Both storms in the middle portion of the screen very quickly showed low min flash area (MFA), again bringing my attention to those storms. The northern most storm quickly developed classic hail signatures prompting a warning. The southern storm quickly fell apart.
Sandor Clegane
Month: April 2019
SJT SVR #1
GLM minimum flash area (bottom left) rapidly fell from ~1600 km^2 to 72 km^2 over a period of 5 minutes in the storm in the NE portions of these images. Maximum Total Optical Energy (TOE) topped out at 228 fJ shortly after the second image below. 
5 minutes later, this storm began displaying classic hail signatures in all-tilts base radar data, prompting a severe thunderstorm warning. I will note that several other storms displayed similar minimum flash area decreases without so quickly turning severe, but the rapid decrease was the first thing that brought attention to this storm.
Severe hail very likely occurred shortly after the warning decision. Shortly thereafter, Max TOE fell off fairly dramatically, to around 50 fj by the end of this loop. However, the difference between the pre and post severe max TOE does not show up well with the default color curve.
Sandor Clegane
Minimum Flash Area: Capturing the “Pulse”
Taking a look at the Minimum Flash Area (MFA) product (part of the GLM suite) and comparing it to the Average Flash Area (AFA), my initial analysis leads me to think that there’s more value in the MFA as opposed to the AFA. The MFA, when watching the evolution of a storm in the far northeastern part of SJT’s CWA, shows the pulsiness rather well. The AFA, in comparison, had a washed-out look during the 19Z hour of analysis. I had a side-by-side of the meso-sector VIS satellite imagery with the MFA on top (1-min.) and the ProbSevere suite with radar products, including reflectivity (GIFs below). The MFA provided a few minute lead time over the ProbSevere and reflectivity products of the storm pulsing up. Thought: watching the trend of the MFA could provide some insight into the overall nature of the storm. In this particular case, the smaller flashes were consistent for several frames/minutes, and the ProbSevere/reflectivity core showed an uptick shortly thereafter. Because the smaller flashes remained confined to a smaller area/grid box (<72km^2), this indicated to me that this storm would remain pulsy and likely either sub-severe or low-end severe. Taking a look at what WFO SJT issued during this hour indicated to me that this thought process likely held true. From a mesoanalyst perspective, this could prove to be invaluable information to provide to a radar operator.
~Gritty
Final Thoughts
Overall – impressed by Inst AzShear. Think this could be a real game changer. Would like to see this fast tracked into AWIPS.
- When used in conjunction with other base and DP products – really supplements storm structure evolution – e.g. gust fronts, UDCZ, RFDs.
- Areas of highest values > 0.01 are easy to see – and give forecasters an easy method to triage what storms need attention.
- Focused much of my time with Inst AzShear at 0.5 deg. Did tilt up on a number of occasions to find values near cloud base – and often times very similar to lowest cut. Would be interesting to see how this performs with mid-level mesos – for things like hail production
- Only real drawback I see is the “noisy” appearance to data – especially in areas of ground clutter, wind farms, etc. But I think forecasters would get used to it quickly.
- Merged products also useful – both individual and tracks – mainly from an SA perspective – trends, tracks of meso, polygonolgy, etc.
Weak Pulse Storm Analysis – GLM
In the below GIF from Top Left to Bottom Right you have:
- GLM Flash Extent Density from GOES-16 (East) and GOES-17 (West)
- GLM Average Flash Area from GOES-16 (East) and GOES-17 (West)
- GLM Total Optical Energy from GOES-16 (East) and GOES-17 (West)
- ENTLN Ground Based Lightning Detection and GOES-16 1 Minute Channel 2 Visible Satellite.
The animation shows the quick pulse life cycle of the storm – you can watch the updraft grow and then the anvil get sheared off of the storm in the visible satellite. The Flash Density and the Optical Energy show a quick uptick and GLM registers at the same moment the first ground indication shows up. The Optical Energy maxes out as the storm grows to its tallest before the top gets sheared off. The Flash Density seems to follow the progression of the updraft eastward as well. The smallest Average Flash Area also follow the ‘newest’ and/or ‘growing’ parts of the storm. While convection continues from the cell (it is sitting along a boundary across the SE Part of MAF’s CWA), it does weaken (looks like lower tops in visible – could certainly check this in other ways) and you use lightning from all networks.
While there is nothing ground breaking here it does show how GLM handles pulse convection and the trends in the GLM data. The one downside that I would note here is the lack of some sort of downtrend before the storm weakens. It seems the data generally peaks then goes away rather than drops before the storm dies. It is important to note that the coverage of the various products did gradually decrease (went from several pixels, to two, then none) as the storm weakened.
-Alexander T.
Viewing storm trends using merged AzShear products
Merged products end up showing trend information as new sails cuts come in. In the cases below, notice that the northern storm’s AzShear is weakening over time and the southern storm is increasing.
Above, you can see on the northern storm that the last tile shows weaker leading AzShear. At the same time, note that the leading AzShear signature in the southern storm is increasing.
The northern storm continues to dissipate as the upper tilts catch up to the sails cuts that were weakening earlier. While the southern storm continues to increase in AzShear intensity as the sails cuts would suggest.
At this point the northern storm is gone and the southern storm is quite strong and large.
So not only does AzShear provide a situational awareness of the strength of a couplet, it can also show rotational trends in a storm over time assuming that SAILS is turned on. Plus, the more sails the more information you’ll get in these plots. In the case above it appears SAILS2 was on as you’d get two leading signatures before the data merged back into one storm.
–FLGatorDon
SJT Mesoscale Discussion
There is a warm front draped across the northern portion of the CWA with widespred cumulus developing along the front and broken skies in the warm sector. GOES-16 Day Cloud Convection RGB and Day Cloud Phase RGB show that a few of these storms have already glaciated indicating the convective initiation is underway. (Below)
Laps All Sky retrievals show a relatively sharp instability gradient along warm front where our convection initiated. Storm motion is likely to be parallel to the front, so convection that does will likely be relatively long lived. (Below)
Laps All Sky LI further indicates that the airmass is relatively unstable and there will be little convective inhibition at least in the midlevels. (Below)
All Sky TPW indicates a relative moist airmass across the CWA with PWs above 1″ for most of the CWA. This matches the 12z sounding from MAF relatively well, and is well above the 90% threshold of the sounding climatology for that location and date. It’s interesting that the mid-level moisture (bottom left) is higher than the low level moisture (bottom right). Not quite sure what to make of that at this point. (Below)
Finally, MRMS RALA indeed shows that showers and storms have developed along the front. When animated (not shown), you can see that the storms are tracking along the front, heightening my concern for flooding. 
Sandor Clegane
AzShear Inflow and RFD/Gust Front Signature
Around 2120Z – interesting structure on a storm SE of KMXX. Can see in Azshear – inflow from the south immediately ahead of the gust front. However, behind this feature – can see an area of enhanced Azshear. Question would be – how much would the PBL be worked over – cooled – and how might that influence tornadic potential? Yet – another easy depiction on what is going on with storm evolution.
QLCS and Single RDA AzShear
The single radar AzShear product continues to show utility in diagnosing storm structure and trends which leads to higher confidence and earlier warning decisions (in addition to the overall SA nature of the product).
In this case you can clearly see the line segment transition from (see captions):
— FLGatorDon
Inst AzShear – Upper Tilts for Large/Sig Hail
Not in this event – but would be interesting to note the trends in Inst AzShear for Sig Hail producing storms – e.g. mid-level mesocyclones.