Photo above shows a double maximum in AzShear. One of these is clearly colocated with a TOR on ground via CC. However, velocity shows two very different situations, possible strong tornado on ground versus convergence along the RFD gust front. If you were just looking at the AzShear product, both of these areas would be cause for concern, when in fact the impacts are extremely different. Would be nice to be able to differentiate between extreme high values of AzShear near a tornado versus more broad convergence on any wind shift area.
This next image shows a region where the AzShear maximum is too broad. This may simply be a color table issue, but clearly there is an area within the white region where a tornado is on the ground, and other regions within that max where only general convergence is occurring.
This image shows a location where some ground clutter is producing a false positive and negative AzShear couplet. Mathematically it makes sense to see this here given the V data, however to an untrained eye or someone just solely depending on AzShear you could be led astray.
A positive of AzShear is that it can highlight areas of potential concern down the road. The above frame was just one of several frames showing strong convergence and high AzShear values preceding a second tornado from the main cell 10 minutes later. This could be particularly helpful in QLCS storm modes where convergence and shear increases in a broad sense along the main line prior to tornadogenesis. Also in this particular case, the V data was not entirely conclusive that convergence was increasing. AzShear was very conclusive here.