Fun with the TDWR and Tropical Cyclone Tornadoes

TRDU TDWR 0.5 SRM image from 2326 UTC 06 September showing cycloninc circualtion near Rolesville NC

TRDU TDWR 0.5 SRM image from 2326 UTC 06 September showing cyclonic circualtion near Rolesville NC

Tornadoes associated with tropical cyclones (TC) typically develop rapidly, with no classical descending mesocyclone. The rotational signatures may last for only 1 or 2 WSR-88D volume scans and require a quick warning decision. Situational awareness along with knowing the environment and how it is changing is critical in these environments. The environment on Tuesday, September 6, 2011 was in many ways consistent with many TC tornado events. These situations are often very challenging for forecasters who balance the need to issue timely and accurate tornado warnings with the large number of rotating thunderstorms that are potential tornado producers.

Tropical Storm Lee moved into the lower Mississippi Valley on Sunday and Monday and then into the southern Appalachians on Tuesday. A cold front dropped south into northern North Carolina on Tuesday morning as the remnant low pressure center moved into the North Carolina mountains with a trailing cold front extending south into South Carolina and eastern Georgia.  Across central North Carolina, specifically near Wake County, a warm and very moist air mass was in place with surface dew points in the lower 70s. Surface based CAPE and mixed-layer CAPE values were analyzed around 500 J/Kg.  Strong south-southeasterly surface winds veered to southerly at around 40kts at 700 and 500 hPa producing sufficient shear for organized convection and some supercells.  Analyzed 0-1km SRH values were analyzed around 250 m2/s2 with the Sig Tor parameter approaching 0.5.

The NWS Raleigh issued more than a dozen tornado warnings on Monday and Tuesday with only three confirmed tornado touchdowns noted. Forecasters were frequently concerned about short lived rotational couplets that would quickly develop and then more often than not, dissipate without producing a tornado.

One particularly interesting cell moved across northeastern Wake County North Carolina near the town of Rolesville between 700 and 745 PM on Tuesday, September 6th. The animation below is a four panel loop of radar data from both the TRDU (TDWR) on the top row and the KRAX (WSR-88D) on the bottom row from 2315 UTC to 2335 UTC on September 6th.  The TRDU 0.2 degree base reflectivity data is shown in the upper left and the 0.2 degrees base velocity imagery is in the upper right with both products available every minute. The KRAX 0.5 degree base reflectivity data is shown in the lower left and the 0.5 degrees base velocity imagery is in the lower right with both products available every 5 minutes.  A similar loop of storm relative velocity imagery which does not highlight the circulation as well is shown at the bottom of the post.

Four panel loop of radar data from the TRDU (TDWR) and the KRAX (WSR-88D) from 2315 UTC to 2335 UTC on 06 September

Four panel loop of radar data from the TRDU TDWR and the KRAX WSR-88D from 2315 UTC to 2335 UTC on 06 September. The TRDU 0.2 degree base reflectivity data is shown in the upper left and the 0.2 degrees base velocity imagery is in the upper right with both products available every minute. The KRAX 0.5 degree base reflectivity data is shown in the lower left and the 0.5 degrees base velocity imagery is in the lower right with both products available every 5 minutes.

The cell as viewed on the KRAX WSR-88D (bottom row of the animation) was rather innocuous with a modest vertical reflectivity structure and only limited amounts of CG lightning. A brief rotational couplet can be seen on the 2325 UTC volume scan just to the southwest of Rolesville with the weak circulation dissipating on the next scan. The TRDU TDWR with its unique VCP provides radar data on the lowest elevation scan every minute. This increased temporal resolution shows the circulation quickly developing and then dissipating.  Inspection of the TRDU and KRAX storm relative velocity imagery at 2326/2325 UTC noted 30/27 kts of rotational velocity for the TRDU and KRAX respectively.  While this storm did not produce a tornado, the NWS Raleigh received multiple reports of a funnel cloud near Rolesville. This storm was one of several dozen storms that forecasters were monitoring during the afternoon and evening and indicative of the difficult warning decisions that take place during these events.

SPC mesoanalysis of surface T, Td, and PMSL from 23 UTC 06 September

SPC mesoanalysis of surface T, Td, and PMSL from 23 UTC 06 September – click to enlarge

SPC mesoanalysis of surface based CAPE from 23 UTC 06 September

SPC mesoanalysis of surface based CAPE from 23 UTC 06 September – click to enlarge

SPC mesoanalysis of 0-1km SRH from 23 UTC 06 September

SPC mesoanalysis of 0-1km SRH from 23 UTC 06 September – click to enlarge

Four panel loop of radar data from the TRDU TDWR and the KRAX WSR-88D from 2315 UTC to 2335 UTC on 06 September. The TRDU 0.2 degree base reflectivity data is shown in the upper left and the 0.2 degrees storm relative velocity imagery is in the upper right with both products available every minute. The KRAX 0.5 degree base reflectivity data is shown in the lower left and the 0.5 degrees storm relative velocity imagery is in the lower right with both products available every 5 minutes.

Four panel loop of radar data from the TRDU TDWR and the KRAX WSR-88D from 2315 UTC to 2335 UTC on 06 September. The TRDU 0.2 degree base reflectivity data is shown in the upper left and the 0.2 degrees storm relative velocity imagery is in the upper right with both products available every minute. The KRAX 0.5 degree base reflectivity data is shown in the lower left and the 0.5 degrees storm relative velocity imagery is in the lower right with both products available every 5 minutes.

Advertisements
This entry was posted in TC Tornadoes. Bookmark the permalink.

One Response to Fun with the TDWR and Tropical Cyclone Tornadoes

  1. Steve Keighton says:

    Good stuff JB, and thanks for posting so quickly. Our issues were mostly on Monday, and we also issued well over a dozen TOR warnings, and have had only two confirmed tornadoes (one, the briefest of brief EF0 touchdowns – path length basically the size of the gas station it hit in Cana VA!). The other tornado was an EF1 near Traphill NC in Wilkes CO NC, with a 1.3 mile length). Both of these were near the base of the Blue Ridge, and fairly close the the surface boundary, but I really have not had a chance to go back and look at many details. Neither one had the quite the strength of rotational signatures that you show here (although at least in the case of Wilkes Co we are at an appreciable distance and looking 6kft aboe the ground). And we had other signatures that were more impressive, and some of these more clearly in the warm sector, while others were probably more in the cool sector, but with terrain we often find it tricky knowing what the chances are of something still mixing down (it’s happened before when we have not warned for that reason). Clearly an area that needs more study (and ideally better observations), and this is a great case to examine more thoroughly.

    Finally, with the fact that one of our tornadoes was so short-lived, but just happened to hit a case station, I frequently wonder how many of the other circulations may ahve done the same thing out in the woods somewhere, and without flying over all areas with a helicopter, we may never know what actually happened. We discovered two other tornado touchdowns after the April 27-28 outbreak many days after the fact. We need to be careful about any correlations we try to make between radar and mesoscale environment characteristics, and tornado likelihood, yet I think we can take cases like this and examine both of these things carefully and make an attempt to refine some of the important factors that can help us with warn/no-warn decisions. Quite a challenge though!

    Steve

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s