Damaging Downburst Prediction and Detection Algorithm (DDPDA)

Theory and History

There are three types of straight-line wind events we are concerned with for the long-term. These include:

Outflows produced by "pulse" thunderstorms ("wet" downbursts)
"Dry" downbursts usually produced by shallow, high-based cumulonimbi
Strong straight-line winds produced by bow echoes

This year, we are primarily concerned with #1: pulse thunderstorm outflows. In fact, this has been most of the work on the algorithm so far.

1. Pulse Thunderstorms

A lot of work has been done in this area, and numerous studies have found similar precursors to damaging wind events at the surface. The more prominent precursors in pulse thunderstorms include:

A high-reflectivity core which begins at a greater elevation than other, non-severe, cells in the same environment;
This high-reflectivity core descends rapidly;
Strong mid-altitude (between about 2-7 km AGL) convergence into the cell.

Eilts et al have shown that the strength of this convergence has a high correlation with the strength of the divergence of the outflow.
Caused by evaporation, melting, and precipitation drag.

Other precursors which have been shown to exist but we believe to be of lesser value for downburst prediction include:

trend and strength of storm-top divergence
rotation within the descending reflectivity core.

These downbursts usually occur in an environment with a lot of low-level moisture and drier mid-level air (at 600-400mb or so).

2. Dry downbursts

These are extremely difficult to detect, as the cells that produce them are usually too weak to be detected by the SCIT algorithm. The greatest number of these occur over high terrain with a very dry lower troposphere and mid-level moisture ("inverted-V" sounding)

A good example of how difficult it is to detect these events can be found in a write-up by Steve Vasiloff called Severe microburst event in Utah. (NWS WR Tech Memo)

Possible future work on the DDPDA may involve incorporating near-storm environment data with a lower-threshold SCIT algorithm and/or satellite data to produce guidance for dry downbursts.

["wet" versus "dry" downbursts -- high-reflectivity downbursts in Arizona occur when there is a much drier lapse rate below 700 mb than with those downburst which occur in Florida]

3. Bow echoes

A substantial percentage of the reports of wind damage in Storm Data are probably caused by bow echoes. Ron Przbylinski (SOO, NWS STL) has collected several bow echo events and found some correlation between the strength of mid-altitude convergence and severe wind gusts at the surface.

Next page
Table of contents