Extremely cold cloud top temperatures

November 22nd, 2008
MTSAT-1R IR imagesMTSAT-1R 10.8 µm IR images

Hat-tip to blog reader Boris Konon for sending us this in an email:

Noted a large area of extremely cold cloud tops on a thunderstorm complex over northern Australia 17-23z 11/22.  So cold it went off our IR color table scale we use here at WSI for MTSAT!

MTSAT-1R 10.8 µm IR images (above) did indeed show an area of convection that exhibited an unusually large canopy of cloud top temperatures colder than -80º C (light purple color enhancement) on 22 November 2008. The coldest IR brightness temperature seen with this particular thunderstorm complex was 174.78º K (-98.37º C), just off the northern coast of Australia at 20:30 UTC.

A comparison of MTSAT-1R visible and IR images at 20:30 UTC (below) showed that there were a number of overshooting tops within the area of cold cloud top temperatures, as the convective system was beginning to move offshore across the Gulf of Carpenteria. This squall line was responsible for producing 25-50 mm of rainfall (BOM rainfall analysis) and wind gusts to 96 km/h (52 knots) at Bradshaw and 68 km/h (37 knots) at Gove (thanks to Jeff Callaghan, Australian Bureau of Meteorlogy for that information!).

MTSAT-1R visible and IR images

MTSAT-1R visible and IR images

A closer view using 1-km resolution NOAA-15 10.8 µm IR data (below) revealed a bit more detail in the cloud top temperature structure around 19:56 UTC. Unfortunately, the image was marred by a large number of “noise” pixels — but the coldest reliable cloud top temperature seen on this AVHRR IR image was 170.2 K (-102.95º C), located within the farthest southeast cold pixel cluster (IR brightness temperatures colder than -100º C are enhanced in the yellow to brown colors at the end of the scale). If this cloud top temperature is accurate, it rivals what is believed to be the coldest cloud top temperature value noted in the literature: -102º C in Ebert and Holland, 1992.

NOAA-15 AVHRR 10.8 µm IR image

NOAA-15 AVHRR 10.8 µm IR image

Rawinsonde data from Darwin Airport at 00:00 UTC on 23 November 2008 (below) indicates that the tropopause was at a very high altitude (around 100 hPa, approximately 16.0 km). Such a high and cold tropopause is not unusal  in the tropics — especially in that particular region, as studies have shown that the average tropopause level is quite high and quite cold over the western tropical Pacific Ocean.

Darwin, Australia rawinsonde report

Darwin, Australia rawinsonde report

Standing wave clouds over northeastern Minnesota

November 21st, 2008
GOES-12 and GOES-13 visible images

GOES-12 and GOES-13 visible images

GOES-12 and GOES-13 visible images (above) showed the development of a narrow band of terrain-forced “standing wave clouds” over extreme northeastern Minnesota on 21 November 2008. Surface wind barbs (plotted in cyan) indicated that the surface winds were generally from the northwest at speeds of 10 knots or less across the region; however, the cloud motions suggested that the northwesterly winds at higher altitudes were a bit stronger. This northwesterly wind direction was perpendicular to the higher terrain of the “North Shore Ridge” — where elevations rise to 2000-2100 feet — which runs from southwest to northeast across the Arrowhead Region of northeastern Minnesota (topography image courtesy of Rick Kohrs, SSEC).

The fact that a thin shadow was cast on the surface along the northern edge of the cloud band indicated that this cloud feature was either fairly deep, or was located at a fairly high altitude. Note that a cirrus plume can be seen that was apparently being sheared off the northern portion of the main standing wave cloud band, which was then carried south-southeastward across Lake Superior by the stronger winds aloft. AWIPS images of the MODIS visible, 11.0 µm IR window, Cloud Top Temperature, and Cloud Phase at 16:22 UTC (below) indicated that a significant portion of the aforementioned cirrus plume was colder than -30º C, with the Cloud Phase product indicating that Ice cloud was present (pink color enhancement).

MODIS visible, IR window, Cloud Top Temperature, and Cloud Phase

MODIS visible, IR window, Cloud Top Temperature, and Cloud Phase

Vertical cross sections of RUC13 model fields (below, courtesy of Dan Miller, Science and Operations Officer at the Duluth MN National Weather Service forecast office) did a fairly realistic job of depicting a deep pocket of upward vertical velocity (Omega, purple contours) within the 800-300 hPa layer that was providing the forcing for the standing wave cloud band — and as moist layers (Relative Humidity greater than 50%, green shading) passed through the deep pocket of Omega, a standing wave cloud band formed that could then seen on satellite imagery. The higher-altitude moist layer arriving at the later time periods seems to correspond to the layer that produced the cirrus plume — and this higher layer was at temperatures colder than -30º C, in agreement with the temperatures seen on the MODIS IR image and Cloud Top Temperature product.

Vertical cross section of RUC13 model fields

Vertical cross sections of RUC13 model fields

A closer view using 250-meter resolution MODIS true color and false color imagery from the SSEC MODIS Today site (below) actually depicted two separate standing wave cloud bands, with the high-altitude cirrus streaming off the upper portion of the bands showing up quite nicely. One could also see that many of the smaller lakes across northeastern Minnesota were either completely frozen or were in the process of becoming ice-covered.

MODIS true color and false color images

MODIS true color and false color images

Freezing drizzle in Colorado

November 20th, 2008
MODIS Fog/Stratus product + topography

MODIS Fog/Stratus product + topography

A cold and moist upslope (northeasterly) flow had pushed up against the Front Range of the Rocky Mountains in Colorado during the pre-dawn hours on 20 November 2008 — a comparison of AWIPS images of the MODIS Fog/Stratus product and the topography (above) showed that the western edge of the stratus deck was backed up against the highest terrain that runs north-south across central Colorado.

AWIPS images of the MODIS Fog/Stratus product, Cloud Top Temperature (CTT) product, and Cloud Phase product (below) indicated that  CTT values were generally in the -8 to -15º C range, with the Cloud Phase product indicating that the cloud was composed of supercooled water droplets. Without the presence of ice crystals in the clouds, the resulting precipitation type across parts of eastern Colorado was freezing drizzle at locations whose surface air temperature had dropped below freezing, as was seen at Denver/Centennial Airport (KAPA) and Elbert Mountain (KMNH) at 10:00 UTC (4 AM local time). Winter Weather Advisories were issued for a number of counties, due to the freezing drizzle causing roadway icing.

MODIS Fog/Stratus product. Cloud Top Temperature product, Cloud Phase product

MODIS Fog/Stratus product, Cloud Top Temperature product, and Cloud Phase product

This extensive stratus cloud deck could be further characterized by examining the GOES-12 Fog/Stratus product in conjunction with the Low Cloud Base product and the sounder-derived Cloud Top Height product (below) — the cloud bases were all below 1000 feet (green color enhancement), while the cloud tops were generally in the 12,000-15,000 foot range (light green to dark green color enhancement). Using the AWIPS “Sample Cloud Heights/Radar Skew-T” functionality, the cloud top height in the Elbert Mountain area (where the MODIS IR brightness temperature was around +9ºC) was placed at around 11,000 feet, in close ‘agreement with the GOES Sounder Cloud Top Height product.

GOES-12 Fog/Stratus product, Low Cloud Base product, Cloud Top Height product

GOES-12 Fog/Stratus product, Low Cloud Base product, and Cloud Top Height product

Finally, it is interesting to point out the improved accuracy of the 1-km resolution MODIS Fog/Stratus product compared to the 4-km resolution GOES-12 Fog/Stratus product (below). In particular, note how the MODIS imagery displayed a “hole” in the stratus deck in the area of Pike’s Peak (located just to the west of Colorado Springs, KCOS), directly  where the higher terrain would have extended above the top of the stratus cloud layer — on the GOES-12 Fog/Stratus product, the “hole” in the stratus deck was incorrectly placed a bit farther to the west.

MODIS and GOES-12 Fog/Stratus product + topography

MODIS and GOES-12 Fog/Stratus product + topography

Interesting “O-shaped clouds” over the Eastern Pacific Ocean

November 18th, 2008
GOES-13 visible images

GOES-13 visible images

We created the “What the heck is this?” category just for the type of case that is shown here: GOES-13 visible images (above) displayed an interesting cluster of O-shaped clouds forming over the Eastern Pacific Ocean (near Isla Guadalupe, off the coast of Baja California) on 18 November 2008. A few hours later, an overpass of the QuikSCAT satellite allowed an overlay of WindSat surface wind data on a GOES-11 3.9 µm shortwave IR image  (below) — and the QuikSCAT wind data seemed to suggest that these O-shaped clouds were actually perturbing the general northwesterly marine boundary layer flow to some extent.

GOES-11 3.9 µm IR image + QuikSCAT winds

GOES-11 3.9 µm shortwave IR image + QuikSCAT winds

MODIS visible, 3.7 µm IR, and 11.0 µm IR images

MODIS visible, 3.7 µm IR, and 11.0 µm IR images

AWIPS images of the MODIS visible, 3.7 µm shortwave IR, and 11.0 µm IR window channels (above, with an overlay of lower-tropospheric MADIS atmospheric motion vectors) provided two important clues about these cloud features: (1) they were composed of supercooled water droplets, which reflected large amounts of solar radiation leading to a display of very warm (> 30º C, darker gray shades) 3.7 µm brightness temperatures, and (2) they were shallow clouds within the marine boundary layer, with fairly warm cloud top IR window brightness temperatures in the 13-14º C range. These points were further confirmed by examining additional MODIS images (below): the MODIS Cloud Top Temperature (CTT) product showed CTT values of 16-17º C (red color enhancement); the MODIS Cloud Phase product indicated these clouds were composed of supercooled water droplets (blue color enhancement); and the GOES-11 Sounder Cloud Top Height product placed the cloud tops in the 3000-5000 foot range (tan to orange color enhancement).

MODIS visible, Cloud Top Temperature, Cloud Phase + GOES Cloud Top Height

MODIS visible, Cloud Top Temperature, Cloud Phase + GOES Cloud Top Height

A closer view using 250-meter resolution MODIS true color image from the SSEC MODIS Today site (below) showed impressive structure to the O-shaped clouds, with hints of fine-scale outflow boundaries along the outer edges of some of the cloud features. These cloud features somewhat resemble  Pockets of Open Cells that have been previously documented — these open cells are apparently related to the formation of areas of precipitation (in this case, drizzle) that then act to dissipate a portion of the cloud to the point that a hole forms in the cloud feature. The downdrafts created by the formation of these pockets of open cells may indeed have had enough of an impact on the surface winds to be apparent in the QuikSCAT surface wind data seen above.

250-m resolution MODIS true color image

250-m resolution MODIS true color image

As an aside, the MODIS Sea Surface Temperature (SST) product (below) showed that SST values were generally in the mid 60s F (darker green colors) over the area where the O-shaped clouds were forming — and there was a well-defined SST gradient just to the south, where SST values rose into the lower 70s F (lighter green to yellow colors).

MODIS Sea Surface Temperature product

MODIS Sea Surface Temperature product