Detection

When the direction of storm propagation is different from the steering wind, it is a signature of the potential for severe convection. Most thunderstorms typically move with the steering wind, calculated as a pressure–weighted mean of the environmental horizontal winds over the depth of the storm. Often, mid–level winds are used as an approximation for the steering wind because the storm’s centre of mass is located there. A storm’s motion could be anomalous in direction and/or speed. To determine if you are seeing a Anomalous Storm Propagation signature, follow the techniques illustrated in the sections below.

Loop of a supercell storm splitting into a dominant left–mover and smaller right–mover. Both of these storms move differently to the steering flow after splitting. It is important to make sure you are tracking the same part of the storm when assessing anomalous storm motion.

Reflectivity: PPI/Plan View

To determine the movement of a storm you need to track the same part of the storm. Usually the easiest part to track is the storm top, as described below.

  1. Loop the latest few scans (say 20–30 minutes worth, weighting towards the most recent scans, due to the possibility that the storm has only just started to move anomalously).
  2. Is the storm moving differently to other thunderstorms or showers in the area? If so, your storm is moving anomalously to steering flow.
  3. If the storm is the only observable feature on radar, a comparison between the storm motion and a representative observed steering wind should be undertaken. The representative steering wind could be obtained from a number of different sources such as a recent sounding, a wind profiler, AMDAR or even the velocity data if your radar is a Doppler radar. If the storm is moving differently to the steering wind, your storm is propagating anomalously.

To find the storm motion vector at a given time, perform the following:

  1. Step back three or four volume scans.
  2. Step up in elevation until you run out of >50 dBZ echoes.
  3. Step down one or two elevation scans, as the storm top may rise or fall over the assessment period.
  4. Centre cursor on the >50 dBZ core.
  5. Set a reference point.
  6. Step forward to the most recent scan
  7. Place your cursor on the >50 dBZ core.
  8. Read off the speed and direction off the cursor data window of your radar display software.

Potential Difficulties in Detection

Storms move anomalously for many different reasons. For severity determinations we are looking primarily for storms that propagate anomalously due to updraft interaction with deep-layer vertical shear, so you may need to confirm the storm isn’t moving anomalously due to the following reasons:

  • Cold pool/shear interactions
  • Storm mergers
  • Boundaries (drylines, fronts, outflow boundaries)
  • Orographic influences
  • Radar sampling – The thunderstorm is too close to the radar, therefore the radar undershoots the storm top used for tracking. It may still be possible to use other recognisable features to track the storm
  • Short storm history – Storms that have just appeared on radar are too young to allow the determination of a storm motion vector due to lack of past storm locations.
  • Very large storms, especially with multiple updrafts – in this situation, it is often difficult to reliably track a persistent feature

Examples of Anomalous Storm Propagation

Loop of a supercell storm splitting into a dominant left–mover and smaller right–mover. Both of these storms move anomalously from steering winds after splitting. It is important to make sure you are tracking the same part of the storm when assessing anomalous storm motion.

The smaller storms move to the east with the steering flow. The two larger storms, one to the northwest and the other in the south are propagating to the north/northeast, that is to the left of the steering flow vector.

The storms initially move to the east with steering, then propagate to the north (to the left of the steering flow vector).

Anomalous Storm Propagation Look–a–likes

  • Elevated core from a separate updraft – More often than not, storms are multicellular in nature with several updrafts co–located in a storm. Sometimes this can lead to mistakenly using different updrafts to calculate the anomalous motion vector.
  • Thunderstorms propagating due to reasons other than updraft/deep layer shear interactions, as mentioned above (these include cold pool interactions, storm mergers, boundary interactions, orographic influences).