Severe thunderstorms in northwestern Kansas

August 5th, 2013
GOES-13 0.63 µm visible channel images with overshooting top detection icons (click image to play animation)

GOES-13 0.63 µm visible channel images with overshooting top detection icons (click image to play animation)

 

AWIPS images of 1-km resolution GOES-13 0.63 µm visible channel images with automated overshooting top detection icons (above; click image to play animation) showed the development of a large mesoscale convective system across northwestern Kansas during the afternoon hours on 05 August 2013. Note that the surface air temperature at Goodland, Kansas (KGLD) dropped from 91º F at 19 UTC to 69º F at 20 UTC (with southeasterly winds gusting to 34 knots).

4-km resolution GOES-13 10.7 µm IR channel images with automated overshooting top detection icons (below; click image to play animation) revealed the formation of a very broad and well-defined “enhanced-V” storm top signature, with cloud-top IR brightness temperatures becoming as cold as -80º C at 23:15 UTC. Overshooting top detection began after 19:45 UTC, once GOES-13 cloud-top IR brightness temperatures became -71º C or colder.

GOES-13 10.7 µm IR channel images with overshooting top detection icons (click image to play animation)

GOES-13 10.7 µm IR channel images with overshooting top detection icons (click image to play animation)

During the early stages of convective development, GOES-13 10.7 µm IR images combined with 15-minute cloud top cooling rates (below) showed the development of significant cloud top cooling rates along the Colorado/Kansas border area at 18:15 UTC (prior to the formation of weak, brief landspout tornadoes just northeast of Goodland, Kansas during the 18:28-18:40 UTC period) — a maximum cloud top cooling rate of 35.3º C in 15 minutes was detected at 18:45 UTC.

GOES-13 10.7 µm IR channel images with Cloud Top Cooling Rates

GOES-13 10.7 µm IR channel images with Cloud Top Cooling Rates

A comparison of 1-km resolution Soumi NPP VIIRS 11.45 µm IR channel and 4.-km resolution GOES-13 10.7 µm images (below) demonstrated the ability of higher spatial resolution VIIRS data to detect much colder IR brightness temperatures associated with the more vigorous overshooting tops (-82º C on VIIRS, vs -71º C on GOES). In addition, a northwestward GOES image parallax shift was seen, due to to the large viewing angle of the GOES-13 satellite positioned at 75º W longitude. Shortly after the time of these images, this storm produced hail of 1.0 inch in diameter at 20:11, 20:21, and 20:54 UTC (SPC storm reports)

Suomi NPP VIIRS 11.45 µm IR and GOES-13 10.7 µm IR images

Suomi NPP VIIRS 11.45 µm IR and GOES-13 10.7 µm IR images

Comparisons of 1-km resolution POES AVHRR Cloud Top Temperature and Cloud Top Height products at 20:21 UTC (above) and 21:16 UTC (below) showed that the coldest cloud top temperatures (-85º C and -87º C, respectively) were seen in regions where the maximum cloud top height values were generally around 15 km. It is likely that the most vigorous overshooting tops associated with the coldest cloud top temperature values were as much as 2-3 km higher than this mean 15 km thunderstorm anvil cloud top height. Conversely, cloud top height values were around 13 km in the “warm wake” region immediately downwind of the coldest overshooting tops.

Convective outflow boundary initiates new convection over Kansas

July 31st, 2013
GOES-13 0.63 µm visible channel images (click image to play animation)

GOES-13 0.63 µm visible channel images (click image to play animation)

AWIPS images of 1-km resolution GOES-13 0.63 µm visible channel data (above; click image to play animation) showed an undular bore marking a surface-based convective outflow boundary which formed over southwestern Nebraska early in the day on 31 July 2013 — this outflow boundary then propagated southwestward during the day and acted as a focus for the formation of severe thunderstorms over southwestern Kansas later that afternoon (SPC storm reports).

The southwestward-propagating outflow boundary / undular bore feature was also seen on 4-km resolution GOES-13 6.5 µm water vapor channel imagery (below; click image to play animation).

GOES-13 6.5 µm water vapor channel images (click image to play animation)

GOES-13 6.5 µm water vapor channel images (click image to play animation)

The GOES-13 imager 6.5 µm water vapor channel weighting function calculated using the 12 UTC rawinsonde data from Dodge City, Kansas (below) indicated that the weighting function peaked much lower in the atmosphere (around 500 hpa) than normal — this allowed a thermal signal (albeit a faint one) of the boundary layer convective outflow boundary cloud features to be seen on the water vapor imagery.

GOES-13 water vapor channel weighting function plot (using Dodge City, Kansas rawinsonde data)

GOES-13 water vapor channel weighting function plot (using Dodge City, Kansas rawinsonde data)

A good view of the undular bore (which was trailing the leading edge of the convective outflow boundary) could be seen on a comparison of 1-km resolution MODIS 0.64 µm visible channel and 11.0 µm IR channel images at 17:05 UTC (below).

MODIS 0.64 µm visible channel and 11.0 µm IR channel images

MODIS 0.64 µm visible channel and 11.0 µm IR channel images

New thunderstorms formed along the old convective outflow boundary (as it encountered increasing instability across southwestern Kansas during the afternoon hours), as seen on 1-km resolution Suomi NPP VIIRS 0.64 µm visible channel and 11.45 µm IR channel images (below). This storm was producing 1-inch diameter hail and wind gusts of 50-60 mph.

Suomi NPP VIIRS 0.64 µm visible channel and 11.45 µm IR channel mages, with SPC reports of hail

Suomi NPP VIIRS 0.64 µm visible channel and 11.45 µm IR channel mages, with SPC reports of hail

A comparison of the 1-km resolution Suomi NPP VIIRS 11.45 µm IR channel image with the corresponding 4-km resolution GOES-13 10.7 µm IR channel image (below) displayed an unusually large 30-degree difference between the coldest cloud-top IR brightness temperatures of the northernmost of the newly-formed thunderstorms in Kansas (-92º C on VIIRS, vs -62º C on GOES).

VIIRS 11.45 µm IR channel and GOES-13 10.7 µm IR channel images

VIIRS 11.45 µm IR channel and GOES-13 10.7 µm IR channel images

 

Cloud-Top Cooling Rate

May 30th, 2012
GOES Sounder DPI Lifted Index 1600 UTC on May 30 2012

GOES Sounder DPI Lifted Index 1600 UTC on May 30 2012

Satellite descriptors of the atmosphere over the central High Plains on May 30th suggested an atmosphere ripe for convection. For example, the GOES Sounder DPI Lifted Index showed an axis of instability from central Oklahoma to northwest Kansas, with widespread values (yellows, reds, purples) at or below -4. (The unstable area overlaps nicely with the slight risk for the day diagnosed for the day by the Storm Prediction Center). The NearCasting product (below, and available on-line here) suggests that convective instability will persist over southern Nebraska and extend southward into Oklahoma for most of the day.

GOES-13 NearCast product (click image to play animation)

GOES-13 NearCast product (click image to play animation)

GOES-13 Visible Imagery and UW Cloud-Top Cooling Rate

GOES-13 Visible Imagery and UW Cloud-Top Cooling Rate (click image to play animation)

Given this type of environment, what can cloud-top cooling rate tell you? The Cloud-Top Cooling Rate is a satellite-based product that diagnoses how quickly the growing convective towers are cooling with time, and strong cooling means rapid vertical convective growth. The strongest cooling is where the strongest convective growth is occurring, and that cooling is well-correlated with the subsequent development of NEXRAD signatures (MESH, maximum VIL, reflectivity at -10 C, etc.) The loop above shows GOES-13 visible imagery with the cloud-top cooling rate superimposed. (Real-time imagery of the cloud-top cooling rate product is available here) The The cloud-top cooling rate product highlights a persistently and quickly growing convective feature near McCook, Nebraska before and shortly after 1700 UTC. By 1800 UTC, that feature is a severe thunderstorm warned for high winds and hail.

The UW Cloud-Top Cooling Rate product is being evaluated at the Hazardous Weather Testbed taking place this month. At the HWT Blog, there are many examples of using the product in concert with knowledge of the synoptic and mesoscale conditions to anticipate strong thunderstorm development.

SPC storm reports show that the system with the strong Cloud-Top Cooling Rate subsequently produced 1-inch hail near Holdrege.

Aircraft turbulence east of Florida on August 9 2011

August 9th, 2011
GOES-13 Visible with Airplane Positions

GOES-13 Visible with Airplane Positions

Turbulence associated with a developing line of convection over the Atlantic Ocean east of Florida was severe enough on August 9th to cause five injuries on a Miami-to-Washington DC flight and force an unscheduled landing of the 737-800 aircraft in Charleston, SC. The GOES-13 image, above, shows the path that the aircraft took through a line of developing convection east of Florida (Click here for a flash-based animation). The time that the satellite was scanning the region east of Florida was between 1935 and 1936 UTC (the nominal time of the image, 1932 UTC, refers to the first line scanned in by GOES-13; it takes almost 5 minutes to completely scan North America. See this NESDIS website for the normal scanning schedule).

Animations of visible imagery, 10.7-micrometer infrared imagery, and 6.5-micrometer infrared imagery (the so-called ‘water vapor channel’) all show a similar evolution, namely strong thunderstorms at the coast of Florida at the start of the loop followed by the development of a line of thunderstorms northeastward. It is through this developing line that the airline penetrated. (The flight path is here).

GOES Satellite data are routinely monitored to detect both the initiation of convection, and the presence of Overshooting Tops and Thermal couplets, the latter two features being well-correlated with severe weather and turbulence. Detection suffers, however, because of the relatively poor spatial and temporal resolution afforded by routine GOES scanning. What was detected on this day?

Overshooting tops were detected over the Florida peninsula at 1715 UTC and at 1732 UTC. The top at 1732 UTC is quite apparent in both the visible and infrared loops. Overshooting tops were not detected in the area again before the turbulence event, but their detection prior to the event suggests an airmass with the potential for strong convective development.

Convective Initiation (CI) was flagged along the southwest-to-northeast line of developing convection at 1832 UTC and 1845 UTC, roughly an hour before the turbulence event. (See this link for all Blog Posts on Convective Initiation). Note that the CI detection in this case — occurring — means that glaciation of the clouds has started. CI is drawing the eye to the convective towers that are growing most rapidly; therefore, their tops cool most quickly. It is to these growing cells that a forecaster must pay attention, particularly when they appear in an environment that will sustain overshooting tops.

(Here is an article on this flight from The Aviation Herald).