17 August 2007 “Cold Season” severe weather outbreak

(Click on thumbnails for larger images and movies)

 

On 17 August 2007, there was a “surprise” severe weather outbreak across upstate NY.  The day before, severe weather occurred as a dew point boundary tracked through upstate NY dropping surface dew points from the mid to upper 60s, to the 40s in some parts of upstate NY.  The dew point boundary stalled and set up across the Berkshires, Taconics and Catskills.  During the afternoon of 17 August, a strong upper-level system was forecasted to track through southern Canada and northern New England, and an upper-level jet segment was forecasted to track south of the region, putting our region in the left exit region of the upper jet.  Boundary layer winds (around 850 hPa) were forecasted to strengthen over the region as well.  So with the combination of the low-level forcing associated with the strong surface dew point boundary and boundary layer jet segment, and the upper dynamics with the cooling aloft and favorable jet dynamics, a low-topped convective complex formed around Lake Ontario and swept through our region. 

There was no outlook for our region from SPC. There was an SWOMCD for the storms in the BUF and BGM area, stating that the storms would diminish around sunset. Then, when the storms blew up east and south of Albany after 9 PM, another SWOMCD was issued.

Thunderstorms formed in PA during the early afternoon, and tracked through
New Jersey, southern New York and southern New England. An area of showers and thunderstorms formed around the Great Lakes, in a region where some convective instability existed, especially with the warm lakes and the rapid cooling aloft. The BUF and PIT soundings showed some weak instability, as did OKX, but the ALB sounding was stable, and all soundings indicated considerable dry air through the atmosphere.  The 0000 UTC 18 August soundings showed conditional instability at ALB, but the 0000 UTC area soundings were available only during the event, just as warning decisions were being made.

There was a strong upper level jet segment that was tracking into our region. During the afternoon, our region was in the left exit region of the upper jet. It is quite likely that the enhanced upper divergence helped support/sustain the convection.

The showers and thunderstorms tracked east and southeast, with the northern half of the area weakening as it headed into the
Adirondacks, probably due to the movement into an environment with dew points in the 40s. The southern half of the area began to form a broad bowed appearance, but reflectivities throughout the area of showers and storms were quite weak (<50 dBZ). BGM and BUF issued warnings and were receiving reports of trees down. Since the convection just became an area of rain over the Adirondacks and Schoharie Creek area, we did not warn for those areas. However, our radar showed around 40 KT in the radar Base Velocity. Hank Crofoot reported a tree and a couple of wires down, with a wind gust around 35 MPH.

We continued with NOWs, including gusty winds. The strongest part of the area of showers and thunderstorms continued to weaken, and by the time the shallow convection reached
Ulster County, there was no lightning indicated. We warned for Ulster County because there was near 50 dBZ in the forward advancing line, and BGM had reports of trees down in Delaware County. We received a couple of reports of a couple of trees down around Schenectady (including one from Brian M.) along with some wires down, and figured that the 40 MPH winds were causing some sporadic problems, but not enough for SVR. We continued with the gusty winds worded into NOWs. Also, we issued an AWW just as the wind gusts were getting to the airport.

We received a report from a spotter in Colonie of a 30 MPH wind gust, and George called from
Clifton Park with a roughly 40 MPH wind gust. ALB airport then reported a gust to 46 KT, and based on the previous reports from Schenectady, we decided that we should warn for the advancing line of 40-45 dBZ echoes that had already tracked through most of Albany County. As the leading edge of the showers and storms tracked east and south of Albany, it strengthened, and we began receiving some calls about trees down in the Albany area. There was still no lightning indicated, although we here at the office did see one distant flash of what must have been LTGIC. ALB airport reported VCTS briefly after the wind gust.

We warned down the line in locations east and south of
Albany, and received reports of trees down in all locations warned, except NW Ulster County. The line of convection strenghtened more as it encountered the dew points in the 60s in the Taconics and Berkshires. There was frequent lightning indicated with the strong convection there. Even ALB dew point increased into the lower 50s when the convection passed through. There was barely a trace of rain in many locations, with up to about a tenth of an inch of rain, until the storms tracked southeast of ALB and strengthened, with reflectivities of 55-60+ dBZ.

What we learned:

The day before this event, operational models did provide some hints that strong convection was possible over our region. There was a concensus that a strong surface dew point gradient would be over our region, although there was disagreement in exactly where the dewpoint boundary would set up. There was also a consensus that the upper jet segment would be tracking into the region, but the exact position and timing was not completely clear, and small changes in timing and position could have big implications on whether we would be in a favorable region of the upper jet. All MOS guidance showed chance POPs for POU and HFD, with slight chances for ALB, PSF and AQW.

On the day of the event, the morning data suggested cooling aloft was advecting into the region, and there was instability out toward BUF and PIT. There was a strong gradient of dew points over the region with 40s to the north and 60s to the south, suggesting there may be enough instability and forcing for strong convection. Once the convection formed over the
Great Lakes and in Canada, it was unclear whether the shallow convection could sustain itself into the lower dew point air. The 00Z soundings showed some "inverted V" type characteristics, but the instability was not indicated in the soundings. With convection present, obviously there was enough instability. There was also enough dry air aloft to enhance downdrafts, and with nearly 40 Kt of wind seen in area VWPs, it supported potentially high wind gusts in any convection.

The convection was much like cold season convection, shallow and low reflectivities. This was very uncharacteristic of August. The low reflectivities were deceiving, and in the future, reports of trees and wires down associated with any system that even has the hint of exhibiting convection, or looked more convective in the past (I.E. higher reflectivities), should be closely analyzed. The radar Base Velocity did not show severe winds until the stronger convection began to develop over the ALB area, but 40 Kt winds should alert a forecaster. A considerable portion of BGMs severe weather reports were associated with a well-defined 50+ Kt wind maximum on their radar in the 4-bit Base Velocity, but not all. We kept an eye on our radar for winds closer to 50 Kt.

Mesoscale analyses were consulted, but the questionably convective nature to the rain, along with the low dew points advecting south suggested any strong wind potential would be with the strongest convection, in southern areas, where there was a strong dew point gradien, and the Taconics and southern Berkshires had dew points in the 60s. There was some convergence seen in the wind barbs, but 3-hour pressure rises and LAPS soundings were not consulted due to the lack of expectation of severe weather, with no SPC outlook and such low surface dew point suggesting lack of instability, plus the seemingly weakening nature of the northern portion of the area of showers and storms.

This event may be interpreted as a gradient wind that evolved into a convective wind gust once the convection re-established itself. In other words, the
Albany and Schenectady wind damage may have been non-thunderstorm winds, but the fact that it was such a short burst of wind, may support the assertion that it was always associated with convection.

 

 

Figure 1.  Upper air sounding from Manawaki, ON from 1200 UTC 17 August.  Note the conditional instability and the warm advection wind profile.

 

a) b)

 

Figure 2.  Upper air soundings from Albany, NY (KALB) for a) 1200 UTC Friday 17 August, and b) 0000 UTC 18 August.  Note the relatively stable soundings and weak wind shear.

 

a) b)

 

Figure 3.  Upper air soundings from Buffalo, NY (KBUF) for a) 1200 UTC Friday 17 August, and b) 0000 UTC 18 August.  Note the relatively stable soundings and weak wind shear.

 

 

Figure 4.  Water vapor satellite imagery at 2210 UTC 17 August.  Note the enhanced moisture over western and northern NY.  Click on the image for a movie of the water vapor imagery.

 

 

Figure 5.  Visible satellite imagery with lightning overlay at 2155 UTC 17 August.  Note the convection over western NY, and a separate cluster in northern NJ.  Click on the image for a movie of the visible imagery.  Note in the movie, that the lightning diminished to just a few strikes by 0035 UTC 18 August.

 

 

Figure 6.  MSAS surface dew point analysis at 2200 UTC 17 August.  Click on image for a movie.  Note in the movie, the impressive dew point gradient across the Berkshires, Taconics and Catskills.  There were also some relatively higher dew points in western NY representing a small area of warm advection ahead of the strong upper impulse.

 

 

Figure 7.  MSAS rise/fall couplets at 2200 UTC 17 August.  Click on image for movie.

 

 

Figure 8.  LAPS CAPE analysis.  Click on image for movie.  Note in the movie that the highest CAPEs remain east and south of the capital district.

 

 

Figure 9.  Severe Weather Outlook for day 1 from the Storm Prediction Center.  Note only the New York City area is in the slight risk for severe weather.

 

 

Figure 10.  Initialized/observed 500 hPa heights and vorticity from the 0000 UTC 18 August NAM80.  Click on image for movie.  Note in the movie, the difference in timing, intensity and track of the upper vorticity center.

 

 

 

Figure 11.  Initialized/observed wind barbs and isotachs at 250 hPa at 0600 UTC 18 August (click on image for larger image).  Click on image at right for movie of forecasts from the 1200 UTC run of the NAM80.  Note the location of the observed jet maximum further east at 0600 UTC as opposed to the forecasted evolution of the upper jet segment in the movie of the NAM80 forecast.

 

 

Figure 12.  Wind barbs and isotachs at 850 hPa initialized/observed from the 0000 UTC 18 August NAM80.  Click on image for movie.  Note in the movie, the differences in the initialized/observed winds (top), the NAM80 forecasted winds (lower left), and the SREF winds (lower right).  Also note that the SREF winds forecast was closest to what actually happened.

 

a) b) c)

 

Figure 13.  LAPS soundings at 0000 UTC 18 August from a) Poughkeepsie, b) Schenectady, c) northwestern Ulster County.  Note the only unstable sounding was at Poughkeepsie where the surface dew points were greatest.

 

 

Figure 14.  Radar reflectivity at 0.5˚ from KTYX early in the event.  Click on the image for a movie.

 

a) b)

 

Figure 15.  Radar Base Velocity at 0.5˚ from KTYX at a) 2247 UTC 17 August, and b) 2259 UTC 17 August.  Note the high base velocities just east of the radar site that weakened slightly as the storms moved east.  The weakening winds might be attributed to the radar returns being slightly higher above ground level.

 

 

Figure 16.  Radar reflectivity at 0.5˚ from KBGM early in the event.  Click on the image for a movie.

 

a) b)

 

Figure 17.  Radar Base Velocity at 0.5˚ at KBGM from a) 2249 UTC 17 August, and b) 0033 UTC 18 August.  Note the high winds northwest of KBGM that tracked through and southeast of KBGM.

 

 

Figure 18.  Storm Prediction Center mesoscale discussion at 2205 UTC 17 August addressing the convection in western and central NY.

 

 

  

 

Figure 19.  Radar reflectivity at 0.5˚ from KENX at 00001 UTC 18 August.  Click on images for movies of the convective line in different stages.  Note the decrease in reflectivity, but the bowing leading edge of the showers and storms.  Also note the speed of movement of the radar echoes, indicative of the strong steering wind flow.  The convection strengthened east and south of the Capital District, with much higher reflectivities developing.

 

 

Figure 20.  Radar Base Velocity at 0.5˚ from KENX at 0001 UTC 18 August.  Note the development of 40Kt+ winds northwest of the Capital District, then 50Kt winds appearing just east of the radar as the leading edge of the showers and storms tracked through the Capital District.  The strong wind signature continued as the convection strengthened east and south of the Capital District.

 

 

Figure 21.  Radar echo tops from KENX at 0105 UTC 18 August.  Note the echo tops of 20,000-25,000ft, illustrating the low-topped nature of the convection.

 

 

Figure 22.  Storm Prediction Center mesoscale discussion at 0122 UTC 18 August addressing the convection in southern NY and southern New England.

 

Map of 070817_rpts's severe weather reports