Starting the day in OHX

Since convection has not fired up yet in the Tennessee/Ohio Valleys, I was able to spend some time analyzing the pre-storm environment with the products we’re testing this week. The first on the list is the NSSL-WRF simulated IR/WV satellite product; I pulled up a four-panel with simulated on the left and actual on the right (below). As of ~1845Z-19Z, the simulated imagery was depicting isolated convection across southern KY and middle TN too soon but not necessarily entirely inaccurate. It is accurately honing in on the cloud cluster centered over IN-IL-IA/MO border, albeit overdoing the coverage and coolness of the cloud temps, as has been observed all this week. Interestingly, it is not picking up on the cirrus overspreading the Ohio Valley and into TN. Accounting for all of this, I can adjust accordingly and remain confident in using this product to develop a mesoscale forecast.

Moving to the next item on the list, I pulled up a four-panel of the NearCast analysis products with 500mb theta-e in the top left, 780mb theta-e in the bottom left, theta-e difference in the top right, and 900-700mb PW in the bottom right (below). A surge of mid-level instability can be seen diving into TN and is also evident in the difference product with the gradient diagonally stretched from the NW corner of TN to the SE corner. PW values/moisture surge matches with where storms are quickly firing up in western and south-central KY. When analysis data is available for these NearCast products, I really like to use these to diagnose and forecast.



Next, I loaded the same four-panel with the NearCast forecast and took a screen shot of the 2-hour forecast (21Z, below). Unfortunately, missing data over the region of interest exists with the 500mb theta-e and theta-e difference products. However, I can still see the lower-level instability and moisture axis draped across western KY into TN in the 780mb theta-e and 900-700mb PW, respectively. Given the success of these products so far, I have good confidence in utilizing the forecasts to accurately predict where the greatest threat will be, given other atmospheric conditions line up for convection as well.



Looking at GOES-14 SRSOR vis imagery and overlaying the CI product, a 60% probability popped up in western KY (below) and within a half hour, storms developed in the vicinity. There was a false alarm 60% probability just to the west of Paducah, KY, probably due to the thin cirrus streaming over a CU field there. In OHX’s CWA, no reliable indications of convective initiation has appeared yet. As for the SRSOR product, I am once again very pleased to have this data available to more accurately assess and see the evolution of the regional atmosphere.



Below, I overlaid OTD on IR SRSOR imagery and thus far, no detections.



~Linda

Lightning Jump

I finally got to see lightning jump data and it was so helpful.  At this frame you can it has the brown color and then as I continued the loop it went to a green color.  It indicated to me that this storm was intensifying and definitely got my attention.  It also was trending up on the Prob Severe so at this point I would probably issue a warning. ~Vollmar

Development of storms in LWX

The forecasters are concentrating on the Sterling WFO and Nashville WFO today. Initiation has already occurred across LWX with the GOES-R CI algorithm picking up pretty well on the initiation. Below are two loops featuring the GOES-R CI product and the subsequent development of radar echoes at the -10 C isotherm. Apologies for the fast loops as the AWIPS tool created these automatically. However, you can see the warmer colors across the LWX CWA develop into radar echoes 10-15 minutes later. The first lightning strike

GOES-R CI product from 1715 -1830 UTC

MRMS -10 C reflectivity from 1828 to 1906 UTC

Simulated Satellite WRF. Cirrus/Insolation issues leading to convection.

Decided to through in a pre-enviornment blog about convection over southern West Virginia and far southwestern Virginia. In this case a plume of cirrus/ cirro-stratus is not showing up in the WRF. Normally that would be rather benign.  However in this case that cirrus is blocking just  enough isolation to prevent storm development over both areas which is keeping storms from popping and having the time to hit the equilibrium level.

Grant H.

Prob Severe Lagging Radar Today

I have been closely monitoring the ProbSvr product today as storms have initially developed over parts of western KY. Analysis of the radar data from the OHX radar data indicates that severe weather would be likely from the storm near Fryer, but the ProbSvr to this point only has around 60% chances.

-JB

Northeast convection

We're operating in Sterling, Virginia, to begin today, but storms fired in Pennsylvania and New York before our shift began. The NOAA/CIMSS ProbSevere model was able to pick up on some of these storms. We can see a cluster of storms below, in Pennsylvania. The featured storm (with the scroll-over) exceeded 50% probability of severe when moderate to strong satellite growth rates were observed (17:54 UTC). The storm was warned by the NWS at 18:08 UTC, when the probability of severe was 97%. A storm to the northwest fluctuated in the 20-35% range before being warned by the NWS. The ProbSevere value then jumped to about 65% soon after the warning. This shows that ProbSevere may give increased confidence and lead time to some storms --- not every storm.

-Cintineo/Sieglaff

NearCast Data Today

If you are following NearCast today, it shows the the Diff. Med Low Theta E that the unstable environment would start over western MD and move across the state into the Delmarva Peninsula.  Behind it some more stable air will move in.  It will be interesting to see if storms follow this track and move from NW to SE along the area and at the same time. ~Vollmar

Got the chance with CI today. YAY!

First chance to really play with the CI tool today in a  partially pre-convective environment. Attempted to grab a gif, but it moves a little too fast. I’ll attempt to write through what I’m seeing here anyways.

At the beginning of these frames the CI tool grabs onto a series of storms across far western NY around 15:15Z. These storms fire and immediately move into north central Pennsylvania Probs with these were roughly 86 to 60%. Good job.  Once the storms fire the CI tool lets go of them almost immediately.

focus then shifts south to 1745Z-1815Z where the colors on the CI then begin to resemble a bowl of fruit loops.  We even get a 91% at 1815 over southeastern Pennsylvania that is moving into th New Jersey and the Delmarva.

This tool seems very good for the synoptic level viewing…. but maybe less so on the mesoscale, as there seems to be a lot of noise above  probability of about 40%. The fact that it leaves fired convection behind and moves onto the new area of interest rapidly might make this tool more useful to SPC than to a WFO by redrawing the attention away from the current convection and towards the next problem area. This could help break a natural psychology problem where/when a forecaster would want to obsess over the current convection and may neglect other developing areas for it. .

Also I would like to add that a graduation of the color on the scale may be better than the current discrete 10% breaks built into the color bar.  I noticed that Percent probs were in increments smaller than 10%. This may help to focus the fruit loop bowl effect more. then again it may not but its worth a try.

Grant H.

Low CI (east) With Initial Activity

Looking at the CI product from GOES-E and it appears to have a marginal handle on the CI ongoing over western KY at the moment. In the loop below the values before CI only reached 45% before a 35 dbz echo was detected by radar. In the last frame of the loop there is a 64% noted within the line of CU, though this has not developed on radar. It will be interesting to see if there is soon higher reflectivitiese in this area.

-JB

GFS Difference in Theta-E and NowCast

Today again appears to be a day where the NearCast model is being degraded by a significant amount of cloud cover over the CWA. I have loaded the difference in theta-e fields from the GFS model to fill in some of the gaps in the model. This appears to do well as a proxi for areas of convective instability in the NearCast model. Using the model difference fields in conjunction with the NearCast can help to fill in areas and allow you to get an idea if the model has a good handle on the situation and provides confidence for later in the day.

-JB