An outbreak of severe thunderstorms occurred during the afternoon hours of 03 March 2019, which produced large hail, damaging winds and tornadoes (SPC storm reports). 1-minute Mesoscale Domain Sector GOES-16 (GOES-East) “Red” Visible (0.64 µm) images (above) showed the development of numerous thunderstorms along and ahead of an advancing cold front (surface analyses); many of those storms exhibited well-defined overshooting tops. Tornado track summaries for Alabama and Georgia are available from NWS Birmingham and NWS Atlanta.

The corresponding GOES-16 “Clean” Infrared Window (10.3 µm) images are shown below. Cloud-top infrared brightness temperatures cooled to around -70ºC (darker black enhancement) with many of the stronger storms — judging from rawinsonde data from Birmingham, Alabama (at 12 UTC) and Peachtree City, Georgia (at 18 UTC), this roughly corresponded to an air parcel rising significantly past the tropopause to an altitude of at least 15 km.

An area which included western Lee County (located in far eastern Alabama, adjacent to the Georgia border) was highlighted by a SPC MCD that was issued at 1900 UTC. Beginning about an hour later, 2 large tornadoes producing EF2 to EF4 damage moved across southern Lee County — initially beginning around 2000 UTC, then again beginning around 2050 UTC — and the formation of prominent overshooting tops was evident in GOES-16 Visible and Infrared imagery (below). Station identifier KAUO in Lee County is the Auburn-Opelika Airport. (side note: later, around 2204 UTC, the Weedon Field Airport KEUF METAR site to the south of Lee County was directly hit by a separate EF2 tornado, and rendered inoperative)

In a plot of the GOES-16 “Clean” Infrared Window coldest brightness temperature for the EF4-tornado storm’s overshooting top as it moved from Macon/Lee Counties in Alabama to Muscogee/Harris/Talbot Counties in Georgia (below), 3 distinct periods of cooling/warming occurred — with the warming indicative of a temporary collapse of the overshooting top pulse. The first (and largest-magnitude) cold/warm pulse (-70.3ºC to -65.6ºC) occurred from 1953-1959 UTC — just prior to the beginning of the Beauregard-Smiths Station EF4 Tornado at 2000 UTC. A second cold/warm pulse (-70.8ºC to -66.9ºC) occurred from 2006-2012 UTC, with a third (-70.0ºC to -66.0ºC) from 2015-2022 UTC. At 2029 UTC the long-track tornado then crossed into Muscogee County in Georgia, producing EF3 damage.

The NOAA/CIMSS ProbSevere product (below) displayed a high tornado probability for the cells that approached Lee County, as discussed by the Hazardous Weather Testbed. The ProbSevere model incorporates GOES-derived Normalized vertical growth rate and Cloud-top glaciation rate as 2 of its predictors.

A comparison of Aqua MODIS Visible (0.65 µm) and Infrared Window (11.0 µm) images along with the Total Precipitable Water derived product at 1836 UTC (below) showed that a few large thunderstorms had begun to develop by that time; TPW values were as high as 43 mm (1.7 inches) over far southwestern Georgia.

An overpass of the Suomi NPP satellite around 1850 UTC provided NUCAPS soundings in non-cloudy areas (above). One of the Good quality (green) NUCAPS soundings in the pre-storm environment was located over southwestern Georgia (circled in magenta) — it showed a Most Unstable CAPE value of 1264 J/kg, with a Lifted Index value of -4 (below).

The GOES-16 All Sky CAPE product (below) showed a trend of destabilization across southern Alabama and southern Georgia during the 5 hours leading up to the fatal tornadoes in Lee County AL.

===== 05 March Update =====

A toggle between before/after (24 February / 05 March 2019) Terra MODIS True Color and False Color Red-Green-Blue (RGB) images from the MODIS Today site (above) showed subtle evidence of portions of a tornado damage path — presumably that of the EF4 tornado that began in/near Lee County, Alabama and ended in far western Georgia. Click an additional time on the image to view at full magnification.

Sentinel-2 True Color images (below) provided a higher-resolution view of the tornado damage path. Imagery courtesy of Sentinel Hub.

The path of the #tornado that hit #Alabama last March 3 is clearly visible from images of #Sentinel2 ?. pic.twitter.com/GLatHrIQ5r

— Annamaria Luongo (@annamaria_84) March 7, 2019

34 tornadoes have been confirmed across Alabama, the Florida panhandle, Georgia and South Carolina from the Sunday March 3, 2019 outbreak. One tornado tracked for nearly 69 miles across AL (EF-4) into GA (EF-3). There were 1 EF-4, 1 EF-3, 7 EF-2, 19 EF-1 and 6 EF-0 tornadoes. pic.twitter.com/qjSQV5poKw

— NWS Eastern Region (@NWSEastern) March 6, 2019

View only this post Read Less

GOES-16 (GOES-East) “Clean” Infrared Window (10.3 µm) images (above) revealed unseasonably cold surface infrared brightness temperatures in the -30 to -40ºC range (dark blue to green color enhancement) across much of Montana on the morning of 03 March 2019. Overnight low temperatures at a number of first-order reporting sites were colder than -30ºF (-34ºC) — with the coldest location across the state (and the entire US, including Alaska) dropping to -44ºF (-42ºC). This was only 1ºF warmer than the all-time record low for Montana during the month of March (record low: -45ºF at Glasgow in 1897, and at Fort Logan in 1906). A few daily/monthly cold temperature records were set in the Great Falls and Billings areas.

A sequence of 3 consecutive VIIRS Infrared Window  (11.45 µm) images (from NOAA-20 and Suomi NPP) with a different color enhancement is shown below — the red shades indicate surface infrared brightness temperatures of -40ºC and colder.

===== 04 March Update =====

On the morning of 04 March, GOES-16 “Clean” Infrared Window images (above) showed a more localized pocket of very cold surface infrared brightness temperatures over the southwest part of Montana. The coldest measured surface air temperature was -46°F at Elk Park (at an elevation of 6292 feet) —  which, if certified, will establish a new all-time March low temperature record for the state of Montana.

A sequence of 3 consecutive VIIRS Infrared Window images from NOAA-20 and Suomi NPP (below) revealed surface infrared brightness temperatures as cold as -46ºC or -51ºF within the red-enhanced areas.

View only this post Read Less

Large bushfires burning in the southern portion of the state of Western Australia produced three pyroCumulonimbus (pyroCb) clouds on 01 March 2019. JMA Himawari-8 “Red” Visible (0.64 µm) images (above) showed that the pyroCb clouds drifted southeastward after formation.

Himawari-8 “Clean” Infrared Window (10.4 µm) images (below) further revealed the 3 distinct pyroCb pulses — 2 originating from the southernmost fire located near 29.5ºS / 124.4ºE, and a smaller one originating from a fire located farther to the northwest. Cloud-top infrared brightness temperatures cooled to the -59 to -63ºC range for the pair of larger pyroCbs (which was close to the tropopause temperature of -64ºC on Perth soundings: plot | data) with temperatures reaching -51ºC with the smaller northernmost pyroCb. Also apparent was a surge of cooler air moving northeastward behind a surface trough, whose arrival appeared to coincide with the pyroCb formation. A time series of surface data from Forrest (YFRT) clearly showed the arrival of the cool, moist air behind the trough.

As shown using RealEarth, an overpass of the Suomi NPP satellite provided a more detailed view of the first (and largest) pyroCb at 0537 UTC (above), with NOAA-20 capturing the second pyroCb cloud about an hour later at 0628 UTC (below). The coldest cloud-top infrared brightness temperature on the 0537 UTC Suomi NPP VIIRS image was -70ºC (darker black enhancement); in addition, there appeared to be an Above-Anvil Cirrus Plume associated with that pyroCb, extending southeastward from a subtle Enhanced-V signature at the upshear (northwestern) edge of the cloud (where the warmest temperature was -48ºC, green enhancement).

On Himawari-8 Shortwave Infrared (3.9 µm) images (below), the pyroCb clouds exhibited a warmer (darker gray) appearance compared to adjacent conventional cumulonimbus clouds — this is due to the fact that ice crystals ejected into the pyroCb anvils are smaller (due to their shorter residence time within the intense updrafts above the fires), and these smaller ice crystals are more effective reflectors of incoming solar radiation. The large flare-up of red-enhanced land during the day is due to highly reflective soils of the Great Victoria Desert that quickly become very hot.

View only this post Read Less

GOES-17 (GOES-West) Low-level (7.3 µm), Mid-level (6.9 µm) and Upper-level (6.2 µm) Water Vapor images (above) displayed subtle thermal signatures of some of the highest-elevation western and central portions of the Alaska Range on 28 February 2019.

Plots of GOES-17 Water Vapor weighting functions, calculated using 12 UTC rawinsonde data from Anchorage, are shown below. Even with a very large satellite viewing angle (or zenith angle) of 70.1 degrees — which would tend to shift the Water Vapor weighting functions to higher altitudes —  the presence of dry air within the entire mid-upper troposphere brought the weighting function peaks downward to pressure levels corresponding to those of the higher elevations of the Alaska Range.

The dry air aloft helped to provide a remarkably cloud-free day over much of south-central Alaska, as seen in GOES-17 “Red” Visible (0.64 µm) images (below). In addition, an example of the transient spikes in daytime solar reflectance seen during this time of year was evident in the Visible imagery — note the brief brightening of a few of the images centered at 2115 UTC. Additional details about this effect are available here.

Just for fun, a closer look at the GOES-17 Visible imagery (below) revealed the tidal ebb and flow of drift ice within Cook Inlet and Turnagain Arm in the Anchorage area — and a similar diurnal flow of ice was also seen on the following day (animated GIF | MP4).

===== 01 March Update =====

With a similar scenario to the previous day (but farther to the east), dry air aloft allowed thermal signatures of the highest summits of the eastern Alaska Range and the Wrangell Mountains to be apparent in GOES-17 Water Vapor imagery — even the Upper-level 6.2 µm band (above).

However, in this case the Water Vapor weighting functions derived using rawinsonde data from nearby Anchorage were not representative of the pocket of dry air farther east over the mountains. An increase in the moisture profile — especially within the mid/upper troposphere — shifted the weighting functions to higher altitudes, due absorption and re-emission by that higher-altitude (and colder) moisture. Note how the weighting function contributions centered around the 300 hPa pressure level increased during the 24 hours from 12 UTC on 28 February to 12 UTC on 01 March, while the relative contributions decreased within the 500-700 layer (below).

An overpass of the Suomi NPP satellite provided a number of NUCAPS sounding profiles across this region (below).

A comparison of a dry NUCAPS sounding over the mountains (Point D) with a moist sounding near Anchorage (Point M) is shown below. As shown here, the moisture profile of the NUCAPS Point M sounding was similar to that of the 12 UTC Anchorage sounding.

View only this post Read Less