Initial thoughts on LUB potential

We are tasked with assessing the pGLM and GOES-R/CIMMS CI products. A very moist and unstable atmosphere is present over much of the LUB CWA with an ongoing MCS exiting the northeastern portions of the CWA. GOES-R/CIMMS layer CAPE values range 400-500 J/KG for the northwestern portions to 1200-1400 J/kg for southeastern portions. PWATs also possessed a sharp moisture gradient from NW to SE, with values near 0.60″ NW to 1.40″ SE, along with dewpoints in the mid 60s building across southern half of the CWA ahead of the eastward moving MCS.

RAP13 forecast sounding centered just south of Lubbock around 18Z indicated MLCAPE around 1600 J/KG, fairly close to the GOES layer CAPE values. Freezing level and -20 C heights are around 20k ft with modest moisture in hail region, though very moist low levels will help overcome some drying aloft. Expect large hail and damaging wind gusts to be an issue with any discrete cells, as well as isolated tornadoes possible due to modest low level shear/helicity. Will monitor CI for possible development south of LUB/ongoing MCS in hopes of catching pGLM/lightning jump with storms as they develop.

Brick Tamland

Amarillo Synoptic/Mesoscale Discussion

Around 1800 UTC, a MCS is going across the southeastern portion of our CWA.  We expect this region to remain stable in the wake of this system as it progresses eastward along the Red River Valley.  Central and western parts of the CWA have seen ample sunshine.  A dry punch is visible moving in from the southwest, and dewpoints have dropped into the 40s through eastern New Mexico, with low to mid 50s through much of the Texas panhandle.  We expect some additional moisture return, especially through southern parts of the CWA…along with a tightening of the dryline feature.  There is some question to the amount of instability, although GOES LAP algorithm indicates 800-900 J/kg through central and southern parts of the CWA…although with heavy clouds in the region, it’s difficult to gage the accuracy of the values.  A 500 mb low is centered over central California, with southwest flow over west Texas.  Any subtle wave or ripple in this flow may be enough to trigger high based convection along the dryline feature.  The 12 UTC AMA RAOB indicates mainly unidirectional flow through the column.

Daily Summary: Week 1 Day 3 (May 6, 2015)

Convective activity began today earlier than was expected across the southern Great Plains. We placed groups in CWA's where the greatest severe threat was expected for the day: Norman, Wichita, and Hastings. As convection was already ongoing when we began operations, attention was focused on warning activities from the start, with the Earth Networks lightning products, Lightning Jump, and ProbSevere Model having a lot of utility.

Tomorrow, we will start activities at 12:00 pm in the Dev Lab for the daily debrief before moving to the HWT for the EFP briefing. Activities will likely shift back to West Texas.

- Bill Line, SPC/HWT Satellite Liaison and Week 1 EWP Coordinator


GOES-R LAP

- Absolute values aren't all that helpful, but the trends and gradients have been very helpful and accurate.
- Yesterday, storms in OK initiated on the nose of increasing CAPE values, where one would expect.
- These fields provide a nice visualization of how the airmass is evolving.


NUCAPS

- I selected a sounding from a clear spot on the NE/KS state line and had to modify it at the low levels as it was to dry and cold. The Mid levels seemed to be pretty good. The modifications drastically increased the CAPE to around 2000 j/kg, which was a better reprsentation of what I thought the environment was.
- The instability values seemed way too low, so after some adjustments to the low levels, it seemed much more accurate, especially with respect to the nearest 12z RAOBs.

CI

- It did a good job as the storms in C Norman were developing upstream, the product was telling me that they would continue to develop. This was especially valuable in this busy situation.
- There was a case where I saw 60s and 70s S of Witchita, 30 min before convection initiated.
- I would like to see another version of the CI tool where the probabilities are for the development of a higher value dBz echo (eg. 50 dBz instead of 35 dBz). This would be more useful for the WFO.
- By 6 o'clock, there was too much anvil cirrus present in the area for the tool to be effective.

ProbSevere

- It is nice to have the outline display so I can still see my radar
- More of a confidence booster yesterday, which was an obvious day. On the more marginal days, it was more useful.
- I see it being especially useful on pulse severe days
- With the text output, it would be helpful to color code or somehow highlight values when they reach certain thresholds, making them stand out.
- Make the actual probability stand out more in the text output.

Lightning Jump

- I took any lightning jump as a sign I should look at the storm
- I had one screen dedicated to LJ witth radar and ProbSevere. With multiple radars, space was tough to find.
- This was not ideal to overlay with radar, it covers too much up.
- There was a cell merger, followed by a lightning jump.
- When we saw the jump, there was no question about the severity of storms, so I am not sure it influenced my warning decision.
- With these obvious severe storms, this tool wasn't as helpful, it was more of a confidence booster.

Takeaways from ICT Today

After a busy day at ICT, we have come to the conclusion that radar base data is still our #1 when it comes to warning operations, while the other tools are “confidence boosters”. With many (if not all) of the warnings we issued today, there was little doubt that the storms were likely to produce severe weather. The radar base data told us what we needed to know to go ahead and issue the warnings, with things like lightning jumps and prob severe being more of a reassurance. In a lot of cases, the prob severe did not substantially increase until we we already in the process of writing up or had already issued a warning. Take this storm in Ellsworth County, KS for example.

When we decided to issue a SVR for this storm, the prob severe was only 13%. Meanwhile, the base data was pointing towards severe hail.

As the warning was being typed up, the next volume scan came in and the prob severe jumped to 85%. Trends in lightning were also more of an afterthought/confidence booster with regards to subsequent storm intensification.



In cases like today, the supplemental tools did not necessarily add any lead time. We do think tools like prob severe and changes in flash rates would be a bigger help with the warning decision making process in marginal severe environments or pulse storms.

Jack Bauer/V. Darkbloom

OUN Wrapup

I should have know a busy day was in store for the OUN area when sitting down and already seeing a storm that needed a Tornado Warning. That continued throughout the rest of the day with numerous TOR/SVR warnings issued (at least 30 and probably over 40). With the quick startup, didn’t get much of a chance to look at the GOES LAP data other than a quick overview to start the shift.

With the size of the OUN area, I was jumping back and forth between 3 different radars and a significant distance between areas made for an overwhelming feeling at times during the shift (especially when trying to look at more than just radar). I used the ProbSevere and lightning jump throughout the shift and found them both to be very useful (especially for situational awareness). In most instances, saw both lightning jumps and rapid increase in ProbSevere before the storms really took off. With as busy as things were, I wish I could have looked at the lightning jump data more in real-time. Unfortunately, there just wasn’t enough screen real-estate to overlay another product (I tried to overlay it on radar data with 50% transparency and it was too much).

All in all, a busy but fun day in the OUN area.

-SRF

A few satellite product observations

Unfortunately we lost GOESR/CIMMS layer CAPE/PWAT products around 22Z. Originally we expected to have a short amount of time to form a mesoscale discussion for GID but quickly transitioned to warning mode, issuing both SVRs and TORs. We were not able to assess the UAH CI product as the area quickly became cloud covered. The Overshooting Top product successfully highlighted the stronger storms, especially those with large hail cores aloft.

Brick Tamland/Alexander Darkband

Wrap up for Day 3

I used mainly lightning jump and Prob of Svr.  This was helpful with increasing confidence in issuing warnings but was not a deciding factor to warn. CI was used the first 3 to 4 hours of the shift before the CI shield became to thick to make it useful. Was able to assist the warning operator during the first two hours of the shift of where to be looking for development based on CI from satellite. This would have been more useful with much better familiarization of the CWA.

Saw some impressive hail spikes and deep convection today.

Lynford

Where to start?

After an eventful afternoon and evening in Norman (here on the south side of town we are fine and continued operations as a tornadic supercell hit northern Norman) it is difficult to pick a storm and blog about it since from Nebraska south to Texas there have numerous severe storms with many severe reports.  The busy weather continues at 730pm CDT as seen in Figure 1.

Figure 1. NOAA/CIMSS ProbSevere, Composite Radar Reflectivity and GOES visible imagery valid at 0024 UTC 07 May 2015.

In general the NOAA/CIMSS ProbSevere model has proved to be quite useful with forecasters in all CWAs today.  One of the forecasters working the OUN WFO today echoed something we heard in the 2014 HWT.  The forecaster had many storms to monitor and issue warnings for, including the tornadic storm approaching Norman.  The forecaster commented one utility of things like NOAA/CIMSS ProbSevere and the lightning jumps/total lightning flash density can be used to draw attention to storms that need further interrogating and possible warnings, which can sometimes be easy to miss when one is focused on a few tornadic supercells.

-Sieglaff

NUCAPS and GOES LAP CAPE

The NUCAPS sounding data became available around 19Z.  We chose a point along the Nebraska/Kansas state line in a cloud free region.  Overall, the sounding appeared valid.  However, the surface dewpoint and temperatures were around ten degrees too low (in comparison to nearby METAR), producing too little CAPE.  After modifying the surface data, CAPE increased to around 2000 J/kg.  This was much higher than the 800-1200 J/kg values being observed from the GOES LAP data.  Considering the severity of resulting convection, the 2000 J/kg values were likely most representative of the environment.  The first image capture is the original, the second is the modified.

BRICK TAMLIN/ALEXANDERS DARK BAND

ProbSevere and Lightning Jump for SA

I was focused on warnings for two areas of the cwa and completely missed out on this severe thunderstorm. Looking back, using ProbSevere and Lightning Jump data could be useful in maintaining situational awareness while in warning ops.

This first image taken at 2308Z, shows both the lightning jump and ProbSevere not indicating anything significant with this storm.

But jumping ahead to 5 minutes to 2313Z, there was a 2 sigma lightning jump as the ProbSevere rises to around 50%.



Then, at 2323Z the ProbSevere was up to 77% and there was a 1 sigma lightning jump.



The storm likely went severe shortly after that and a warning was issued later. Having a screen dedicated to those products could be useful in maintaining awareness where there were numerous warnings out. -SRF