Storm speed ?

[Glen Romine]

For what it's worth - the Bunker's technique really isn't that complicated either from a graphical perspective. If you have a hodograph that includes height markers - you just draw a line from the mean 0-500 m wind to the 5.5-6km wind, then find the middle of the hodograph - and draw a line perpendicular to the first line you drew that crosses through the point on the middle of the hodograph. Finally, moving 7.5 m/s to the left or right of this crossing point along the second line gives the left or right mover estimated cell motion - with the point in the middle for the non-rotating cell motion. Of course - you rarely have a hodograph for the location and time you expect a storm to develop - but you can look at a forecast sounding for the target location and then apply this technique to estimate what cell motions to expect. For the graphically minded, I've included a sample below.

So here, a right mover (labelled R) would be expected to track NNE at ~18 m/s.

Glen

 

[Gene Moore]

I've used the Bunker's technique and found that supercells, especially if they are not in a strong dynamic environment, (strong winds aloft) go through a range of motions. Trying to pin down a supercell's motion (as in the example) to a given set of motions may lead to chase frustration. This is much like trying to catagorize a storm as LP, Classic and HP. Fact is many storms go through stages like LP to Classic and later HP. During these times the storm may initially move fast to the northeast, slowing during the classic phase and finally grinding to a stop or turning completely south during the HP mode. The causes of this were outlined in a earlier post and may result from increased convergence due to outflow, propagation, backbuilding or merging cells. All these factors may impact a supercell's motion, especially if it last 4-6 hours. As a chaser in the field I believe a storm's change in motion can be anticipated for both photography and safety's sake (large hail). The main clues are associated with the character of the inflow or flanking lines and the appearance of the core, or main precip mass. It's not uncommon for chasers to get overrun by the forward flank hail shaft as the cell makes its initial deviation to the right. I think schemes like Bunkers work best with situations like a strong pushing dryline where a storm's motion is most likely dominated by a strong synoptic feature.....as opposed to a lone supercell that may be enfluenced more by "storm scale" motions. Regardless, it's good to have all these techniques in your head so you can anticipate the storms changes before you get overrun, or worse yet, left behind.

 

[Jeff Snyder]

For those who are interested:

Edwards, R., R.L. Thompson and J.A. Hart, 2002: Verification of Supercell Motion Forecasting Techniques. Preprints, 21st Conf. Severe Local Storms, San Antonio. [125K PDF]
http://www.spc.noaa.gov/publications/edwar...ards/motion.pdf

Corfidi, S.F., 1998: Forecasting MCS Mode and Motion. Preprints, 19th Conf. Severe Local Storms, Minneapolis MN [39K HTML]
http://www.spc.noaa.gov/publications/corfi...di/mcs_mode.htm

Corfidi, S.F., 2003: Cold Pools and MCS Propagation: Forecasting the Motion of Downwind-Developing MCSs. Published in Weather and Forecasting. [3156K PDF]
http://www.spc.noaa.gov/publications/corfi...idi/mcs2003.pdf


As others have mentioned, forecasting supercell motion is difficult in many instances. Unfortunately, one of the primary supercell and tornado forecast parameters -- Storm-Relative Helicity -- is extremely dependent upon supercell motion. Thus, one can see how model forecasts or SPC/RUC mesoanalysis can be in large error if the storm motion is hightly deviant. In other words, just because you see the NAM only forecasting 100 m2/s2 0-3km SRH does not mean that there is no risk of tornadoes... If there's still sufficient deep-layer shear for supercells, there may be highly deviant storm motion which may markedly increase the SRH for that storm.

 

[Glen Romine]

Originally posted by Gene Moore
Trying to pin down a supercell's motion (as in the example) to a given set of motions may lead to chase frustration.

I wouldn't disagree with you there - but of course we usually are only concerned with particularly cases where keeping up with a storm simply may not be practical. Storm environments yeilding slow storm motions are really not much trouble all even if they deviate greatly from what you expect in terms of direction from the perspective of chase positioning. It's the cases where the expected storm direction deviates greatly from the available road network when combined with fast motions (e.g., E-W roads, ne @ 45 knt storm motion is bad - but with a more easterly (or northerly) cell motion you might have a better chance of sticking with a storm). I think in these cases, Bunker's offers a decent ballpark estimation of what you might encounter - and really it's much better than having no clue at all - and can be a tool for deciding if an event is 'chaseable'. As for storm type - I've generally found the storm-relative mid-level winds to be a pretty reasonable predictor. Of course not perfect - but HP's are often well predicted with forecasts of very weak mid-level storm-relative winds - which you can't get without a storm motion estimate to substract from the midlevel winds. As you suggested though - in the field once you have a storm you should forget the forecast storm motion and pay attention to what the real thing is doing to avoid costly positioning errors.

Glen

 

[Gene Moore]

We are in agreement, although I've seen HP's develop in some strong wind environments, that is ....with over 100 knots aloft. One such case was May 25, 97. We had a slow moving tornadic supercell tied to the warm front. And, that might be part of the key, the source of convergence and a boundary. I've seen lumbering supercells with wind profiles that should have yielded 30-40 knot movements especially along warm fronts and OFB's. Again, these slower situations are least likely to occur where storms form on a pushing boundary like the recent dryline bulge in Kansas.

 

[Glen Romine]

Getting a bit off topic, but has anyone ever used the Stensrud et al. (1997) BRN shear as a storm type predictor? The bonus to using this is supposed to be independence from needing storm motion - which is needed to find storm type following the Thompson (1998) storm-relative mid-level winds. The former suggested BRN shear values of 40-100 m^2/s^2 were supposed to be most favorable for tornadic storms (when all other conditions favor storms to develop). I've generally trusted the storm-relative wind forecasts - but maybe I should be conferring with the BRN shear values as well.

Glen