Confused about the cap...

[Steve Conley]

Hey all,

So I was on the same bust a lot of folks were on here in Iowa last Thursday. As I read back through some threads, I believe the long story short is that the cap just held a lot longer than expected. While I'm not 100% clear on the CAP just yet (save that one for later), I do have a question regarding the visualization of this taking place (IE, towers not being able to 'bust through' the cap).

As I understand it, CU's form anvils when their updraft bumps into the cap, so to say. Anvils are obviously pretty easy to identify. Then, when the cap gives way, you can get an overshooting top, which indicates there is some seriously impressive updrafts.

So my question is this. On a day like last Thursday, where I watched towers just go up and down all night, if it was the cap holding them back, why wouldn't I have seen anvils in this case? In other words, if the cap is holding, what conditions dictate whether the clouds actually form an anvil, or just fall back down again.

I know those are a lot of lamens terms, I hope I made it clear, sorry if this has already been addressed.

Thanks in advance for any help.

 

[Adam Lucio]

A storm doesn't form an anvil when it hits the cap. The anvil forms when the updraft reaches glaciation and the jet stream "blows off" the top of the tower and flattens it out. The overshooting top is indicative that the updraft is still active and healthy.

When visualizing a cap, just picture everything you saw on that day. Some weak Cu that can barely get going along with some high based cirrus clouds or even some mid level stratus. Once the sky was still looking like this at 6pm I became worried, at that point I decided to check meso analysis as well at the latest RUC to confirm my suspicions. I deemed it game over, abandoned my target and went to meet up with everybody else at the Flying J.

 

[Danny Neal]

Picture your garden variety Cu field. Low tops sometimes high based.

http://upload.wikimedia.org/wikipedia/commons/b/b5/Cumulus_clouds_in_fair_weather.jpeg

Picture above shows a cumulus. Picture pockets of these growing to a certain point and then collapsing over and over and over again.

Now look at what happens when one of these Cu breaks the cap shown below:

http://www.windows.ucar.edu/earth/Atmosphere/clouds/images/cumulus_cloud_di00168_big.jpg

Big difference?

Basically the cap is what suppresses convection, I.E rain/thunder/hail/storm in general. It is that warm layer above, that refuses to cool and allow those shallow cumulus to grow and condense vertically. Generally everything will appear to be flat.

 

[Jeff Snyder]

Steve,

The "cap" that we often refer to is a layer of warm air often at the base of an elevated mixed layer (i.e. around 850-750 mb). This warm air acts as a lid to prevent the warm, moist air below (i.e. near the surface) from rising. There are a couple of ways to remove, essentially, this cap -- you can heat the near-surface air (e.g. by insolation, etc.) or you can cool the capping layer (e.g. by large-scale ascent downstream of a shortwave trough, etc.).

Oftentimes, the initiation of surface-based thunderstorms (meaning thunderstorms whose updraft is largely comprised of air from near the surface) is a battle between the cap strength and low-level convergence. If you heat the low-level air enough to remove the cap, you only need very weak surface convergence most of the time to initiate deep moist convection. If you have strong surface forcing, such as a cold front plowing through, you'll often see widespread initiation (which usually isn't good for discrete supercells). On the other hand, if you have a stronger cap, only weak low-level convergence will not be able to "push" low-level air through the cap on a persistent manner. In such cases of a stronger cap, you need stronger low-level convergence.

The anvils that you are referring to occur when thunderstorms reach the equilibrium level (oftentimes the tropopause, but certainly not always!); above the EL, negative buoyancy (i.e. updraft colder than the environment) causes the updraft air to "slow down", which in turn results in upper-level convergence and horizontal divergence that you see as the anvil spreading. Note that the top of the thunderstorm is often quite a way ABOVE the EL, since the velocities of the updraft may be very high and may take a good time to slow down to 0 m/s.

Oh yeah, and note that this is essentially something that we call parcel theory, and it does not always "work". I can think of several chases (well, busts) that ended with no thunderstorms despite a nearby sounding showing little or no capping in the presence of at least weak surface convergence. For example, check out THIS 00z OUN sounding from 6/7/07, and note that there is relatively low CINH (less at a couple of nearby locales that were reporting higher temperature and dewpoint). There were some weak, brief storms that initiated very near that sounding location, but all failed to sustain themselves. Some have speculated that the failure of sustained supercells can be attributed to too limited a time that parcels remained in the zone of low-level convergence.

 

[Brett Roberts]

Hey all,

So I was on the same bust a lot of folks were on here in Iowa last Thursday. As I read back through some threads, I believe the long story short is that the cap just held a lot longer than expected. While I'm not 100% clear on the CAP just yet (save that one for later), I do have a question regarding the visualization of this taking place (IE, towers not being able to 'bust through' the cap).

As I understand it, CU's form anvils when their updraft bumps into the cap, so to say. Anvils are obviously pretty easy to identify. Then, when the cap gives way, you can get an overshooting top, which indicates there is some seriously impressive updrafts.

So my question is this. On a day like last Thursday, where I watched towers just go up and down all night, if it was the cap holding them back, why wouldn't I have seen anvils in this case? In other words, if the cap is holding, what conditions dictate whether the clouds actually form an anvil, or just fall back down again.

I know those are a lot of lamens terms, I hope I made it clear, sorry if this has already been addressed.

Thanks in advance for any help.

The short answer is that a "cap" exists in the low-levels of the atmosphere -- within or near the 850 mb to 700 mb layer, oftentimes -- and can prevent thunderstorms from forming at all. "Bubbles" of air may begin rising from the surface, but in the presence of a sufficiently-strong cap, those bubbles will decelerate and ultimately stop rising completely somewhere between 5,000-10,000 ft. above the surface (just as a rough idea). The result, which is probably what you observed, is that any "towers" will top off at a pitifully low altitude, and never even become thunderstorms at all. In fact, the "towers" in such a scenario may be no taller than they are wide.

Anvils, on the other hand, only exist with actual thunderstorms which have successfully penetrated any low-level cap (if there is one at all) and continued accelerating upward through the troposphere. They are the result of an updraft reaching the tropopause*, which is the boundary between the troposphere and stratosphere. The height of the tropopause varies over space and time, but is often more like 40,000-50,000 ft. above the surface -- far, far above the height of any low-level temperature inversion we would call a "cap."

Both the tropopause and the "cap" are examples of temperature inversions in the atmosphere (i.e., vertical layers of air in which temperature either increases with height or does not decrease very much with height). Both phenomena cause air parcels attempting to rise through them to slow down, stop, and eventually descend. The difference is that the "cap" exists much, much lower in the atmosphere, and any rising pockets of air that hit it haven't had nearly enough time accelerating upward to gather the speed necessary to form an overshooting top and/or anvil. In contrast, an air parcel which has been accelerating up through the entire depth of the troposphere (several tens of thousands of feet) in a powerful thunderstorm updraft will have a much higher upward vertical velocity; thus the much more impressive visual result!

*Technically, the air parcels will stop accelerating upward at the Equilibrium Level, but it's usually fairly close to the tropopause as far as I know. The definition of the EL is a bit harder to describe without getting into CAPE, buoyancy, parcel temperature deficits, etc.

Disclaimer: this is just what I know off the top of my head from my undergrad courses and independent reading, so if there are any inaccuracies, hopefully a professional will chime in.

 

[Steve Conley]

Ok, you guys are shedding a lot of light on some incorrect assumptions I've been operating under, the first of which was obviously a misunderstanding of what actually formed the anvil.

JEFF: There are a couple of ways to remove, essentially, this cap -- you can heat the near-surface air (e.g. by insolation, etc.) or you can cool the capping layer (e.g. by large-scale ascent downstream of a shortwave trough, etc.).

The warming of the near-surface air makes sense, and explains why my boss (who used to study meteorology) was so excited about the sun breaking through on this particular day. As to cooling the capping layer, would an upper level front be an example of something that would produce the cooling you're talking about?

JEFF: In such cases of a stronger cap, you need stronger low-level convergence.

This is because by introducing low-level convergence, you're also introducing additional "rising air", correct?

The visual aids helped out a lot guys, I appreciate everyone's help.

 

[Skip Talbot]

Keep in mind that last Thursday's chase was not technically a cap bust (unless your forecast was limited to a very local area). Both the 4km WRF and RUC showed storms firing at around 8 pm, and indeed storms did fire near Waterloo shortly after 8 pm. These storms became surface based, and one dominant one became supercellular and carried a few tornado warnings as it crossed the river into Wisconsin. Often times storms do develop after dark in a capped environment, but these storms form above or at the capping inversion. Their inflow does not come from the surface, and thus these storms typically do not produce tornadoes.

Although that chase was a bust for tornadoes, we did get a surface based supercell off. I think most of us were hoping storms would fire earlier so we'd have more daylight to work with, but indeed the models were correct (for once). The placement was also very awkward as the Mississippi and terrain on either side is a real bear to work with on a chase.

 

[Jeff Snyder]

As to cooling the capping layer, would an upper level front be an example of something that would produce the cooling you're talking about?

Steve -- cold-air advection can help cool the capping layer, but that's not terribly common on "normal" tornadic supercell chases on the Plains (again, though, it does happen on some chases). Associated with cold-air advection in this part of the world is a wind profile characterized by backing with height (e.g. winds that go from west to southwest to south as you go "up" in the atmosphere), which often creates wind profiles and hodographs that are not particularly favorable for cyclonic supercells.

More common is rising motion in response to differential positive vorticity advection (DPVA) downstream of a trough axis. This is probably more technical than you want, but just think "rising motion" when you read about an approaching shortwave trough or "vort max". There are also other processes that can create upward motion through some depth of the troposphere, including ageostrophic curvature divergence (which is mid- and upper-level divergence associated with the curvature of the flow downtream of a trough and upstream of a ridge), mid- and upper-level divergence associated with the transverse circulation in the quadrants of a jet streak (i.e. divergence in the right-entrance and left-exit regions), and so forth. The specific height of these "disturbances" and the static stability of the atmosphere (which you can think of by looking at lapse rates aloft ... The steeper the lapse rate, the less the change in potential temperature with height, and the lower the static stability) dictates the degree to and depth through which this forcing creates upward motion in the atmosphere.

This is because by introducing low-level convergence, you're also introducing additional "rising air", correct?

If convergence is maximized in the low-levels, you will see upward motion just above (i.e. upward motion above the level of max quasi-horizontal convergence). This convergence CAN help storm initiation by locally weakening the cap, and it helps provide an environment for parcels to be pushed up and through the capping layer. You can have steep low-level lapse rates all you want, but you need something to push those potentially unstable parcels upward; this push upward is done through low-level convergence.

 

[Steve Conley]

Ok, that makes sense Jeff, appreciate the less technical indications in there as well.

Skip, I agree that I shouldn't have called it a total bust, just a tornado bust. When I first started out, I thought tornadoes were the only thing I was after. Now that I've learned more about all the structures that go along with severe weather, I'm just as happy to sit back and enjoy the beauty that is weather.

Unfortunately, I was too far from Waterloo by the time it started firing, and figured I'd run out of light shortly after catching up with it, so I bailed. It turned out that holding up where I was (in the hopes of getting a wider look at a decent supercell), I basically took myself out of the running....live and learn.

Thanks again guys.

 

[Andy Wehrle]

While we're on the subject of the cap, here's something I'm wondering about:

How does the temperature at a given level (H7, for example) translate into j/kg CINH? Does it vary from day to day given how much CAPE is present?

 

[Chad Cowan]

While we're on the subject of the cap, here's something I'm wondering about:

How does the temperature at a given level (H7, for example) translate into j/kg CINH? Does it vary from day to day given how much CAPE is present?

The 700mb temp is usually a decent indicator for cap strength but for me, nothing beats looking at a sounding to see the amount of CINH present. Here's a Jon Davies page "Using 700mb temperatures as an estimation of the "cap" and to limit tornado potential": http://www.jondavies.net/700mbTcap/700mbTcap.htm

 

[Clarence Bennett]

The temperature at a given level (H7) translates into J/kg of CINH relative to the temperature of the air in the H7 environment and the parcel being lifted into H7. If a parcel of air lifted into H7 is greater in temperature, you go positive and can begin calculating CAPE. If that parcel of air lifted into H7 is cooler than the surrounding environment, you can begin to calculate CINH. CINH over 50 is going to lead to a moderate to strong cap.

Keep in mind that CINH is in the area from the surface to where positive CAPE begins. It is not necessarily isolated to a specific height.

I know this just scratches the surface, but in basic terms this is it.