05/01/08 multi-faceted severe thunderstorm environment




I was on mids during this event.  With the seemingly best environments for tornadoes shaping up well to my south (off the dryline in the accelerating LLJ axis) and well to my north (on the warm front), I determined that things were just far enough out of reach (annoying!) and opted to watch things unfold from home.  I did consider heading up I-35 when wannabe discrete convection seemed to want to pass north of the airport, which would have afforded me a view of a tornado-warned cell near Plattsburg... but chose not to.  The next tornado-warned cell down the line passed just south of my house, but looked pretty pathetic via SkyFox. 


One interesting thing about the day was the impressively deep/rich low-level moisture which rapidly backed up to the retreating dryline during the late afternoon and evening (see 20Z and 00Z OUN RAOB evolution here)... particularly since that was seen as a huge question mark with regard to severe potential given a cold frontal intrusion into the majority of the GOM 3 days prior.  Though the instability environment ahead of the severe bow echo MCS that moved through Kansas City was probably slightly "elevated"  (requiring the momentum to be pushed through some CINH)... I assume the relatively rich and not well-anticipated moisture that evolved in the lowest couple kilometers "helped" a good amount with the potential instability and thus the severity of the bow.  Note that this moisture wasn't any richer than progged by the NAM-WRF for a few days!  Kind of a throwback to April 21, 2007, when similar moisture questions were raised for a set-up in west TX... yet deep moisture evolved and was sufficient for a few cyclic/damaging tornadic supercells.


I mainly wanted to focus on the zany event up in far northwest IA associated with the warm frontal boundary stemming from the synoptic surface low.  An arc of storms formed on and then moved north of the warm front, aided by a significant and elongated shortwave impulse rotating northward across the area.  Within this arc, a beefy supercell with impressive velocity signatures developed and passed southwest through west of Sioux Falls, producing destructive hail up to softball size but no tornadoes.  Farther east, storm cells with comparatively weaker/smaller reflectivity and velocity structures eventually developed and produced a few tornadoes in far northwest IA.  In particular, the long-track "Rock Valley" EF-2 tornado touched down around 7pm in association with a very young storm cell, with base reflectivity suggesting the storm to be "mostly updraft" with 40dbz echoes just starting to appear (though the cell strengthened with regard to reflectivity and velocity signatures thereafter).  I haven't looked at this event with a magnifying glass so I hate to make assumptions, but my gut feeling is that the Sioux Falls storm failed to tornado due to the surface temperatures falling off pretty quickly with northward extent/time (from ~70 F to ~64 F)... particularly as the warm front seemed to occlude by 00Z (warm sector "pinched off").  Given the relatively warm profile aloft and unimpressive low-level moisture, surface temperatures in the mid 60s F were likely insufficient to allow for appreciable (any?) surface-based instability... especially if moisture depth was an issue farther west.  Thus, the storm may have been drawing its inflow from the 800-700mb level as time went on, a layer which was saturated and moderately unstable when lifting a parcel from that level.  (This "elevated" environment was also sampled by RAOBs farther north at ABR and MPX).  Meanwhile, farther east, temperatures north of the warm front remained a little warmer (near 70 F) with similar dewpoints; assumedly this allowed for the maintenance of modest surface-based instability to the cool side of the warm front and thus tornadogenesis. 


Below is a proximity sounding based off the 00Z NAM initialization at 00Z over Rock Valley Iowa.  I moistened the boundary layer just slightly based on TAMDAR trends at Sioux City, but note that the moisture is still quite modest.  Probably a good thing, or every storm along that arc probably would have been significantly tornadic based on the strong low-level shear.  (Given the fact that the lifted parcel path barely cleared the environmental temperature profile, all the thermodynamic calculations were pretty "fickle" and highly dependent on the low-level moisture quality... so take them with a small grain of salt.)  Note that the hodograph shows a classic sickle shape (just rotated on its side), with intense/backed flow at 1km agl.  I haven't looked into whether these were true supercellular tornadoes, or were more tied to modest low-level stretching/vorticity.


00Z surface analysis



00Z Rock Valley IA modified NAM sounding/hodograph



Sfc T/Td: 68/55F

MLCAPE: 946 j/kg

MLCINH: 35 j/kg

0-3 km MLCAPE: 49 j/kg

MLLCL: 1159 m

MLLFC: 1989 m

storm motion: 160 @ 27 kts

0-1 km SRH: 222 m2/s2

0-3 km SRH: 329 m2/s2



Meanwhile, vigorous convective initiation occurred on the dryline in eastern KS.  This evolved into a chunky line, with more discrete cells on its northern and southern ends (the northern ones I described earlier near KC).  Storms generally struggled to produce long-lived low-level mesos and tornadoes, probably due to marginal low-level shear.  The LLJ began accelerating into southeast KS after 02-03Z, but capping was likely becoming a problem by then.  Farther south, in spite of weaker low-level convergence, the capping thermal ridge subsided enough to result in almost complete CINH removal and explosive instability (as shown previously in OUN RAOB evolution) and attendant isolated initiation on the OK portions of the dryline by 00Z.  The northernmost of these became a significant tornadic supercell over Pawnee/Osage counties between 01Z and 02Z.  Below is the 00Z OUN RAOB modified for conditions over that area at that time.  The 500-1000 m flow accelerated a bit above what was sampled by the RAOB, likely increasing 0-1 km SRH to values of 200-250 m2/s2.


00Z surface analysis



00Z Norman OK RAOB (sounding modified for 01Z conditions at Ralston OK)



Sfc T/Td: 76/64 F

MLCAPE: 3623 j/kg

MLCINH: 42 j/kg

0-3 km MLCAPE: 113 j/kg

MLLCL: 1157 m

MLLFC: 1777 m


storm motion: 216 @ 23 kts

0-1 km SRH: 104 m2/s2 (underestimate)

0-3 km SRH: 195 m2/s2 (underestimate)

0-6 km bulk shear: 45 kts (underestimate)



Throughout the late evening, warm sector capping increased from the southwest as progged, gradually "killing off" the dryline supercells from south to north.  Meanwhile, when intense large scale ascent/upper difluence associated with the primary shortwave trough intercepted the surging/sharpening cold front, the damaging bow echo MCS evolved and raced east-northeast through KC metro.  An EF-2 (bookend vortex?) tornado developed 0.7 miles east of my home in Gladstone MO, destroying a few residences.  I was very relieved to find only minor/isolated tree damage at my home upon getting off work.  As stated above, some CINH to surface parcels likely existed as temps fell toward the mid 60s F ahead of the bow echo... but apparently not enough to significantly impact momentum transfer.  Below is the modified 06Z SGF sounding for conditions in Kansas City.  The sounding is definitely somewhat of a best guess, given that significant thermodynamic changes were likely occuring via intense forcing for ascent (possible dynamic/adiabatic cooling) and a secondary surge of low-level moisture from the south.  Note, I don't have any effective SRH or effective deep layer shear values as our NSHARP won't calculate them.


06Z surface analysis



06Z Springfield MO RAOB (modified for 07Z conditions at Gladstone MO)



Sfc T/Td: 67/60 F

MLCAPE: 1753 j/kg

MLCINH: 77 j/kg

MUCAPE: 2222 j/kg

MUCINH: 17 j/kg


other links:


damage survey from WFOFSD

damage survey from WFOEAX

damage survey from WFOTSA