29 May tornadoes and
(click on images to enlarge)
(click on images to enlarge)
SPC had our area outlooked in a "Slight Risk" for Severe Thunderstorms, a couple days before the event occurred. On Wednesday, a warm front began to press into the region. Showers and a few rumbles of thunder, associated with isentropic lift, preceded the warm front early in the day. Then, breaks of sunshine took place even ahead of the warm front. Initially there was no convection; in fact, more debris clouds filled in, obscuring the sunshine at times. By mid afternoon, the surface warm front was draped right over the Capital region, and by late afternoon, it was over our northeastern zones. The right entrance region of the upper 70KT jet moved just to our north...putting our region in a favorable region for moderate ascent. A potent upper level disturbance also worked over the region toward evening. This upper-level disturbance was noticeable on water vapor imagery as a dry tongue punched southward near Lake Ontario.
Mesoscale analysis from SPC was very useful in identifying the severe weather and tornado threat. The environment just ahead of the convection evolved into one of moderate cape and high shear. Most unstable cape values reached 1000-1500 j/kg with 0-1 km SRH of 150-200 m2/s2 and 0-3 km SRH of 200-300 m2/s2. A 0-1 km shear vector of 25-30 was noted effective bulk shear reaching 35-45 kts along with surface based LIs of -2 to -4. Low level lapse rates were 7 to 7.5 C/km with a supercell composite parameter between 4 and 8.
Convection began to initially fire up along the Lake Ontario breeze front between 3-4 PM on Wednesday. Buffalo and eventually Binghamton warned on a few cells, including a tornado warning in Ostego County. This convection continued to progress eastward riding along a low-level theta-e gradient/low-level lapse rate gradient that was aided by a differential heating zone that had set up east-west the Mohawk River valley and into the Capital region. This differential heating zone acted as a secondary warm front helping to focus and reinvigorate the initial convection in Buffalo's CWA that had moved away from the lake breeze. Both Buffalo and Binghamton issued SVRs based on reports of wind damage and marginally-severe hail.
This event further emphasizes the importance of boundaries. There were several boundary interactions that helped enhanced the convection and produce the tornado. The tornadic cell merged with a wall of wind (WOW). This interaction allowed the convective complex to evolve into a line echo wave pattern (LEWP) with a classic northern bookend vortex (BEV). The BEV produced most of the extensive damage in the central and eastern portion of the Greater Capital Region. Further details below...but this looked like a classic LEWP within a mesoscale convective system. -
A discrete cell then developed just west of the Herkimer County line. This cell was able to tap into rich BL moisture and instability, as SBCAPE had crept up to around 1500 J/kg in a 340+K theta-e environment. Forecasters wasted no time and issued a warning for southern Herkimer county based on the fact that the same cell took out a quite a few trees in Syracuse. Some warning thresholds were established from the morning soundings with VIL of the day around 56, 50 dBz core to just over 30,000 feet based on -20 C level and expected storm motion of 35-45 kts from west to east.
The first batch of cells weakened briefly as they advanced further into the Mohawk county. However, a stronger line of cells upstream developed. This line became a mesoscale complex, eventually with a classic book end vorticity. While not large enough to be considered a derecho, it definitely had the look of a mini one. The whole line turned suddenly southeast, and looked to become anchored to the warm front, and possibly a theta-e gradient aloft. The circulation initially created pretty impressive inflow which actually resulted in leading cells to arc back into the circulation. Then an outflow boundary emerged which raced to the cells ahead of it in Mohawk valley, resulting in them becoming quite strong again.
The whole line caught up to the cells invoking a full cell merger. Specifically where wind fields of the line and cells took place, two cells rapidly began rotating and ultimately produced tornados. In fact, our area experience its first EF2 tornado since July 21 as a tornado touched down near Florida in Montgomery County tracking east quickly at interstate speeds, before lifting back up in Rotterdam Schenectady county. The tornado left a swath of destruction including collapsing a large barn in Mariaville removing a person from it. Luckily the person survived but was injured. The tornado also snapped large power lines in Mariaville.
The tornadic supercell transformed into a bookend vorticity and while no additional tornadoes, there was a swath of enhanced wind damage associated with it as it moved across northern Albany County into Renssalear and southern Washington counties.
The second tornado touched down between Summit and east Jefferson in Schoharie County. This was an EF-1 tornado with damage to mostly large trees. This was the first day also since July 21 2003 that our area had two more tornadoes in one day.
There was some speculation that a wall of wind (associated with the bowing echo) actually might have disrupted the tornado circulation before it could have caused any more damage.
The remainder of the meso-scale complex caused plenty of additional sporadic damage of its own throughout the Capital District and south into the northern portions of the Mohawk Valley...mostly tree and wire damages. However a building partially collapsed in the city of Hudson. There were a tremendous number of cloud to lightning strikes with these storms which help cause structural fires and damage and added to more customers knocked out of power (along with the tornadoes and straight line winds).
In conjunction with the highest reflectivity core surpassing the -20C level, when looking at the dual pol products, the first indication that a cell may have hail beginning to form is by looking at the CC. The CC will begin to lower into the 0.9-0.97 range first, coincident with a spike in KDP (if not blocked out by CC <0.90) of around 2-3 deg/km. ZDR values at this time are still high, generally greater than 2 dB, which would signal big drops. Use all tilts and FSI to go up through the updraft to see how far these high ZDR values extend. This is a rough estimate for the strength of the updraft and how much liquid water is being pumped into the midlevels of the storm in the form of the ZDR column. Although CC initially tails off to values that would support possible hail, ZDR values remain high. This may be due to the ZDR column artifact and a delay in hail production as big water droplets are lofted into the updraft/meso. When the ZDR values drop below 2 dB and even less than 1 dB, I know that hail is being formed in the updraft and this is verified by the CC values continuing to drop along with an even bigger increase in KDP up to 3-5+ deg/km. When the ZDR drops off in conjunction with low CC and a high reflectivity core that extends well to -20C this can sometimes allow at least 5-15 minute lead time between warning issuance and time of first report, because the ZDR dropping off signifies hail growth, however there is a delay as the hail must grow big enough in size to escape the updraft and fall to the ground. The higher the ZDR column, the better lead time since a higher ZDR column is correlated to a stronger updraft that can recycle hail longer, especially if a meso is present.
The Vr-Shear analysis was very helpful issuing the TOR warnings. Emphasis was not only placed on the LLVr with the Vr-Shear tool, but cross-sections in FSI were used to determine the the mesoscyclone couplet at low-levels, but also the mesocyclone aloft (Vm). The Tornado near Florida, NY spun up quickly with a clear hook echo and a debris ball signature a few scans later. The 0.5 DEG SRM LLVr was close to 50 kts gate to gate with a shear value of 0.06 s-1 at 0.5 nm across the couplet at 2252Z over Mariaville Lake. The Tornado Warning was issued a few scans earlier at 2242 UTC with the 0.5 DEG SRM with a cyclonic convergence signature with a LLVr of 39 kts with a shear value of 0.04 s-1 just west of Scotch Bush (Florida) where the tornado touchdown about 5 min laters. Eye Witness heard the tornado warning in these locations when the TOR was issued at 2245 UTC. The TVS triggered at 2242 UTC too...but then it disappeared at 2247 UTC as the circulation broadened before reintensifying as it returned at 2252 UTC. This case will be added to the Tornado study. The Schoharie County tornado spun up quickly along the line. It looked like a QLCS type tornado. It was there for a few scans and then disappeared.
Above: Loops of NAM80 850 hPa theta-e (left) and 850 hPa winds (right). There was a theta-e gradient of ≥ 15K that supported enhanced low level forcing and a low level jet segment that tracked through the region.
Above: Loop of the High Resolution Rapid Refresh (HRRR) forecasted reflectivity. The forecast showed a mesoscale convective complex tracking near the Capital District, similar to what actually happneded.
Above: Soundings from Albany, NY at 1200 UTC 29 May (left) and 0000 UTC 30 May (right). Notice the sounding were not as unstable as other surrounding areas. There were no elevated mixed layers.
Above: Soundings from Buffalo, NY at 1200 UTC 29 May (left) and 0000 UTC 30 May (right). Notice the instability and shear. There were no elevated mixed layers.
Above: Soundings from Upton, NY at 1200 UTC 29 May (left) and 0000 UTC 30 May (right).
SPC Mesoscale analyses - Shear and instability supported severe weather.
Above: Loops of SPC mesoscale analyses of (left to right) 3 hour change in CAPE, 6 Km shear, Bulk Richardson Number, DCAPE, and derecho composite.
Above: Loops of SPC mesoscale analyses of (left to right) deep moisture convergence, effective helicity, effective shear, energy helicity index, and hail parameter.
Above: Loops of SPC mesoscale analyses of (left to right) lifted condensation level, lifted index, low-level lapse rates, moisture convergence, and midlevel lapse rates.
Above: Loops of SPC mesoscale analyses of (left to right) mixed layer CAPE, maximum unstable CAPE, NCAPE, Suface-based CAPE and supercell composite parameter.
Above: Loops of SPC mesoscale analyses of (left to right) surface to 2Km winds, surface to 1 Km shear vector, significant tornado parameter, 0-1 Km storm-relative helicity and 0-3 Km storm-relative helicity.
Above: Loops of SPC mesoscale analyses of (left to right) theta-e, NCAPE, VGP shear vector and surface observations.
Above: Loops of SPC mesoscale analyses of (left to right) 850 hPa plot, 700 hPa plot and 300 hPa plot.
Above: Loops of water vapor satellite imagery (left) and visible satellite imagery (right).
Above: Loops of KENX reflectivity with warning polygons overlayed across the region (left) and local tornadic supercell in the Mohawk Valley area (right).
Above: Loops of KENX base velocity (left) and Storm-Relative Velocity (right).
Above: Evolution of the tornadic supercell in the KENX Storm-Relative Velocity.
Above: Four panel KENX Dual Pol analyses of the tornadic supercell showing Base Reflectivity (upper left), ZDR (upper right), KDP (lower left) and CC (lower right).
Above: KENX Base Reflectivity (left), Storm-Relative Velocity (center) and Dual Pol CC (right).
Above: Loop of KENX Base Reflectivity and Storm-relative velocity showing the evolution of the tornadic signature on radar.
Above: Severe weather report plot from the SPC showing tornadic activity in the Great Lakes on 5/28 moved east into New York on 5/29.
Above: Pictures of the tornado from Schenectady (left and center) and a rope tornado near Niskayuna (right).
Above: Track of the Schenectady County Tornado in Google Earth.
Above: Track of the Schoharie County tronado in Google Earth.
Above: Track of the Saratoga County, Vischer Ferry tornado in Google Earth.
Above: Damage pictures from north of Mariaville, NY near the Montgomery/Schenectady County border.
Above: Damage pictures from around Rt. 159 between Mariaville, NY and Rotterdam, NY in Schenectady County.
Above: Damage survey pictures from Schenectady County.
Above: Damage survey pictures from Saratoga County.
Above: Damage pictures in Schoharie County.