Afternoon Forecast Trends

Working ALY desk today. A frontal boundary is slipping south and east across the southern portion of the forecast area. Significant convection is already occurring along the boundary including a few large hail producing supercells. Moderately unstable airmass is shown over the region south of the boundary, with CAPE values on the order of 1500-2000 J/kg. Strong bulk shear around 60kts is more than sufficient to produce the rotating storms that are evident.

Alex Brown

19Z surface plot 1926Z visible satellite clearly show boundary & intense convection

18Z AllSkyLAP CAPE showing moderate instability over far southern forecast area

18Z NUCAPS sounding over far southern forecast area

Very High ProbTor Value

In the image below, note the high ProbTor (85 percent by sampling) at 1850z over southern Columbia County in the ALY CWA (lower right panel). This  was a rapid increase from 27 percent 10 minutes prior. I believe the ALY office had already issued a TOR warning on this cell, but the ProbTor product should increase confidence.

A.Cope

Convective Initiation

Working the ALY forecast area today. Strong/severe convection developed rapidly around 18z ahead of the main line over PA. Loop below shows
probability of severe storms, overlaid on VIS imagery.  Strongest storms  developed along Ulster, Greene and Columbia counties, but not too well anticipated by the CI products which were highlighting the cu field farther SW.
A.Cope

ProbTor "cell" size

The inclusion of the MRMS data into the ProbTor product has really turned it into a good situational awareness tool when triaging storms on the wide view of the radar.

Fig 1. ProbTor product at 1914 UTC.

However, linear storms can lead to an undesired result when it comes to the identification of storm cells. This is generally OK in a severe wind threat, but the tornadic potential is usually quite limited. In the example above, the identified cell is 56 miles long. That doesn't provide a lot of value other than to draw ones eyes to that general area. Having a way to reduce the size of detected cells, especially in QLCS events would be a great improvement.

wall_cloud

NUCAPS data planes

I have found the NUCAPS soundings to be quite beneficial, especially in areas where radiosonde data is sparse. This will be quite useful in my CWA (MEG) where the nearest sounding is roughly 150 miles away. Having the METOP A/B data at the WFO level would help offset the temporal issues.

I think the gridded NUCAPS will be useful as well, especially on when viewing on a plane like we do many other fields. However, I have found the data planes in the volume browser to be quite cumbersome and not exactly intuitive. I think that using more standard levels like most models/observations would be beneficial to the user and would ensure more use in the field.

wall_cloud

Forecast 5/15/18

Moderate instability exists today across northern PA, southern NY, even extending east into portions of MA/CT/NJ. AllSkyLAP CAPE values are generally in the 1500-2500 J/kg across the Mount Holly CWA with surface temperatures already in the mid 80s to near 90 degrees. Strong unidirectional flow aloft is observed across the region with slightly backed low-level winds in the warm sector. The resultant deep layer vertical shear of 35-40 kts (0-6km BWD) will help maintain the QLCS moving east at this time. Strong, organized bowing segments are likely within the QLCS which may result in widespread damaging wind across the CWA. Large hail cannot be ruled out, but this is certainly not the primary concern today. In addition, a few tornadoes will be possible should winds remain backed ahead of the main line. This system will propagate eastward across the CWA very quickly this afternoon.

wall_cloud

GLM Parallax

The image below compares cloud lightning flashes from Earth networks with GLM total energy. The GLM image gives a much more complete picture of the extent of lightning, however there is a noticeable north/south discrepancy between the two. The GLM max just southeast of the main line is shifted northward by about 10 miles relative to the ENI lightning cluster.

A. Cope

Week 3 Day 2 Operations

The forecasters will be operating today in the Northeast starting out in the Albany, NY and Mount Holly, NJ CWAs. We will be monitoring a fast moving convective line across Pennsylvania and New York and later moving into the Boston area with the potential for damaging winds and a few tornadoes.

-Michael

Day 1 Wrap Up

Below are comments from yesterday's activity:

GLM

Came in pretty open minded and didn’t know what to expect

Found it useful overlayed with the ENI data

Helped provide more details than what was there before

Liked having the aerial extent of lightning for lightning safety

Found it useful to incorporate within a TV broadcast, especially in showing the coverage of the lightning
within a storm

Still useful to compare with current lightning network to get the CG strokes in conjunction with the GLM
data

Not sure what to make of the event density and group density with it being the first time seeing it

With the 5 minute running sum, a lot of the cores would run off the top of the color scale

ProbSevere

Seemed to work pretty well with the convective line in Virginia

ProbTor had a bit of a ramp up as the storms approached D.C. as some rotation developed around D.C.
on the north end of the line

The color scale seems not intuitive where the low end and high end fit both in the cool color range

Provides good triage in busy situations and works very well

Tried overlaying it on MRMS products instead of having it on base data and liked having it that way
pretty well

Having more discrete colors instead of a gradient or have increasing line thickness with higher
probabilities

ProbTor should probably have a different color scale loaded more to the lower end instead of the
same as wind and hail.

Think it’s better to have all of the individual hazards contours and readouts than the whole big long
readout

It would be useful to have a time series tool for ProbSevere

NUCAPS

Modified didn’t come in yesterday would have been a useful product yesterday.

Surface temperatures and dew points were quite a bit too cold in the regular NUCAPS

The spatial availability is a great aspect of the soundings especially in areas of sparse coverage

Thought the data looked pretty good particularly in the mid and upper levels

Training needs to be hit harder to know that these soundings exist now

Gridded products need to be simplified to have options of surface data instead of just pressure levels

Having the more passes the better

Useful for these purposes of having greater observational coverage

For boundary layer identification having the more timely data is very important, when looking at lapse
rates or something having the less timely data can still be useful

Bad soundings were advertised as bad

CI and All-Sky LAP

The All-Sky LAP was found useful but could definitely see where the model is seen

Might be better to use the RAP or to have a little better blending to make it less blocky

The CI didn’t seem to handle the towering Cu in NW Oklahoma very well where convection ended up
firing later in the day

There were quite a few false alarm areas in the field as well

Would be able to filter out some of the noise by raising the lower end threshold to eliminate some of
the very low probabilities

Didn’t look at the severe since regular CI values were all pretty low

Tornado Potential Near Washington D.C.

A line of thunderstorms from southern Pennsylvania to northern West Virginia moved southeast into Virginia and Maryland around the Washington, D.C. area. For much of the afternoon, the main threat was from damaging wind and hail.

As the line moved through northern Virginia, it began to interact and eventually merge with a couple of isolated severe thunderstorms that formed earlier in the afternoon ahead of the main line. A field of broken cumulus clouds in between the line and the isolated cells developed very rapidly into a line of severe thunderstorms that produced a number of reports of large hail and damaging straight-line winds.

As the northern end of the line approached the Washington, D.C. area, the southern end of the line surged southeastward into the area south of the isolated cells. The northern end of the line slowed as the cells merged with the northern end of the line. As a result, a very nice mesocyclone became very evident in the NEXRAD imagery as it passed near the Sterling radar site. Tornado warnings were issued, although no touchdowns were reported (a funnel cloud was observed at the Sterling office).

Of particular interest to me was the increase in lightning as the merging between the cells and the line took place, followed by a decrease in lightning as the mesocyclone formed. Also evident was the progression of overshooting cloud tops along the line as the mergers took place, with the overshooting tops traveling in an east-northeasterly direction toward the DC area while the line of storms moved steadily to the southeast. as the more robust thunderstorm activity moved southeast of the DC area, you could see how by using the NEXRAD data along with the satellite and lightning data the tornado threat might have been on the increase toward the DC area, while a more significant wind, hail, and heavy rain threat was favoring areas to the south.

Finally, the ProbTor product took that into account and while not overly high, showed an increase to about 30% as the storms moved toward Washington, D.C., then decreased very rapidly after that as the mesocyclone became rain-wrapped and fell behind the main line of thunderstorms as they headed south of the DC area.

Gary Cannalte