


Thanks for the radar image.

I'm going to retract this newly italicized text pending analysis of older forecasts which I don't have access to right now. The 18Z GFS last night did have a significant rain signal to Western PA.


******* READ WITH CAUTION.. I'm maybe wrong **
However I'm not ready to just wash this away as missing a band location. The area where the majority of these bands set up was 200km to the west of forecast and that's a synoptic scale miss. The miss may be due to convective feedback which some forecast discussions a few days ago mentioned as "spurious" but has turned out real.

** END CAUTION **

Sorry to be skating on thin ice here but I am busier than usual and can't do a proper job of tracking and the above conclusion was not due to careful analysis. I'ts pouring here right now 2"/hour (which will last for maybe ten minutes). The rain is still welcome.




NWS busted the forecast here big time. I am up here on Long Island where subsidence and high pressure had prevailed. We have received 0.02 inches of precipitation and it has just remained overcast with most of the precip falling south and west. The models did not do a good job handling this situation. A lot of us here were welcoming the rain because we need it but this was a bust forecast...though a difficult forecast nonetheless with QPF and weak steering currents
Member Since: December 27, 2009 Posts: 0 Comments: 10
9. georgevandenberghe
4:15 PM GMT on October 11, 2013
Quoting 8. 24hourprof:
George and everybody...

I added the radar image below as an update to this blog.


Thanks for the radar image.

I'm going to retract this newly italicized text pending analysis of older forecasts which I don't have access to right now. The 18Z GFS last night did have a significant rain signal to Western PA.


******* READ WITH CAUTION.. I'm maybe wrong **
However I'm not ready to just wash this away as missing a band location. The area where the majority of these bands set up was 200km to the west of forecast and that's a synoptic scale miss. The miss may be due to convective feedback which some forecast discussions a few days ago mentioned as "spurious" but has turned out real.

** END CAUTION **

Sorry to be skating on thin ice here but I am busier than usual and can't do a proper job of tracking and the above conclusion was not due to careful analysis. I'ts pouring here right now 2"/hour (which will last for maybe ten minutes). The rain is still welcome.


Member Since: February 1, 2012 Posts: 17 Comments: 1585
8. 24hourprof
3:46 PM GMT on October 11, 2013
George and everybody...

I added the radar image below as an update to this blog.
Member Since: October 24, 2012 Posts: 90 Comments: 798
7. 24hourprof
3:08 PM GMT on October 11, 2013
Quoting 6. georgevandenberghe:
I've gotten 2.5" of rain from this system so far in DC and it was much needed.


Exactly my point, George. There's no way to predict the mesoscale bands of convectively enhanced rain (except for very short-range forecasts). See 1455Z mosaic of composite reflectivity below (larger image).



Member Since: October 24, 2012 Posts: 90 Comments: 798
6. georgevandenberghe
1:54 PM GMT on October 11, 2013
I've gotten 2.5" of rain from this system so far in DC and it was much needed.
Member Since: February 1, 2012 Posts: 17 Comments: 1585
5. 24hourprof
12:43 PM GMT on October 11, 2013
Nathan,

It took a while to get on the IMD Web site, but IMD uses a 3-minute average and does not take the maximum value from the 1-minute, 3-minute, and 10-minute measurements (reference).

Lee
Member Since: October 24, 2012 Posts: 90 Comments: 798
4. 24hourprof
12:24 PM GMT on October 11, 2013
Quoting 1. Astrometeor:
Indeed a short blog, thanks anyways.

Random question: Which, in your opinion, is better to use when referencing a cyclone/hurricane's strength: 1-minute, 3-minute, or 10-minute wind speed estimates?

The IMD is getting slammed right now over on Masters' blog for stating "Phailin won't become a super cyclone". Link


Well, in the U.S., it's clearly a one-minute average.

Reference: HRD

This method is the de facto standard for assessing storm maximum wind speed. However, some agencies still prefer other standards.

I remember reading in a WMO document that the maximum sustained wind speed was the maximum value of the ten-minute, three-minute, and one-minute measurements. I admit I'm not sure that the IMD adheres to this method.



Member Since: October 24, 2012 Posts: 90 Comments: 798
3. georgevandenberghe
3:24 AM GMT on October 11, 2013
Quoting 2. rpointwx:
Could you describe the genesis of double barreled low pressure system in relation to upper level low?

Also what is the best way to forecast whether the upper level low will generate a surface low? forecasting strengthening or weakening tips or rules?


I don't have any good rules because there are different mass and temperature configurations that can produce upper lows. Lee has already stated that the ones cut off from the steering flow will move little and hang around for days.

Upper lows that have no isotherms going through them are barotropic and will have the same intensity at all levels. Lows that have isotherms parallel to the isobars are said to be equivalent barotropic. In this case the cold pool associated with the low is also cut off from the main temperature gradient region and it will also likely not move. Barotropic lows dissipate through friction on a time scale of days to a week if something else doesn't happen to affect them first. But equivalent barotropic lows, much weaker at the surface than aloft and in the absence of steering or topography can theoretically last for several weeks to over a month. Something usually happens to the steering around them first so we don't see them persist that long and radiative heating will also slowly erase them.. slowly is order ten days, faster than the friction only month plus I assumed earlier. The third case is lows that have isotherms not parallel to the isobars. These are baroclinic and evolve to something else fairly quickly. They also move. Lee can probably say this with fewer words and decent pictures.

A rule of thumb for intensification is that if the upper low tilts into the cold air with height, surface cyclogenesis will occur on the warm side. This is usually the case when the low is embedded in a temperature gradient, the "third case" above.

(one of the things that bothers me about this rule is that ANY upper low with isotherms running through it will tilt into the cold side with height.. it has to in a hydrostatic atmosphere)

Baroclinic lows, the typical ones with fronts ahead of and behind them, intensify much more rapidly in a less statically stable atmosphere. This effect is stronger for shorter wavelengths. The combination of high static stability and short wavelength CRUSHES them, they dissipate. Longer wavelengths are not as much affected. The reason why is that the vertical motions are stronger for the short waves and cause cooling ahead of and heating behind them which more than cancels the warming ahead from horizontal warm advection and cooling behind from horizontal cold advection. The temperature changes from vertical motion are smaller in a less statically stable atmosphere and zero in a neutrally stable one (rare on the synoptic scale) One interesting special case of this is when deep cold air plunges into the U.S. and warms rapidly at the surface but not much aloft. The resultant vortex is exceptionally sensitive to development and surprisingly small and insignificant short waves forming in it or moving into it can trigger rapid cyclogenesis. One of these occurred in mid April 1983, a strong polar outbreak in the Ohio Valley with a short wave plunging down the back side resulted in strong cyclogenesis over the lower Ohio Valley and Northeast resulting in accumulating snow in Central New Jersey on April 19.

Since I'm tapping old memories, critical comment is especially solicited. It's also likely in a response this long I've made at least one mistake.



modification 10/11. The current mid atlantic cutoff has thickness (isotherm) lines running through it and tilts into the cold air. By the simple Rule of Thumb (ROT) I stated above, it should develop and move. It is persisting and converting thermal gradient energy (Available Potential Energy also called APE or "baroclinic energy" by forecasters) to kinetic energy but it is not forecast to move much. This is another case where a simple conceptual model is overwhelmed by reality. The forecast models are handling it passably well so far although I think they missed the westward extent of heavy precipitation into Central Pa. I haven't been tracking it closely.. too busy with work.
Member Since: February 1, 2012 Posts: 17 Comments: 1585
2. rpointwx
12:59 AM GMT on October 11, 2013
Could you describe the genesis of double barreled low pressure system in relation to upper level low?

Also what is the best way to forecast whether the upper level low will generate a surface low? forecasting strengthening or weakening tips or rules?
Member Since: December 27, 2009 Posts: 0 Comments: 10
1. Astrometeor
10:26 PM GMT on October 10, 2013
Indeed a short blog, thanks anyways.

Random question: Which, in your opinion, is better to use when referencing a cyclone/hurricane's strength: 1-minute, 3-minute, or 10-minute wind speed estimates?

The IMD is getting slammed right now over on Masters' blog for stating "Phailin won't become a super cyclone". Link
Member Since: July 2, 2012 Posts: 94 Comments: 9659

Viewing: 17 - 1

Page: 1 — Blog Index

Top of Page

About 24hourprof

Retired senior lecturer in the Department of Meteorology at Penn State, where he was lead faculty for PSU's online certificate in forecasting.

24hourprof's Recent Photos

Portrait