Here is an example of how UAH Convective Initiation (CI) and CIMSS Cloud Top Cooling (CTC) can be used to anticipate convective development. Here in the first image at 1940Z, let’s look at the CI product in the top left panel:
Focus on the Lake/Marion/Sumter County line. There is a maximum in CI there of 80, indicating likely convective development. If we look at the CTC in the same area on the next image (top right panel), we see a developing max CTC over -20C/15min, which is a strong indicator of a developing severe thunderstorm.
Although this did not ultimately become a severe thunderstorm, we did observe a few lightning strikes shortly after this time. This progression from strong CI signal to strong CTC can be a good indicator of a developing thunderstorm.–Gordon Strassberg for WFO MLB
No computer-based algorithm is perfect, right? We had a failure of the Cloud Top Cooling (CTC). Here is a CTC image from 2345Z, in the top right panel:
The developing storm over York County has values over -30C/15min, which is usually a strong indicator of a developing severe thunderstorm. As this storm developed and approached Seward County from WFO GID, a warning was issued based on CTC. Maximum Expected Size of Hail (MESH) was 0.87″ and increasing. However, as the storm entered Seward County, it weakened. No severe reports were received, and the MESH dropped pretty rapidly. Here is an image as the storm entered the CWA:
The moral of the story? You can’t win ’em all, CTC!–Gordon Strassberg for WFO OAX.
Issued a severe thunderstorm warning strictly off of explosive CTC at 2315 UTC (previous post), with very little showing up on radar at the time. As would be expected, through the following 15-20 minutes, the storm exploded, with MRMS MESH spiking up to 1″ by 2334 UTC, and a quarter size hail report by 2336 UTC. This gave us 21 minutes of lead time from the initial warning issuance off of CTC. Would have likely issued the warning based on traditional base products by 2325 UTC, 10 minutes after the CTC-based issuance.
Here in WFO ILM, we have had some great success with the CIMMS Cloud Top Cooling (CTC) product this afternoon (see our previous post). Here is another success story from Horry County, SC. First we will look at the Cloud Top Cooling product at 1845Z (Top Right panel):
There are two “bull’s eye’s” over SC. Our storm is the northern “bull’s eye”. The CTC showed cooling rates of -21C/15minutes AT 1845Z, which typically correlates to strong convection with significant lead times averaging 45 minutes before Maximum Expected Size of Hail (MESH) exceeds 1″. Now let’s look at some verification. First is the MESH image from 1936Z (lower left panel). This is the first time the MESH exceeded 1″ for a LEAD TIME OF 51 minutes!
And now for a conventional radar image from 1947Z from the KLTX Radar:
The cursor readout shows this storm has 60dbz echos up over 28,000ft…which is above even the -30C level from the 12Z CHS sounding (~25,000ft). We finally got verification of quarter size hail in NW Horry County by around 20Z, though the reports seem to have not come from directly beneath the core of the storm so some larger hail was possibly produced. Now let’s examine some NSSL 3DVAR Imagery:
You can easily identify this storm based on the Max Divergence above 8km, Updraft Composite, and Max Vorticity Images as all show local, significant maxima. Today, the CTC product has shown significant value in lead time for severe hail beneath the Southeastern CONUS upper low.–Gordon Strassberg for WFO ILM.
The CIMMS cloud top cooling algorithm first detected cooling rates of approximately -10 C/15 min with a thunderstorm developing across eastern Hoke and western Cumberland counties on the 1815z image. The algorithm continued to show strong cooling rates on the next two images with accumulated CTC values of approximately -30 to -35 C/15 min before the cell was dropped from the display after 1845z. A report of quarter sized hail was received in far southwestern Harnett county with this thunderstorm at 1937z, almost an hour and a half after the algorithm first detected stronger cooling with this storm.
In the 1815 UTC vis satellite image (upper right of top 3-panel), cloud-top cooling was on the order of -17C/15min with a developing thunderstorm. By 1858 UTC, MRMS MESH product reached 1″ hail size (middle image), and just over 2″ hail size by 1906 UTC. The UW-CTC product lead to 45-50 minute lead time for 1-2″ hail (as estimated by MRMS MESH). This is one of many instances we’ve noticed the CTC give decent lead time for 1″ or larger hail.
The CIMSS-CTC algorithm showed cloud top cooling rates of between -10 to -15 C/15min on several images with a thunderstorm developing across Sampson county in North Carolina. The CTC first appeared on the 1715z image and continued to highlight this cell through the 1815z image. As the algorithm continued to show moderate cooling rates with the cell the radar presentation became more impressive and a three body scatter spike developed with the cell. A report of penny sized hail was received with this cell 1823z 5 miles WSW of Newton Grove, NC.
1745z image indicates developing cumulus over extreme northeast NM. In the 1830z image, UAH-CI product indicates a cluster of cumulus with a high confidence of convectively initiating. UW-CTC product also indicates rapid cloud-top cooling with the eastern-most convective element. Given the deeply mixed airmass with minimal cap, this would lend confidence to convective initiation soon. Sure enough, by 1915z the 15-minute lightning plot indicates some CG strikes with the eastern-most convective element, with a lead time to convective initiation of over one hour.
Another severe thunderstorm warning was recently issued (2148 UTC) for a portion of southeastern Georgia (Appling and Jeff Davis Counties), west of Vidalia. It’s worth noting that both the UAH CI and the UW CTC products flagged this storm quite some time ago.
The first indication on this cell occurred with a 59 strength-of-signal indication from the UAH CI product.
The SOS increased to 68 on the next satellite scan, then 71 on the subsequent scan (2015 UTC). The UW CTC flagged the storm for the first time at 2045 UTC with CTC rates of -13 to -14 C/15min, and that decreased to -16 C/15 min on the next image.
The visible imagery becomes quite impressive with overshooting tops by 2130 UTC. The SVR was issued at 2148 UTC, based mostly on distant radars. The 3DVAR analysis domain was just recently expanded to include this area, and MRMS POSH and MEHS were not particularly impressive.