Cold air builds in Alaska

December 31st, 2008
GOES-11 10.7 µm IR images

GOES-11 10.7 µm IR images

Very cold air was becoming established across the interior of Alaska during the last few days of December 2008. AWIPS images of the 4-km resolution GOES-11 10.7 µm IR channel (above) showed large areas exhibiting very cold IR brightness temperatures (colder than -40º C, darker blue color enhancement) which were increasing in areal coverage as the surface temperatures continued to drop. The coldest air temperature measured in Alaska on 30 December 2008 was -49º C (-57º F) at O’Brien Creek; the temperature at the Fairbanks airport dropped to -41º C (41º F), with the coldest location in the Fairbanks urban corridor (the Woodsmoke subdivision at North Pole) reaching -43º C (-46º F).

A closer look using 1-km resolution 10.8 µm IR data from the AVHRR instrument on the NOAA series of polar-orbiting satellites (below) showed remarkable detail in the cold surface temperatures, whose patterns were strongly influenced by elevation. Due to cold air drainage, the coldest surface IR temperatures (indicated by the darkest blue color enhancement) were found in lower elevations such as river valleys and the Yukon Flats area of interior Alaska — note that the Yukon Flats region (located in the lower right quadrant of the images) exhibited a progressively darker blue enhancement on these 3 IR images as the surface temperatures plummeted during that time period. The coldest IR brightness temperature seen on the 15:57 UTC image was 222º K (-51º C, or -60º F) — and the air temperature at Fort Yukon was -46º C (-51º F) at that particular time.

NOAA-15 / NOAA-17 / NOAA-18 AVHRR 10.8 µm IR images

NOAA-15 / NOAA-17 / NOAA-18 AVHRR 10.8 µm IR images

With such cold temperatures becoming established up in Alaska, parts of the Lower 48 states should be on notice for a possible arctic outbreak in the coming weeks…

– 04 JANUARY 2009 UPDATE –

The coldest temperatures during this particular streak of cold temperatures occurred on 04 January 2009, when -54º C (-65º F) was reported at O’Brien Creek. A NOAA-17 AVHRR 10.8 µm IR image (below) indicated that IR brightness temperatures in the Yukon Flats region (located in the southeastern portion of the image) were as cold as -52º C (-62º F) at 20:51 UTC (11:51 AM local time).

NOAA-17 AVHRR 10.8 µm IR image

NOAA-17 AVHRR 10.8 µm IR image

– 10 January 2009 Update –

The coldest official temperatures recorded in Alaska during this 15-day-long cold spell were -56º C (-68º F) at O’Brien Creek (on 10 January) and at Chicken (on 08 January). In Fairbanks, the temperature remained below -40º C/F for a 24-hour period on 05-06 January.

Strong winds over the Upper Midwest region

December 29th, 2008
GOES-13 6.5 µm water vapor images

GOES-13 6.5 µm water vapor images

A potent shortwave trough was propagating eastward across the Upper Midwest region on 29 December 2008, and AWIPS images of the GOES-13 6.5 µm “water vapor channel” (above) showed a well-defined signature of mid-level dry air (yellow colors) moving rapidly east-southeastward over parts of Minnesota, Wisconsin, and Lower Michigan during the morning and afternoon hours. MODIS 6.7 µm water vapor channel images from consecutive overpasses of the Terra and Aqua satellites (below) showed a similar signature, albeit with a slightly larger areal coverage of the “dry air” pocket.

MODIS 6.7 µm water vapor images

MODIS 6.7 µm water vapor images

A comparison of the three GOES-13 Sounder water vapor channels (6.5 µm, 7.0 µm, and 7.4 µm) with the GOES-13 Imager 6.5 µm water vapor channel (below) revealed that a more distinct “dry air signature” (indicated by the brighter yellow colors) was evident when examining the Sounder 7.0 µm and 7.4 µm channel imagery — those water vapor channel’s weighting functions peak at lower altitudes than either the Sounder 6.5 µm or the Imager 6.5 µm water vapor channels.

GOES Sounder and Imager water vapor channel data

GOES-13 Sounder and Imager water vapor channel data

Note that the core of this dry air pocket appeared to have moved over the NOAA Wind Profiler site at Blue River in southwestern Wisconsin (station identifier BLRW3) after about 15:00 UTC — and the wind profiler data at that site (below) revealed a nice “descending jet” signature from the late morning into the early afternoon hours, with wind speeds of 100 knots and higher (red colors) deepening and moving downward from around the 10 km altitude at 13:00 UTC to near the 4-6 km altitude by 20:00 UTC. As the high momentum aloft (associated with the fast-moving mid-tropospheric dry air seen on the water vapor imagery) was gradually transferred downward to lower altitudes, the surface winds gusted to 51 mph at Austin in southern Minnesota (at 15:15 UTC),  47 mph at Sheboygan in eastern Wisconsin (at 18:24 UTC), and 65 mph at Charlevoix in northern Lower Michigan (at 00:25 UTC).

Blue River, Wisconsin wind profiler data

Blue River, Wisconsin wind profiler data

The core of the mid-tropospheric dry air had moved over southern Lower Michigan by late afternoon, and the GOES-13 Sounder and Imager water vapor channel weighting functions calculated using the 00:00 UTC rawinsonde data from Detroit (below) indicated that the layer of radiation being detected by the Sounder 7.4 µm channel was peaking at a very low altitude (red plot), with a significant component of that radiation likely coming from the surface.

Detroit, Michigan GOES-13 water vapor channel weighting functions

Detroit, Michigan GOES-13 water vapor channel weighting functions

It is interesting to note that an animation of the GOES-13 Sounder 7.4 µm water vapor channel imagery with the map overlay removed (below) displayed a clear signature of the outline of portions of Lake Superior and Lake Michigan after the driest air had moved over the region — the strong thermal signature of the “cold land / warmer water” surface boundary was able to reach the satellite, since there was very little middle and upper tropospheric water vapor present to attenuate the signal.

GOES-13 Sounder 7.4 µm water vapor images (with map overlay removed)

GOES-13 Sounder 7.4 µm water vapor images (with map overlay removed)

Thunderstorm causes a power blackout in Oahu, Hawaii?

December 27th, 2008
GOES-11 10.7 µm IR images

GOES-11 10.7 µm InfraRed (IR) images

A strong thunderstorm which produced heavy rain and frequent lightning was apparently a factor in triggering a power blackout that affected the Hawaiian island of Oahu for at least 12 hours, beginning on the evening of 26 December 2008 — in fact, this was the first time all of Oahu had lost power since October 2006, when a 6.7 magnitude earthquake shook the Hawaiian Islands and knocked out power on Oahu and parts of other islands for up to two days. GOES-11 10.7 µm IR images (above) showed the rapid development — and the subsequent rapid dissipation — of this thunderstorm over Oahu (the island located at the center of the images), which formed around 04:00 UTC on 27 December (6 PM local time on 26 December). The IR cloud top brightness temperatures quickly cooled to a minimum value of -45º C (violet color enhancement) at 04:30 UTC, but then cloud top temperatures increased to values warmer than -40º C as the storm moved northwestward away from the island of Oahu by 06:00 UTC.

An upper-level low had been located to the east of the Hawaiian Islands for several days, and GOES-11 6.7 µm water vapor images (below) showed that this low moved slowly westward during the 25-27 December period. Also note the roll-up of smaller cyclonic “vorticies” along the northern edge of the moisture field of the upper low — CIMSS water vapor winds products indicated that these vorticies were forming in an environment of low wind shear that existed over the northern periphery of the low (and the wind shear tendency product showed that shear had been decreasing over that region during the previous 24 hours).

GOES-11 6.7 µm water vapor images

GOES-11 6.7 µm water vapor images

An AWIPS image of the CIMSS MIMIC Total Precipitable Water product (below) indicated that TPW values were certainly not as high over the Hawaiian Islands as those seen farther to the south in the region of the Inter-Tropical Convergence Zone (ITCZ), but PW values in excess of 40 mm (green to yellow color enhancement) were in place as the upper low moved westward across the region. Over the Big Island of Hawaii, several locations had received more than 10 inches of rainfall in 24 hours, with Waiakea Uka reporting 13.15 inches (and about a foot of snow fell at the higher elevations of Mauna Kea and Mauna Loa).

MIMIC Total Precipitable Water (TPW)

MIMIC Total Precipitable Water (TPW)

In advance of the arrival of the higher TPW values associated with the upper-level low, a pocket of dry air had been moving southwestward across the Hawaiian Islands on 25 December. It is interesting to note that GOES-11 water vapor images (below, with the map overlay removed) revealed the presence of “standing waves” due to the strong northeasterly winds interacting with the high terrain of the island chain.

GOES-11 6.7 µm water vapor images (with map overlay removed)

GOES-11 6.7 µm water vapor images (with map overlay removed)

The GOES-11 water vapor channel weighting functions (below) showed that the mid-tropospheric dry air present over Hilo, Hawaii at 12:00 UTC on 25 December had the effect of shifting the peak of the water vapor weighting function downward, allowing features at a lower altitude to be resolved on the water vapor imagery (compared to what would be viewed at that location if a slightly cooler but more moist “US Standard Atmosphere” were in place).

GOES-11 water vapor channel weighting functions

GOES-11 water vapor channel weighting functions

Ice forming in Lake Michigan

December 22nd, 2008
MODIS true color image

MODIS true color image

Unusually cold temperatures were experienced across much of the Upper Midwest and Great Lakes region during  the month of December 2008 — as of 21 December, monthly temperatures were 6.4º F below normal at Milwaukee WI (their low that morning was -5º F), 7.2º F below normal at Chicago IL (their low that morning was -6º F), and 8.6º F below normal at Madison WI (their low that morning was -10º F). On the following day (22 December 2008), 250-meter resolution MODIS Red/Green/Blue (RGB) true color imagery from the SSEC MODIS Today site (above) revealed that significant amounts of ice had begun forming in the western and southern near-shore waters of Lake Michigan. During the previous winter season, similar ice formation was not seen until 19-20 January.

The satellite image confirms that there was abundant snow cover over the area (snow depths on the morning of 22 December were as great as 18 inches in southeastern WI, 12 inches in northeastern IL, and 30 inches in southwestern Lower Michigan). Also note that some of the smaller inland lakes in southeastern WI remained unfrozen — those particular lakes are quite deep, and take longer to freeze.

Another item of interest was the fact that there was a “convergence” of the Lake Michigan cloud bands seen over southwestern Lower Michigan: winds (and the resulting cloud bands) were oriented southwest-to-northeast over the southern portion of Lake Michigan, and oriented northwest-to-southeast farther to the north. Lake-effect snowfall amounts in that part of southwestern Lower Michigan included 13.4 inches at Muskegon and 12.6 inches at Grandville.

Farther to the north, ice was also seen forming in the far southwestern portions of Lake Superior — and the long tornado damage path from the 07 June 2007 EF3 tornado in northeastern WI was still quite visible in the MODIS true color imagery.