Sunday, July 01, 2012

What's CAPE


A reader (The Saguaro) has asked in a comment about CAPE. It's complicated, but I'll take a shot at an explanation. CAPE stands for Convective Available Potential Energy and is therefore a measure of energy that could possibly be realized within deep, convective clouds.


First, let's consider buoyancy. If you hold a cork under water (as per one of the wine bottle corks above) in your sink or bathtub and then release it, it rises quickly to the surface. The cork has buoyancy because it is less dense than the water. It rises because its buoyancy is greater than gravity, which would like to pull the cork down to the bottom.


Or, consider hot-air balloons, as seen above near Shiprock, New Mexico. Such a balloon is filled with ambient air, but the air has been heated at the burner. Thus, the balloon is filled with air that is less dense than the outside air - when the buoyancy of the balloon and its cargo exceeds gravity's pull, the balloon rises. CAPE is a quantification of the potential buoyant energy that an air parcel could have in the atmosphere.


CAPE is usually estimated using a plot of an upper-air sounding on a thermodynamic diagram. There are many types of such diagrams, but I show skewT/log P diagrams (as above) on this blog. The skewT/logP diagram is an equal area diagram - this means that equal areas on the diagram represent the same amount of energy. An estimate of CAPE usually is made via computer software and displayed along with the sounding - see "CAPE" above at right - red shading. In fact, such software usually computes a large number of various atmospheric parameters. It is important to understand how any given set of sounding analysis software is actually calculating CAPE, since the assumptions used to build the software will be reflected in the output.

How does an air parcel become less dense than the surrounding environment? As an air parcel rises and becomes saturated, further lifting causes water vapor to condense. This releases the latent heat of condensation, which warms the air, allowing a parcel to sometimes become less dense than the ambient air. Thus, the phase changes of water are very important for convective processes and thunderstorms.

Consider the sounding plotted above - it is a morning sounding, but there is a well-mixed layer above the radiation-cooled surface layer. I have estimated CAPE for this elevated layer. The lifted parcel is shown and the CAPE area is shaded red. Units of CAPE are J/kg or m**2/s**2. Other levels identified above: LCL is the Lifted Condensation Level (the level at which an air parcel becomes saturated and clouds form); LFC is the Level of Free Convection (the density of a rising air parcel is equal to that of the ambient air); and EL is the Equilibrium Level (the density of a rising air parcel again equals that of the ambient air). The LCL is the height of cloud base [note that air from near the surface can have a lower cloud base than that of the well-mixed boundary layer (BL)]. This is the reason for the distinct lowered structure of a wall cloud. The EL is the level at which the cumulonimbus anvil will spread out.

Two other areas are shaded blue. CIN is Convective Inhibition and results when a layer exists where a rising air parcel has negative buoyancy (i.e., would sink due to gravity unless some process is forcing it upward against gravity). The blue area above the EL is another layer where an air parcel would have negative buoyancy. An air parcel will overshoot this level due to its momentum and continue to rise until its vertical velocity is zero, and then sink back toward the anvil layer.

The extreme challenge of evaluating CAPE lies in the fact that soundings taken at a single time (usually either 12 or 00 UTC) must be used to forecast the subcloud environment's thermodynamic conditions immediately before deep convection develops. Thus, forecasters today usually rely upon model-predicted soundings to evaluate the convective storm environment. Even high-resolution models can have difficulty accurately predicting the time-evolution of the BL. Finally, the larger the CAPE the more intense will be the storm (assuming storms develop); the larger the CIN the more intense must be local forcing for upward motion before a parcel can be lifted to the LFC.

Whew - I said it was complicated!

3 comments:

  1. It's even a bit more complicated than that. Buoyancy ia an unbalanced pressure gradient force that is due to spatial variations in density. See:

    Doswell, C.A. III, and P.M. Markowski, 2004: Is buoyancy a relative quantity? Mon. Wea. Rev., 132, 853-863.

    The "model" of deep convection as represented by the parcel path on a sounding diagram is not a very good representation of what is actually going on ... parcel buoyancy is not determined by a comparing the parcel temperature to its "environment". How would a parcel know its density compared to its surroundings? Where in the surroundings would it look? The reality is that differences in density result in changes to the vertical (perturbation) pressure gradient force, and the parcel responds to that.

    which can be downloaded free from my "publications" Webpage. Kerry Emanual's book on Convection, and Bob Houze's book both give proper explanations, but these aren't widely understood because we've all been taught the "hot air balloon" argument. Archimedean buoyancy is OK, but it doesn't tell the whole story re buoyancy for air parcel.s

    For air parcels, there is no well-defined boundary between it and its environment - air parcels have no definite size or boundary. As a matter of fact, the distinction between a parcel and its environment is entirely artificial. All air responds to the forces described in the equations of motion, whether it's in a cloud or miles away from that cloud.

    ReplyDelete
  2. Sorry .. the paragraph order got jumbled in the preceding! I don't think I can fix it.

    ReplyDelete
  3. Sorry again ... Yes, I know you're wanting a non-technical explanation, but the hot-air balloon model is just not a very good one. I'll try to give the notions in our paper a bit more thought in terms of a layperson's explanation.

    ReplyDelete