LJDA & Prob. Severe Model with CAE CWA T-Storms

My co-worker and I continue to monitor a few thunderstorms across our assigned CWA this afternoon. The Prob. Severe Model and Lightning Jump Detection Algorithm tools continue to be very helpful today with the decision to issue warnings.

Doppler radar image at 1922z with Prob. Severe & LJDA overlayed

The early lightning jump rose to a 4.0 sigma value, which to me hinted that this storm was getting stronger by the minute.

Plus, we were tracking the Prob. Severe figure and at first it was 48%, then 61%, followed by 83%, and finally 92% before I decided to issue a Severe Thunderstorm Warning for the very southern fringe of the CWA.

Prob. Severe output indicating 92% with also a hail spike shown on a previous frame of the radar data. I decided to issue a warning.

The warning did end up verifying in Burke County 10 miles northwest of the city of Girard with golf ball sized hail reported. That was reported at 2004z.

-BW

NUCAPS Sounding Near KJNX Observation 1800 UTC - Adjusting It Is Best For Use By Mesoscale Analyst In Severe Ops.

Here is a NUCAPS sounding near KJNX in the east central portion of the RAH CWA. Note the near surface temperature was about 21C/70F and dew point temperature was about 13C/55F.

Here is a visible satelllite image with the NUCAPS sites and surface observations overlaid. Note the cursor readout of temperature at 24C/75F and dew point temperature at 19C/66F at KJNX, which is very close to the NUCAPS sounding site.

Here is the adjusted NUCAPS sounding near KJNX, with a more representative temperature and dew point temperature profile in the low levels. Note the much higher values of CAPE in the adjusted sounding. They may be too high, as the temperature profile may need more adjustment further up. Still, it gives an idea of how unstable the airmass ahead of the convection is. In addition, the -20C level did not change much (about 21,500 feet above ground level), compared to the observed 12 UTC sounding at Greesnboro, NC (perhaps 100km/60miles west northwest of the KJNX area). This made sense given the synoptic setup, and would help the warning forecaster with where this level is over the eastern portions of the RAH CWA.

Overall, it seems that having to adjust the low levels of the NUCAPS sounding would be best handled by the mesoscale forecaster, NOT the warning forecaster. Once you get the hang of adjusting these NUCAPS profiles, they can be useful for near storm environmental monitoring. This would likely be too distracting for the warning forecaster to do, while trying to watch the storms themselves. JJW

NUCAPS vs GOES LAP

Comparison of NUCAPS sounding data with GOES LAP.  Note the cursor rreadout in the second image indicating LAPS derived CAPE of 1709 J/kg versus surface initialized T/TD of 82/62 which gives a NUCAPS derived CAPE of 1690 J/kg...quite good I'd say.

NUCAPS 18Z Indicies

18Z GOES LAP

Tornadotod

ProbSevere Prompts Warning

One of the first convective cells of the afternoon approached the RAH forecast area...but was pulsing in strength. The ProbSevere product (overlaid on the visible imagery) also fluctuated between 50 and 70% as the cell approached the RAH border. But once the probability reached 75% I issued a warning. At the same time, there was a 2.0 Sigma jump in the lightning flash rate, 50 dBZ at the -20C level and plenty of ice present (in the VIL shown in the top right panel). The cell slowly progressed eastward and began weakening (evident in both the ProbSevere output and reflectivity). Once the ProbSevere dropped below 50%, I cancelled the warning.

~Lilly Miller~

UPDATE: This cell weakened after entering the CWA, prompting a cancellation of the warning. However...the cell continued its pulsing life cycle for at least a couple of more hours and eventually strengthening into a stout supercell and dropped severe hail once it reached better instability.

GFS Underestimating CAPE, LAP Helps With Adjusting It Upwards.

The LAP imagery 4 panel below at 1800 UTC shows CAPE (upper left), LI (lower left), Total PW (lower right) and Layer Cloud Type (upper right). The sample tool shows a rather low CAPE estimate area, which is being derived from the GFS model (upper right image). The higher CAPE values to the northwest of there (upper left) shows values being derived from clear or cloudy skies. Thus, the GFS is underestimating the CAPE values. The LAP data is helping adjust it upwards to more realistic values. This is helpful to know where the GFS is being used, and that its values should be higher than shown. JJW

CI Detections Verifying With Radar in RAH CWA.

In the four panel image below, the upper left image shows CI detections across the RAH CWA at 1815 UTC. The sample is detecting a 59% chance of CI in southern Randolph County.

In the radar image below from KRAX at 1832 UTC showed a 44 dBZ radar echo in southern Randolph County. Thus, the CI product did a good job with this cell. It provided about 17 minutes of lead time, which was cool to see! JJW

Predictability with GOES CI

Interesting GOES CI animation showing good advance predictability with small but growing convective cluster over central portions of GSP's CWA.

Tornadotod

LAPS Pre-convective Environment - KRAH

A bit more difficult to assess the LAPS data over the KRAH forecast area due to early morning cloudiness. You can see over the RAH area...you have a mix of cloudiness, GFS data, and clear sky (shown in the top right panel). However...as the clouds move eastward...you can see a clear increase in CAPE (top left panel) and LI (bottom left panel) values. The layer PW however shows a decrease in overall moisture indicative of the drier upper level air encroaching on KRAH.

In a situation like this (with more cloud cover than clear sky) I will tend not to trust the LAPS data as much since its relying so much on NWP data. However...I will trust the data trends...which shows an increase in instability as the cold front approaches the region.  Overall...I feel that the LAPS data is a good resource for assessing the pre-convective environment.

~Lilly Miller~

LAP Precip Water 4 Panel and Sat Winds - Helpful With Moisture/Shear Over RAH CWA, But Memory Intensive.

This image is a 4 panel of the LAP Total PW and satellite winds (upper left), then 900 mb PW/winds (lower left), 900-700 mb PW/winds (lower right), and 700-300 mb PW/winds (upper right). It gave a quick at-a-glance of the moisture and wind profiles in the RAH CWA at around 1700 UTC. The readout shows total PW around 1.40 inches, with around 35 knots of shear in the 700 mb to 300 mb layer. The last several images (not shown) were showing drying in the middle to upper levels in the RAH CWA, so lapse rates and instability in this area should be increasing. This trend is expected to continue into the early to mid afternoon hours. One issue to note: This image takes awhile to load up in AWIPS II, and seems to use a fair amount of memory. This would be an issue when working severe weather events, as this may lead to AWIPS II crashing (combined with other memory-intensive products). A solution may be to unload some of the winds, or just use the top left image as a quick look image. JJW

GOES-R LAP 4 Panel Before Thunderstorms (CAE CWA)

This is a great product to look at before the anticipated thunderstorms this afternoon. As of 1730z, there are just a few, non-severe, thunderstorms across the far southern portion of the CAE CWA.

The 4 panel does include the GFS model as part of the analysis. While hovering over the CWA, I noticed that the GFS is under doing the LI, CAPE, and PW when comparing to the Charleston, SC 12z sounding and what we expect in the area today.

GFS output moused over

When looking at the satellite output, the LI, CAPE, and PW values greatly increased to figures we would expect with thunderstorm development today.

Satellite output moused over

The satellite data does help to increase our confidence when forecasting overall convection for the afternoon and early evening.

-BW