Season: Every day
Phenomena: Dry Circulation, Vorticity Minima, and Vertical Velocity
Tools: GOES water vapour
Vorticity minima signatures are every bit as common as vorticity maxima signatures. They indicate areas of descending circulation and atmospheric forcing and can be used to diagnose dynamic features such as the axis of maximum winds and deformation zones.
Note: All examples and conceptual models in this module are set in the northern hemisphere. All graphics are oriented north to south with north on top.
The shape of the anticyclonic comma pattern reveals the location of the vorticity minimum. The vorticity minimum is located at the point of inflection and is the center of the cloud rotation in the atmospheric frame of reference. The shape of the concave inflow and convex outflow arcs is related to both the relative intensity of the vorticity minimum and the length of time the vorticity minimum has been acting on these arcs. The arcs become more concave or convex with both vorticity minimum intensity and age.
The vorticity minimum (N) is the addition of both rotational vorticity (R) and horizontal wind shear (S)—hereinafter simply referred to as shear. The following graphics illustrate the concept. The size of the symbols indicates relative intensity.
Given pure rotation, this conceptual cloud line will evolve in symmetrical fashion. The position of the vorticity minimum is at the center of the rotation, which is also the point of inflection. The rotational vorticity is the sole component of vorticity. There is no shear vorticity component.
With increased shear on the equatorial side of the rotation, the point of inflection and the total vorticity is shifted towards that region of greater shear. As a result, the concave arc is enhanced by stronger relative flow. The convex arc is not enhanced and is created purely by the rotational component of the vorticity, which remains unchanged and located at the center of the circle. The point of inflection is still the center of rotation. It is also the location of the total vorticity minimum resulting from the sum of the rotational and shear vorticities.
To summarize, this point of inflection is an anticyclonic outward cusp formed by the intersection of two anticyclonically curved arcs directed toward the cusp. Typically, an anticyclonic outward cusp is created by easterly speed shear that is equatorward of the cusp and marks the location of a vorticity minimum.
With even greater shear on the equatorial side, the shear component of the vorticity increases and there is more displacement. The total vorticity and point of inflection shift toward the region of shear and away from the position of the rotational vorticity centre. The concave arc is again enhanced with stronger relative flow.
If the shear is on the poleward side of the rotation, the same shifting of the point of inflection takes place, again with the center of total vorticity shifting towards the region of shear. In contrast, however, the convex arc in the location of the shear wind maximum is enhanced. The concave arc is not enhanced and is created purely by the rotational component of the vorticity, which remains unchanged and located at the center of the circle. The vorticity minimum will be located within the moisture area as opposed to the previous illustrations where it is located in the dry portion of the circulation.
Poleward shear comma clouds are common with the northeast trade winds in the tropics.
To summarize, this point of inflection is an anticyclonic inward cusp formed by the intersection of two anticyclonically curved arcs directed toward the cusp. Typically, an anticyclonic inward cusp is created by westerly speed shear that is poleward of the cusp and marks the location of a vorticity minimum. This pattern is very common due to the prevailing westerly circulations around the globe.
It's important to note that in these idealized examples the moisture extends right to the vorticity centre. In the real atmosphere the moisture may dry out near the circulation centre, especially as the circulation ages. As a result the actual point of inflection may be displaced upstream from the apparent point of inflection revealed by the moisture patterns.
Roger Weldon originally identified the "comma" cloud pattern in the early 1980's. The anticomma pattern is a mirror interpretation of that. Weldon showed correlations between the clouds, streamlines, and absolute vorticity isopleths. Note that absolute vorticity isopleths are used as an operational surrogate for system relative streamlines. Variation in cloud heights from 500-hPa and time differences between the analysis and the satellite data can lead to errors. The centre of cloud rotation, as identified by the point of inflection, is the vorticity minimum to a very close operational approximation.
The concave arc associated with the shear wind maximum defines this type of anticomma. The anticomma head is downstream from the vorticity minimum. Negative vorticity advection will be strongest in the anticomma head region.
The convex arc associated with the poleward wind maximum defines this type of anticomma. The concave arc region is minimal and the point of inflection is within the moisture area. The anticomma head is still downstream from the vorticity minimum.
In the following interaction, take a look at the GOES water vapor loop, then analyze it for vorticity minima, vorticity maxima, and the axes of maximum winds. After identifying these features, compare your analysis to the one provided.
Look for vorticity minima in this GOES 4-km water vapour loop
After viewing the loop, click the Your Analysis tab to access the last image of the loop (1115 UTC) and mark the location of the vorticity minima, vorticity maxima, and axes of maximum winds.
The following exercise cannot be completed in Internet Explorer 11 or older versions. Please use Microsoft Edge or other modern browsers to complete this exercise.
Mark the location of the most prominent vorticity minima, vorticity maxima, and axes of maximum winds within the highlighted area.
Click the Done button to compare your analysis to the "Expert's Analysis".
Tool: | Tool Size: | Color: |
---|---|---|
|
|
The most prominent anticomma formation is N1. The formation's classic anticomma shape is comprised of convective cloud that has smudged together.
There is an older vorticity minimum, N2, just barely visible to the northeast and a younger one, N3, developing to the southwest.
All are to the right of the flow looking downstream.
A vorticity maximum, X, is located to the left of the flow looking downstream.
Let's take a closer look at this situation. First we'll look at the large scale atmospheric circulations, then we'll focus on the main anticomma and vorticity minimum and finally, we'll take a closer look at the moisture patterns associated with the series of anticommas.
South of the main deformation zone, the flow is southerly. North of the deformation zone, the general flow meanders from the west. There are eddies in both of these general flows. The anticyclonic eddies are confined in the regions to the right of the general flow looking downstream. The cyclonic eddies are confined in areas to the left of the general flow looking downstream.
Let's focus in on the largest anticomma. Some key points:
From a moisture edge perspective, in order to identify the vorticity minimum, you are typically interested in the concave or inflow arc of the anticyclonic pattern.
N1
This pattern happens to be comprised of well-developed convective cloud. The cirrus anvils of the convection have blended together to reveal the circulations at the cirrus level that would otherwise be difficult to discern. The convection is present as a result of upper divergence upstream from the vorticity minimum.
This is good example of the anticomma with the moisture edge wrapping around the anticyclonic center of rotation. The point of inflection separates the inflow concave arc and the outflow convex moisture arc. The shape of the inflow concave arc is used to locate the vorticity minimum. The best position of the vorticity minimum can be estimated by visualizing the oval shaped circulation that best fits the inside moisture edge near the point of inflection. The shape of this inflow arc is molded by the rotation. The vorticity centre is certainly within the “best fit” oval and most likely at the point of inflection.
N2
The same principles apply to identifying this anticomma although the circulation is much drier and more difficult to discern. Note the greater depth of the concave inflow arc. If the vorticity minima are all of similar strengths (which is a good approximation since they are all the result of the same relatively straight axis of maximum winds), then this circulation has been acting on the concave inflow arc of N2 longer than either of the others. This implies that the associated vorticity minimum is the oldest of the three and the most likely to be identified correctly by NWP.
N3
Anticomma N3 can be identified solely by convection. In a similar discussion to the above, this vorticity minimum is the youngest of the three and is the least likely to be well identified by NWP.
Notice that the wavelength between the vorticity minima is quite uniform. This is because the axis of maximum winds generating the vorticity minima is relatively straight and because of the cyclical nature of short waves in the atmosphere.
In this case, the NWP and observed vorticity minima correspond fairly well. The observed location of the vorticity minimum is based on the point of inflection at the atmospheric level of the moisture. This moisture level may not correspond exactly with the pressure level of the NWP analysis (500 hPa in this case). This mismatch in level could result in a discrepancy between the observed location of the vorticity minimum and the NWP position.
The NWP vorticity minimum corresponding to the observed N1 is very well placed, as would be expected with such a well-organized anticomma pattern. The position and strength of the circulation have been well integrated into this model run. Vorticity minima that have been correctly analyzed by NWP in previous analyses are more likely to be correctly analyzed in the current and future analyses. Thus older vorticity minima are more likely to be correctly analyzed than new vorticity minima which are just spinning up.
The same is true with N2, a relatively older vorticity minimum. But as the system moves further from the land area and data sensors used in model initialization, there is more likelihood that features such as this will be lost. Here the blue highlight is the implied vorticity minimum that must exist between the two analyzed maxima. The accuracy of the NWP positioning of the vorticity minimum depends upon the amount and quality of the observational data in the area., Oceanic areas are typically data sparse so that the vorticity minimum over oceans can be misplaced through a lack of data.
Using the satellite animation, N3 is placed near the leading edge of new convection. The model inappropriately places the analyzed vorticity minimum upstream from the convection. The NWP error in this case is best illustrated by the NWP vorticity maximum lobe which attempts to include the two separate anti-commas evident on satellite imagery, into a single pattern. The NWP fails to resolve the two separate and distinct real vorticity maximum centres and instead creates a single vorticity maximum centrally located within an elongated vorticity maximum lobe. This is most likely due to model resolution issues.
The correct placement of the vorticity minima is vital to the placement of related dynamic features such as the axis of maximum winds and deformation zones. All of these dynamic features must fit cohesively into the atmospheric puzzle. The correct placement of a vorticity minimum might be the piece of the puzzle required to correctly identify and place other dynamic features.
Here are some important attributes of vorticity minima to keep in mind when identifying them:
Here are a few examples of vorticity minima. As is often true on the forecast desk, cases are not always as straightforward as the conceptual models. These real-world cases may not be the best examples of vorticity minima, but then, how often do you see any perfect examples at the desk?
Carefully examine the loop before doing the analysis. Be sure to compare your analysis to the ones provided.
Focus in on the highlighted area of this GOES 4-km water vapour loop. Look for the well-developed vorticity minimum with a a signature showing very little shearing influences.
After viewing the loop, click the Your Analysis tab to access the last image of the loop (2130 UTC) and mark the location of the most prominent vorticity minimum, vorticity maximum, and axis of maximum winds.
The following exercise cannot be completed in Internet Explorer 11 or older versions. Please use Microsoft Edge or other modern browsers to complete this exercise.
Mark the location of the most prominent vorticity minimum, vorticity maxima, and axis of maximum winds within the highlighted area.
Click the Done button to compare your analysis to the "Expert's Analysis".
Tool: | Tool Size: | Color: |
---|---|---|
|
|
The most prominent anti-comma formation is well displaced to the right of the axis of maximum winds. Shear influences are minimal and the vorticity minimum is the result of rotation.
The GFS180 500-hPa absolute vorticity for 00-hr 0000 UTC 28 July, which is a bit later in time than the satellite image, is in decent agreement with the placement of the vorticity centres. The vorticity maximum is downstream as anticipated. And also as expected, the vorticity minimum is not as well placed.
Carefully examine the loop before doing the analysis. Be sure to compare your analysis to the ones provided.
There is a subtle but fairly well-developed vorticity minimum in this GOES 4-km water vapour loop. It is showing the signature of having equatorial shear.
After viewing the loop, click the Your Analysis tab to access the last image of the loop (1730 UTC) and mark the location of the most prominent vorticity minimum, vorticity maximum, and axis of maximum winds.
The following exercise cannot be completed in Internet Explorer 11 or older versions. Please use Microsoft Edge or other modern browsers to complete this exercise.
Mark the location of the most prominent vorticity minimum, vorticity maxima, and axis of maximum winds within the highlighted area.
Click the Done button to compare your analysis to the "Expert's Analysis".
Tool: | Tool Size: | Color: |
---|---|---|
|
|
Carefully examine the loop before doing the analysis. Be sure to compare your analysis to the ones provided.
Within the highlighted area of this GOES 4-km water vapor loop, look for a subtle vorticity minimum with the slight signature of having poleward shear.
After viewing the loop, click the Your Analysis tab to access the last image of the loop (1230 UTC) and mark the location of the most prominent vorticity minimum, vorticity maximum, and axis of maximum winds.
The following exercise cannot be completed in Internet Explorer 11 or older versions. Please use Microsoft Edge or other modern browsers to complete this exercise.
Mark the location of the most prominent vorticity minimum, vorticity maxima, and axis of maximum winds within the highlighted area.
Click the Done button to compare your analysis to the "Expert's Analysis".
Tool: | Tool Size: | Color: |
---|---|---|
|
|
This concludes Vorticity Minima and Anticomma Patterns, please take a few minutes to take the Quiz, and then fill out our User Survey.