A highly-amplified upper air and jet stream pattern allowed a lobe of the polar vortex to migrate southward across southern Canada and the north-central US — leading to an outbreak of arctic air throughout the Upper Midwest and Great Lakes during the 29 January – 30 January 2019 period. The path and expansion of the cold arctic air was apparent in GOES-16 (GOES-East) Air Mass RGB images from the AOS site (above) — which first became evident over the Canadian arctic beginning on 28 January. The coldest air exhibited pale shades of yellow to beige in the Air Mass RGB images.

“It Came from Nunavut”
It sounds like the title of a good movie to describe this cold weather. Have you been wondering where this system came from? The upper-level low pressure system dived down from northwest Nunavut, in far northern Canada, to Minnesota. pic.twitter.com/GZSMCcO1eR

— NWS Duluth (@NWSduluth) January 30, 2019

GOES-16 “Clean” Infrared Window (10.3 µm) images (below) also showed the southward expansion of arctic air into the north-central US — surface infrared brightness temperatures of -30 to -40ºC (darker blue to green enhancement) covered a large area. Such cold infrared brightness temperatures are normally associated with clouds in the middle to upper troposphere. Surface air temperatures of -20 to -40ºF were widespread, along with wind chill values of -40 to -70ºF, leading to numerous school and business closures. Two of the coldest official temperatures in the US on the morning of 30 January were -48ºF at Norris Camp, Minnesota and -44ºF at Bottineau, North Dakota (the high temperature in Bottineau on the previous day, 29 January, was only -26ºF); however, there were a few North Dakota Department of Transportation roadside sensors that reported low temperatures of -49ºF.

More than 130 hourly observation stations across the Upper Midwest recorded a wind chill of -50°F or colder overnight into this morning. Here are some select station minimums from across the region. pic.twitter.com/A7FXAKdP8y

— MRCC (@MidwestClimate) January 30, 2019

GOES-16 True Color RGB images (below) revealed a variety of multiple-band and single-band lake effect snow features as the arctic air moved across the Great Lakes. In addition, elongated and long-lived cloud bands created snow squall conditions across parts of Ohio and Pennsylvania.

VIIRS True Color RGB and Infrared Window (11.45 µm) images from NOAA-20 (at 1802 UTC) and Suomi NPP (at 1852 UTC) viewed using RealEarth (below) provided a closer look at the cloud bands across Ohio and Pennsylvania.

=======================================================

In addition to being extremely cold, the airmass over the Upper Midwest was extremely dry. The image above shows the Baseline GOES-R Total Precipitable Water product. The default AWIPS color enhancement has been modified to better capture the extreme dryness. Regions in light blue over western Minnesota and the eastern Dakotas curving through Iowa into northern Illinois show TPW values around 0.01″ ; green shadings over Wisconsin and eastern Minnesota correspond to values closer to 0.03″. In such dry airmasses, it is possible to see surface features in the infrared 7.3 low-level ABI Water Vapor Channel, Band 10, and the morning of 30 January was no exception, below. Surface features like rivers are notable in Illinois, for example. Even the heat island of the Minneapolis/St. Paul is apparent (albeit barely).  Surface features over northern Minnesota and Wisconsin aren’t quite so apparent, perhaps because of the increased amounts of moisture there.  There is likely less surface temperature contrast there, also, as rivers/lakes are more likely frozen.  It is the temperature contrast — as best exemplified by the Great Lakes shorelines — that allows features to appear in the Water Vapor imagery.

Weighting Functions (in real time, from this site) allow for an estimate of where information in different water vapor channels will be detected by the satellite. In the 0000 UTC 30 January 2019 example, below, from Chanhassen, MN (when total precipitable water there was 0.01″), a large signal is apparent from the low-level water vapor channel (7.3 µm); in fact, most of the information detected by the satellite was coming from the surface!  Even the mid-level water vapor (6.9 µm) had a component from the surface.  Weighting Functions for Davenport Iowa (The axis of the driest air shifted from near Chanhassen at 0000 UTC to near Davenport at 1200 UTC) at 0000 UTC and 1200 UTC are shown here; Note in the toggle that the level from which information is received by the satellite drops from 0000 to 1200 UTC as dry air moves in.

The GOES-16 Baseline Land Surface Temperature product, below, from 1200 UTC, shows many values at/below -45 F (purple shading) over Minnesota.  Dark blue values are around -25 F.  Note the relatively warm region over western Iowa, in cyan.  That part of Iowa lacks snowcover and exceptional cold rarely happens over bare ground.

===== 31 January Update =====

Across much of the Upper Midwest, the coldest temperatures occurred on the morning of 31 January. GOES-16 Infrared images (above) showed much of northeastern Minnesota and far northwestern Wisconsin — the low temperature of -56ºF at Cotton was only 4 degrees warmer than the all-time record low for Minnesota, and the low temperature of -47ºF at Butternut was 8 degrees warmer than the all-time record  low for Wisconsin (both of those state records were set in early February 1996). The -56ºF in Cotton was not only the coldest temperature in the Lower 48 states on 31 January, but was also significantly colder than any official reporting station in Alaska that day. Also of interest in northeastern Minnesota, note the warmer plumes (darker blue enhancement; brighter greens are coldest) coming from power plants and industrial sites in the Iron Range area.

Farther to the south, GOES-16 Infrared images covering the Minnesota/Wisconsin/Iowa/Illinois region (below) also showed widespread cold surface brightness temperatures (shades of green). All-time record low temperatures were set at Cedar Rapids in Iowa (-30ºF) and at Moline (-33ºF) and Rockford (-31ºF) in Illinois. The cooperative observer at Mt. Carroll in northwestern Illinois reported a low of -38ºF — this was certified as a new all-time record minimum temperature for the state of Illinois.

The recent stretch of days with cold air in place had helped the ice coverage to increase significantly in western Lake Superior — and the transition from northerly/northwesterly cold air advection to southwesterly warm air advection at the surface began to fracture a lot of this newly-formed lake ice (below).

Ice coverage had also increased across much of western/central Lake Erie, although areas of open water continued to supply latent heat to help generate lake effect snow bands (below).

A sequence of Terra MODIS True Color RGB images (below) showed substantial growth of nearshore ice in the southern end of Lake Michigan from 29 to 31 January.

.A summary of this cold outbreak was compiled by NWS Duluth, NWS La Crosse, NWS Twin Cities and NWS Grand Forks.

View only this post Read Less

* GOES-17 images shown here are preliminary and non-operational *

GOES-17 Air Mass RGB images from the AOS site (above) showed the characteristic signature of a Potential Vorticity (PV) anomaly over the East Pacific Ocean on 29 January 2019. A descending tropopause brought ozone-rich air from the stratosphere down to very low altitudes — and this ozone-rich air was highlighted by deeper shades of red (the Air Mass RGB uses the 9.6 µm Ozone band to calculate the Green component). A Gale Force low became better organized beneath this PV anomaly, quickly reaching it’s occluded stage (surface analyses).

The deepening midlatitude cyclone also exhibited a vivid signature on GOES-17 Mid-level Water Vapor (6.9 µm) imagery (below). An overlay of the GFS75 model PV1.5 pressure — a parameter commonly used to diagnose the level of the “dynamic tropopause” — showed values as low as 800 hPa at 18 UTC.

View only this post Read Less

On 25 January 2019 a GOES-16 (GOES-East) Mesoscale Domain Sector was positioned over Argentina in support of the RELAMPAGO-CACTI field experiment, providing imagery at 1-minute intervals. “Red” Visible (0.64 µm) images (above) showed thunderstorms that developed and moved northward over the Córdoba (SACO) area — surface observations there (plot | list) showed a sharp drop in temperatures along with wind gusts to 37 knots during the thunderstorm, which also produced hail and heavy rainfall. Two important features were revealed in the imagery: (1) an outflow boundary (from the decay of a large and long-lived Mesoscale Convective System to the southeast) which was moving slowly northward between between Rio Cuarto (SAOC) and Córdoba, likely helping to enhance boundary layer convergence and lift, and (2) a southward/southwestward flow of moist, unstable air — indicated by a plume of agitated cumulus clouds — approaching Córdoba. Toward the end of the day, the presence of an Above-Anvil Cirrus Plume also became evident in the Visible imagery.

The corresponding GOES-16 “Clean” Infrared Window (10.3 µm) images (below) showed that infrared brightness temperatures of the pulsing thunderstorm overshooting tops were frequently -90ºC or colder (yellow pixels embedded within darker purple). This indicates a significant overshoot of the tropopause, which had an air temperature of -72.1ºC at an altitude of 15.2 km on 12 UTC rawinsonde data. Also note the development of a pronounced cold/warm thermal couplet over the core region of the storm, as an enhanced-V storm top signature formed.

A side-by-side comparison of GOES-16 Visible and Infrared images is displayed below.

A toggle between NOAA-20 VIIRS True Color Red-Green-Blue (RGB) and Infrared Window (11.45 µm) images at 1734 UTC as viewed using RealEarth (below) showed the early stage of convective development south of Córdoba, as well as the large decaying MCS to the southeast. Cloud-top infrared brightness temperatures with the developing storms were already -80–C and colder (violet enhancement), about 10ºC colder than what was observed using lower-resolution GOES-16 imagery at that same time.

We observed some of the deepest storms I’ve ever seen on ground-based radar near Córdoba, Argentina today – comparable to some of the strongest ever observed with TRMM @NSF_GEO @RELAMPAGO2018 pic.twitter.com/d0Te1BEPSl

— Steve Nesbitt (@70_dbz) January 25, 2019

URGENTE | Fuerte tormenta en Córdoba Capital

Ahora mismo se registra una fuerte tormenta con ráfagas de viento y fuertes lluvias en la ciudad de Córdoba. [Más información en: https://t.co/0573RLZtPS ] pic.twitter.com/2pwP5DdSrC

— TormentasDelLitoral (@TDLtiempo) January 25, 2019

View only this post Read Less

JMA Himawari-8 “Red” Visible (0.64 µm), Shortwave Infrared (3.7 µm) and Infrared Window (10.3 µm) images (above) showed the development of a pyroCumulonimbus (pyroCb) cloud from a bushfire that was burning in the eucalypt forests of eastern Victoria, Australia on 25 January 2019. A rapid-scan “Target” sector was positioned over the region beginning at 0522 UTC, providing images every 2.5 minutes (instead of the routine 10-minute interval). Cloud-top infrared brightness temperatures became colder than -40ºC (the threshold for pyroCb classification) after 0230 UTC, and eventually cooled to around -55ºC (orange enhancement). This temperature roughly corresponded to an altitude around 12 km, according to nearby Melbourne rawinsonde data (plot | text).

A closer view of Himawari-8 “Red” Visible (0.64 µm) and Shortwave Infrared (3.7 µm) images (below) revealed the rapid southeastward run of the fire, as shown by the growth of the “hot spot” (black to red pixels) on Shortwave Infrared images. The darker gray appearance of the pyroCb cloud is due to the presence of smaller ice crystals at the cloud top — these smaller ice crystals are more efficient reflectors of incoming solar radiation, making the cloud tops appear warmer than those of conventional cumulonimbus. Vigorous updrafts driven by the intense heat of the fire limit the in-cloud residence time for ice crystal growth, which leads to smaller particles being ejected at the pyroCb cloud top.

In a comparison of VIIRS True Color Red-Green-Blue (RGB) and Infrared Window (11.45 µm) images from Suomi NPP (at 0311 UTC) and NOAA-20 (at 0501 UTC) images viewed using RealEarth (below), cloud-top infrared brightness temperatures were in the -55 to -58ºC range (darker shades of orange).

Large bushfire in #EastGippsland producing a huge plume of smoke this afternoon stretching over eastern Bass Strait. Check out the satellite picture at https://t.co/9md40P2b4k pic.twitter.com/5ceZV5nf0T

— Bureau of Meteorology, Victoria (@BOM_Vic) January 25, 2019

View only this post Read Less