Topic: Non-supercellular tornadoes

[Macca]

Hi all,

In light of the events which occurred in central Victoria last week, I've been doing some reading into non-supercellular tornadoes and the Denver Convergence Vorticity Zone "DCVZ" (a frequent convergence zone which sets up on the eastern side of the Rockies over Denver, Colorado during the summer months and has been known to produce quite a few non-supercellular tornadic events) in an effort to conceptualize the non-supercellular tornado event sighted by Clyve H in central Victoria last week. Photos of this event can be seen in the briefing room here:

www.weatherzone.com.au/cgi-bin/ultimatebb.cgi?ubb=get_topic;f=17;t=000222;p=3

A quick bit of searching on non-supercellular tornadoes found this...

"A new conceptual model for at least one type of non-supercell tornado grew out of the Colorado front range observations of tornadoes arising in ways quite reminiscent of the waterspout, leading to the term "landspout" as a way of denoting this
model. In this model, "misocyclones" (using Fujita's term) existed at low levels along some sort of boundary before convection even developed. In the conceptual model, these pre-existing, low-level misocyclones would be overridden by
developing updrafts, stretching the existing low-level vorticity to tornadic proportions. This suggests an upward development, seemingly quite distinct from the supercell paradigm." - C.Doswell (circa. 1996).

And a study of the Denver Convergence Vorticity Zone noted the following:

The convergence zones create these "misocyclones" at the surface which are quite small and short-lived (less than 2km and 5-15mins). Hot weather (and convergence) causes upward motion in the vicinity of this "misocyclone". Ultimately, the overall upward motion generates new convective development under which this miso-cyclone is now being stretched further as a result of the intense convection in the unstable environment. The result being a non-supercellular tornado (The Denver Cyclone and Tornadeos 25 Years Later: The continued challenge of forecasting non-supercellular tornadoes - Szoke, 2006).

This paper has a great illustration of this process and i'm trying to find a non-pdf version so I can post it here...but here is the link to the paper anyway so if you are interested you can have a look:

http://ams.confex.com/ams/pdfpapers/115160.pdf

A bit more digging brings me to some information from the NOAA on "favourable conditions leading to pulse storm tornadoes".

1. An environment with steep lapse rates, strong surface heating, and no CIN (Convective INhibition).

2. A well defined boundary (convergence zone).

3. A rapidly growing updraft situated close to or over the boundary.

These three conditions fit with the description of events noted above in Szoke's paper with the boundary/convergenze zone providing the potential for "misocyclones" to exist. The strong surface heating allows for upward motion in the vicinity of this convergence zone (and any existing "misocyclones") and the steep lapse rates and low/no CIN allow for intense convection to develop over or close to the convergence zone.

Let's now apply these ideas to the events which have occurred in central Victoria this past week. As you have read and seen on this forum, Clyve H witnessed a non-supercellular tornado on Thursday north of Melton and Lindsay ("Loo") Knowles observed a second non-supercellular tornado on Friday. I don't have much information on the event near Pakenham on Friday so I'll focus more on Thursday.

From what I can get from the posts on here, temperatures on Thursday (and Friday) were well into the mid-30's with convection initiating around midday-1pm on both days. So we have automatically satisfied one of the "favourable conditions" being strong surface heating.

Harald Richter commented on the austpacwx list "surface data show a wind shift line in the area providing the initial vertical vorticity for this stretching process" when referring to Thursday's event. This wind-shift line mayhave been sea-
breeze related. A quick look back at the observations from this day shows both Laverton and Sheoaks with generally SE'ly winds most of the day with temps of 30-32C and DP's in the low-mid teens (12-14C). Melbourne AP seemed to be sitting quite
close to the convergence zone with its wind shifting between SE'ly and NNW'ly a few times during the afternoon but the shift that catches my eye is the 1:00-1:30pm where the temperature rose by 1C but the DP dropped by 1.5C (indicating drier non-seabreeze air). Another interesting observation is Ballarat. Although it is elevated (approx. 500m ASL), it shows northerly winds for much of the day (and during the relevant time period of midday - 2pm) with temperatures of 30-32C (note it is slightly cooler due to its elevation) and DP's of around 10C. So without having a wind streamline chart to look at (unless
someone has one saved), as noted by Harald, it appears there was a convergence zone running in a general east-west line from near Melbourne AP across to near Bacchus Marsh. So we appear to have a well defined boundary (convergence zone) to generate these "misocyclones".

A quick look at an analysis sounding (thanks to BSCH) from 00Zon Thursday and re-plotted for the surrounding conditions (36/11) shows very steep lapse rates with 0-6km rates of 8.5C/km. Instability wasn't an issue either with LI's of about -4 and CAPE of 1600j/kg. Again we have satisfied more of the favourable conditions required for these non-supercellular tornadoes.

Clyve's brief report on the forum stated that "this spouty tornado appeared under a small developing storm...severe storms developed in this area over the next hour". From this report it appears there was some rapidly developing updrafts in the near vicinity to the event.

So...with all the information we have above, it appears that a weak non-supercellular tornado has occurred in quite text-book fashion (if you can get that with these events).

Loo may be able to provide some other details about the "Pakky" event (ie timing, conditions, other convection(?) in the vicinity) so as to be able to make an assessment if a similar process took place the following day. Given the similar set up's, I wouldn't be surprised to see this second event being quite similar in nature.

It also makes you think back to the Paraparap non-supercellular tornado reported by Jane ONeill on 9th January 2000. Here are some extracts from the report written by C.Herbert on this event...

"At between 1530 and 1550AEDST on the 9th January, 2000, a small compact multicell thunderstorm moved across the Paraparap

district 21kms southwest of Geelong & 10km southwest of Geelong Airport Automatic Weather Station. At approximately 1541AEDST (0441UTC), a small tornado developed along the short flanking line raising dust from ground level to cloud base."

"Although this small, compact multicell did not exhibit severe characteristics, the temperature profile may show that a steep potentially superadiabatic lapse rate may have existed between the surface and as high as 700hPa. There is also evidence of strong lapse rate potential between 500 and 300hPa."

"The nearest official BoM weather station is located approximately 10 kilometres northeast of the Paraparap district at Geelong Airport (Mt Duneed). At 1530AEDST this station reported an air temperature of 31.9C and a wind direction of 70degrees at 10 knots and appears to be effected by a weak seabreeze extending from Port Phillip Bay. At 1600AEDST, the wind had shifted to 130degrees, which may indicate the arrival of a southeast sea breeze extending from Bass Strait (common in this district)."

The full Paraparap tornado report can be found here:

www.stormchasers.au.com/paraindex.htm

Again, we have the "ingredients" required for a non-supercellular tornado as set out above - steep lapse rates, strong instability, convergence zone and rapidly developing updrafts.

Perhaps this is something we need to be considering in future events where we are expecting sea breeze convergence to initiate storms in an unstable environment that is favourable for non-supercellular tornadoes. It would not be at all inconceivable for the sea-breeze convergence zone to be sitting over the outer northern suburbs of Melbourne on days such as what we have just seen, meaning these weak tornadic events could occur over populated areas of Melbourne. It seems as though we have something similar to the Denver Convergence Vorticity Zone on days such as this. Perhaps we should call it the Melbourne Seabreeze Convergence Zone ("MSCZ"). Definitely something to keep in the back of ones mind on a chase day with
moderate-strong instability, low shear, hot temperatures and the MSCZ.

Macca

[David C]

Definitely Macca, it is worth chasers noting this info - in particular having an idea on the location and movement of 'the boundary' when the thermodynamic environment is favourable. Slow moving or stationary boundaries (or slow movement of the updrafts/incipient cells relative to the boundary) are favoured in that the processes are afforded more time.

Judging by the photos in this case, the cumulus/congestus were quite shallow, just going by what I can see in the photo, I'd imagine that the key to this event (in addition to the convergence boundary) would have been very steep low-level lapse rates (eg surface up to a few kms), allowing for rapid upmotion of parcels under the updraft.

Here is another example of a nice landspout from a congestus (South Australia)

[Jeff Brislane]

I would like to see a video clip which shows base connection. To me from the photos it's looks like just a dust devil which would not be un common in a hot enviroment like that.

[Jimmy Deguara]

Hi Jeff,

I cannot comment any further given that the photos were not very clear. Having said that, a quote from Harald Richter and I guess to a certain extent reiterated by David Croan was that the tight structure of the feature leads it to be more a landspout than a dust devil. If only there were clearer pictures.... David must have good eye sight seeing as he can note a funnel in one of the pictures. Harald also confirmed a boundary in the area that is one of the ingredients coinciding with the development of landspouts. Landspouts would not be uncommon in this country I guess as compared to the mesocyclonic cousin.

I cannot say I am totally convinced it is one nor the other despite the tight struture. Consider the Canberra dust devil video from a few years ago at the Australian Storm Chasers meeting. The video shows a very tight structure and also rapid upmotion. So really can we lean between one or the other? I wasn't there so cannot comment beyond this point.

Regards,

Jimmy Deguara

[Michael Thomas]

Correct me if I am wrong but a common ingredient for landspout tornadoes, waterspouts and dust devils is steep low level lapse rates, potentially even super-adiabatic lapse rates (where the air cools faster than the dry adiabatic lapse rate with increasing altitude). So for landspouts and dust devils strong solar heating of the surface can quite easily give super-adiabatic lapse rates while for waterspouts it is heating of air by warm water. Given this it would seem that a difference between landspouts and dust devils is the amount of lower level moisture. So if the same processes to form a large dust devil (ie.air moving vertically rapidly) where to occur on a day with higher surface relative humdity then condensation is going to occur at the LCL and you would get a landspout instead.

A bit simplistic but hopefully its clear what I'm trying to say.

Michael

[Jimmy Deguara]

Hi Michael,

Given it is the case that relative humidity separates the landspout and dust devil dynamics in regions with steep lapse rates, what are your conclusions in this particular case? If I recall, there was a mention that this boundary may have been a sea breeze boundary. The cumulus were very high based so the air was either mixing out or was generally in an environment with relatively low relative humidities. But in proximity with a sea breeze boundary, the relative humidity increases dramatically.

I have by the way experienced dust devils with the onset of the sea breeze front but perhaps not necessarily along it.

Regards,

Jimmy Deguara

[David C]

I take back my comments on this particular dust-devil 'looking' like a tornado. This was based on my own observations of dust devils only. After actually looking around, there are some impressive examples of dust devils around which look more like a 'tornado' than the vic example eg http://www.wrh.noaa.gov/fgz/science/dustdvl.php?wfo=fgz 

The only other thing that others may or may not have observed with dust devils is the dust 'cloud' which demaracates the top of the vortex where dust / debris cloud becomes diffuse and is no longer lifted (see above example). Then again, this is not always evident with dust-devils (and not evident in the Vic pics) so is hardly definitive.

I think it is reasonable to conclude that vortex appearance cannot be used to distinguish between these phenomena at all.

Regarding the funnel, when I looked at the photos initially I thought I could see a slight needle protrusion at cloud base in one of the photos which indicated a possible funnel - I can't see it now (at least on this mac with the gamma settings out).