GLM and Convective Inititation

Saw an interesting example of a developing CB (over land at the Haiti/Dominican Republic border) where GLM flashed before development was recognizable on satellite.

First, a longer loop.  Then a shorter loop focused on three frames.  All data is time matched to satellite.  In addition to GLM, ENI is plotted on the upper left (small yellow dots).

Satellite/Lightning 2002-2057Z

At the center of each of the four images is the developing cell.  The first flash occurs on frame 4 of 12.  It's most noticeable on the lower right with VIS and Total Energy.  Then, notice on the upper right the enhanced IR.  The flash seemed to practically come from nowhere, but then the cloud becomes more recognizably convective a couple frames later.

Three framed loop.  Initial flash is frame 2.

Satellite times above are 2012-2017-2022Z, with the initial flash at 2017Z.  Again, notice the upper right enhanced IR with little to see.  Max tops estimated by cloud top temp in the sequence was FL280-FL320 (flash)-FL350.  The cloud top ended up a frame or two later peaking at FL380.  ENI didn't pick this up for a few frames after this loop.

Sequence of GLM product values for the event:

Total Energy:  0fJ - 52 - 0.4
Flash Extent Density:  0 - 1 - 1
Event Density:  0 - 42 - 3
Area Extent.   0km^2 - 1122 - 70.2

Eric happened by while I was checking this out and made an interesting comment.  The intense, single flash was likely seen very clearly from the satellite since the icing/charge must have been restricted to the cloud top.  Thus, GLM can be useful for recognizing convective initiation.

-Forrest

To Warn or Not To Warn…

…we didn’t warn. Despite ProbSevere jumping to 75% and decent looking reflectivities, we decided not to warn on the cell in Jackson Co, KS. The MESH estimates were only up to .7″.

But the bigger deciding factor was unimpressive lightning data (seen below). lightningnojump

-BT

Severe Omaha Charley

Have a quickly developing cell in our southern CWA.

Decided to issue a severe based on a quickly growing lightning time series (with little change in areal extent) as well as a high ProbSevere.

Renville Co Storm MPX

An increase in the CIMSS prob severe gave us higher confidence to issue this warning for this storm over you can see a jump in the lightning also near the time the prob severe increased it’s prob.

Charley…Most Useful Products Today

Overall with the pattern in place it seemed the DVIL and ProbSevere helped me to make warning decisions more than other products.  I used base velocity for a few of the SPS/warning decisions.  DVIL and clusters of total lightning helped focus on the stronger storms.  However there were several storms that we issued on with lower lightning data.  In particular the cell over Williamson county did not have a lot of lightning with it, but it did produce several reports of hail…and we later learned it had some wind damage as well.

We had some hiccups with the SRS satellite data not coming in.

Lightning time series data were not as difficult to work with as I anticipated.  I would expect a problem though if I had more than one storm to focus on.  Perhaps with experience that will get easier.  It was straight-forward to understand when cells falsely jumped in lightning data, but because of cell mergers.

PNGA/lightning data on Warned storm Williamson Co

We were debating on whether or not to reissue a warning on storm and noticed that the PGNA product was showing an increase in total lightning and decided on issuing.

The time series product also showed an increase right before we re-warned…but quickly began to tail off again after we re-issued.

Looking at Lightning Jump

We’ve got a lot of action going on across our CWA this afternoon, and most of the storms have had at least a 1 sigma lightning jump. Some of them have had a significant jump, up to 7 sigs.

One cell we were tracking (eventually turning more linear) was moving across Perry Co, PA. We spotted an 80 flash/min jump (from 20-100) in 10 minutes.

This area did receive reports of a tree down (~1950z) and pea-sized hail (~1950z) just before that huge jump. There was still a smaller jump taking place at that time (from 10-20/min over 10 minutes).

We did notice there were some tracking issues in terms of cells merging and determining a jump. We account this to the amount and speed of convection in our area. Visual jumps were spotted on the time series graph, but it did not show well on the LJDA map.

-BT

Prob SVR and LJ in VA severe storm.

First SVR report from downed trees in Shenandoah county at 1848 UTC 8 June 2015. Note the NOAA/CIMSS Prob. Severe Model showed an increase in Prob from  36 to 50 to 84% at 1834 UTC then a rapid decrease back to 36% by 1842 UTC. The using 84% would have resulted in a lead time of 14 minutes.  The animation below shows this evolution with the storm headed in to the ground clutter near the RDA.

There was also a notable lightning jump (LJ) from 1830-1835 UTC.    It might be useful to include a maximum lighting rate and rate of change value in the meta data of the lighting jump time section. MrSnow/Wacha.

Warning 1

Issued a warning on cell over Humphrey’s County…storm was in an environment that had already produced storms with severe hail. Noticed that it had not yet produced an overshooting top but looked close using the one minute Super rapid scan imagery also noticed local maximum area of cape out ahead of where the storm was developing with clear skies

Notice the lightning jump on the ENI time series…this may have added urgency to issue a warning had I used it to begin with. Did get wind damage reported and power outages with this cell.

Inthecards

ENI Time Series and Prob Severe Influence Warning Decision

I noticed a very dramatic increase in lightning activity and prob severe as a storm approached the Jacksonville FA (from Lanier County), so I decided to go ahead and pull the trigger on a severe thunderstorm warning. The prob severe didn’t quite make it to the 80% threshold (only 78%), but since it accompanied the ‘lightning jump’, my confidence was higher that it was intensifying and was capable of producing severe weather. There was also an overshooting top associated with this storm as seen by the overshooting top detection.

The warning has not verified yet.

(Click image to see animation)

-Helen Hunt

First Severe for CYS

ProbSevere consistently climbed with this first storm.  When we got to 74% severe we issued a warning for the storm based on hail.  The ENI total lightning data jumped as well suggesting that there is a lot more ice with this storm.

MacGyver and John Pendergrast

Overshooting Top and ENI Time Series Comparison

We noticed a significant ‘lightning jump’ around 1820Z in the ENI time series which was indicative of the updraft increasing in strength and size. Not long after that at 1830Z, the overshooting top detection picked up on an overshooting stop associated with the same storm. It makes sense that we would see the lightning jump first, then get the overshooting top detection a bit later as the momentum of the growing updraft catapulted it over the tropopause.

-Helen Hunt

ProbSevere jump

The ProbSevere algorithm jumped from 36% to 88% between 0052z and 0056z. At 0102z, a large TBSS appeared, so there was certainly lead time on that feature showing up.

Lightning has also increased enough to flag a Significant Thunderstorm Alert. While it didn’t initially look like storm was going produce much, it ended up intensifying – the ProbSevere definitely gave a few extra minutes of lead time.

Jason Williams

Convection continues in Montana

Storms are still festering in the NW BYZ CWA. This one near Hailstone Wildlife Refuge (okay, its name was a big factor for this post) went from 12% to 36% between 0046z and 0048z.

Lightning trends have not been impressive – only the northern storm, just outside the radar screenshot, has had a thunderstorm alert.

Jason Williams

ENI Data with Storm Split

This was an interesting case where the storm split and the ENI data highlighted the stronger southern cell which was very beneficial in the warning process.

The ENI products all worked beautifully in this case. Initially, I noticed a slowly increasing trend in Prob Severe around 2346Z over the cell. By 2355Z, I noticed that there was an ENI DTA Thunderstorm Alert on the storm. At 0000Z, the ENI DTA Thunderstorm Alert increased to Significant and the Prob Severe split into two ovals since the storms had split by this point. By 0007Z, I  began to see that the southern storm had become the stronger of the two cells as the severe prob had increased to 72% while the northern storm prob severe was only at 42%. At 009Z, prob severe on the southern cell increased to 90% while it was still only 42% on the northern cell. At this point, I decided to go ahead and pull the trigger on the severe thunderstorm warning for the southern cell. Knowing that the prob severe was much lower on the northern cell increased my confidence that I should only include the southern cell in my warning polygon. The warning took a long time to get out due to some technical difficulties, so I didn’t get the lead time that I would have liked, but the warnings did verify. There was a report of 1.25” hail at 7:25 PM, 2” hail at 7:23 PM and 1” hail at 7:52 PM.

(Click image to animate)

The ENI time series data was also helpful in this case as it showed lightning jumps around the same time I was seeing other signs of intensification. This increased my confidence that the storms were indeed intensifying and that a severe thunderstorm warning was needed.

-Helen Hunt

Using ProbSevere as a key piece of info for warning decision.

Issued a SVR at 23:53z over northern Stutsman County. Lightning flash rate increased prior to intensification from 2340Z to 2347Z (fig 1) then the cell produced 55 dbZ to >30kft.  The cell was getting to be far from the radar with the 0.5 scan reaching only 12.5kft. ProbSevere showed probability of 94% at this time (Fig 2).  The limited low level information with a farther distance from the radar made the use of ProbSevere more heavily weighted in the warning decision. –CattyWampus

Fig 1: Increasing flash rate from 2340Z to 2347Z.

Fig 2: ProbSevere 94% at 2352z.

Coastal WFOs will benefit from the total lightning presentations

Coastal sites challenged with open water warning requirements while facing movement of cells away from the coast, eventual degradation of radar data will lead to increased usefulness of rapid refresh of ENI lightning trends for cell interrogation. This should lead to higher confidence on issuance of marine warnings and statements.

John Pendergrast

No-warn using ENI and ProbSevere in Montana

ProbSevere was increasing after 2300z, and reached above 80% at the 2324z radar scan. However, the lightning data doesn’t show anything particularly notable as far as an increase in rate.

Hail potential is evident on radar, although ProbSevere MESH is only at about 1″. Given the marginal nature of the MESH and the lack of increase in lightning, this is a situation in which I would hedge toward not warning because of the addition of the experimental products. If I just had NEXRAD, I would have leaned toward warning, despite the distance from radar.

Jason Williams

ENI compared to LJ

I’ve noticed a few cases today where the Lightning Jump Algorithm and ENI polygons and time series data do not line up as I might have imagined. Lightning jump is based on radar data whereas the ENI polygons and time series data use the Earth Network lightning data. Beam blockage due to the terrain in the Pocatello CWA may have affected the radar data going into the lightning jump which may account for some of the differences.

In the example below, the lightning jump algorithm shows a value of 6 sigma at 2301Z whereas the ENI time series shows only a relatively small increase in lightning activity at this time.

-Helen Hunt

6/1/15: 1st Severe TS Warning

We issued our 1st SVR at 2246Z. This decision was based on a combination of basic base data interrogation. Solid core of 50dbz with 65+dbz at 0.5 degrees. Dual-Pol showed signs of a TBSS in ZDR and CC radially behind the core.

In terms of experimental tools we noticed the Sigma Jump of 2 at 2237Z, then next scan showed an 11 Sigma. This can be seen in the gif loop below.

Figure: Gif Loop of reflectivity and Sigma for lightning

Fig 2: Plot of the the ENI lightning jump near the time of the warning.

It is interesting to note that the ENI lightning jump plot showed this rapid jump as well (fig 2). At this time the rapid jump could have also demonstrated a cell merger in conjunction with intensification of the southern cell which also demonstrated a low level mesocyclone along with a POSH/POH of 80/100 and MESH of 1.25″ (fig 3).

Fig 3: 2246Z, Low level circulation and hail indicator on the HC plot.

Forecasters: Cattywampus/CoonieCatEye