POES AVHRR fog/stratus product + METAR surface reports

An AWIPS image of the 1-km resolution POES AVHRR 11.0-3.7 µm “fog/stratus product” (above) revealed a very interesting banded structure in the fog and stratus features (yellow to red color enhancement) that covered a large part of the western Great Lakes region at 07:49 UTC on 16 January 2010. With stagnant high pressure in place and light winds over much of the area, there was little boundary layer mixing to mitigate the formation and persistence of the low clouds that were trapped under a very strong low-level temperature inversion (Rawinsonde data: Davenport IA | Lincoln IL | Wilmington OH; NAM12 Line A-A’ west-to-east cross section).

An animation of the 4-km resolution GOES-12 11.0-3.9 µm fog/stratus product (below) suggested that there might be a thin veil of high-altitude cirrus clouds (which show up as darker black features) drifting over the banded portion of the fog and stratus — the presence of overlying high clouds could have been contributing to the unusual appearance of those low cloud features on the fog/stratus product imagery.

GOES-12 fog/stratus product + surface pressure/fronts

To further explore whether or not there were indeed cirrus clouds overhead that might be masking the low-altitude fog/stratus signal and creating the false appearance of banding as seen on the fog/stratus product, let’s examine a few additional satellite images. The POES AVHRR 10.8 µm IR image (below) did show some thin filaments that exhibited slightly colder IR brightness temperatures, but most were warmer than -20º C — much warmer than would be expected of typical cirrus clouds. However, with very thin cirrus features, a significant amount of thermal energy reaches the satellite IR detectors from the warmer surfaces below; this then leads to IR brightness temperatures values that are much warmer than those of the actual cirrus clouds themselves.

POES AVHRR 10.8 µm IR image

Multi-spectral POES AVHRR derived products such as Cloud Type, Cloud Top Temperature, and Cloud Top Height (below) did a better job at identifying more of the thin features as Cirrus (orange color enhancement on the Cloud Type product) — with cloud top temperatures of -30º C to -40º C and cloud top heights of 7-8 km — but some of the thinnest cirrus filaments were still flagged as supercooled water droplet clouds (cyan color enhancement on the Cloud Type product).

POES AVHRR Cloud Type, Cloud Top Temperature, and Cloud Top Height products

In this case, perhaps the best satellite product to confirm the presence of thin cirrus filaments aloft was the 1-km resolution MODIS 6.7 µm water vapor channel (below). Since the weighting function of such a water vapor channel normally peaks at higher altitudes in the middle troposphere, it is generally immune to the effects of warm surface radiation that can sometimes plague proper cirrus cloud classification using conventional IR imagery alone.

MODIS 6.7 µm water vapor image

As it turns out, a broad region of high-altitude “transverse banding” had formed from the southern Great Lakes to the mid-Atlantic region — this banding was generally located along the axis of a strong (120 to 130 knot) anticyclonically-curved jet stream axis, as confirmed by an overlay of CRAS model Level of Maximum Wind isotachs (below).

MODIS 6.7 µm water vapor image + CRAS45 Level of Maximum Wind isotachs

This transverse banding is a satellite signature that indicates a potential for high-altitude turbulence — and there were a few pilot reports of moderate turbulence over the area during the 02-11 UTC time period, shown overlaid on 4-km resolution GOES-12 6.5 µm water vapor images (below).

GOES-12 6.5 µm water vapor images + pilot reports of turbulence

View only this post Read Less

GOES-12 and GOES-14 visible images

McIDAS images of GOES-12 and GOES-14 visible channel data (above) showed that large chunks of ice (known as “ice fields”) were drifting north-northeastward across the western portion of Lake Erie on 14 January 2010. Southerly to southwesterly winds were beginning to increase on that day, helping to move the ice features across the surface of the lake.

AWIPS images of MODIS false-color Red/Green/Blue (RGB) composites (below) confirmed that these were indeed ice features — snow and ice appear darker red on this false color imagery, in contrast to supercooled water droplet clouds which appear as brighter shades of cyan to white. Surface METAR reports plotted on the imagery indicated that wind speeds were generally in the 10-20 knot range. This example offers a glimpse at the type of RGB image capability that should be available with the upcoming AWIPS II software.

MODIS false color RGB images

A closer look using 250-meter resolution MODIS true color images from the SSEC MODIS Today site (below) revealed how far the ice fields drifted between the 16:58 UTC overpass of the Terra satellite and the 18:42 UTC overpass of the Aqua satellite.

MODIS true color images

View only this post Read Less

MODIS Sea Surface Temperature + RTG_SST_HR surface temperature analysis

AWIPS images of the MODIS Sea Surface Temperature (SST) product and the High-Resolution RTG_SST model surface temperature analysis (above) revealed some important differences between satellite-measured and model-derived SST features on 11 January 2010. In particular, note the very strong MODIS SST gradient across the Florida Keys, with water temperatures cooling from the low 70s F (yellow color enhancement) to the low 60s F (darker green color enhancement) over a very short distance. Farther to the north, MODIS SST values off the coast of Louisiana were only in the upper 30s to low 40s F (darker blue color enhancement), in contrast to the RTG_SST analysis values in the low 50s F.

In addition, MODIS true color and 11.0 µm IR images from the SSEC MODIS Direct Broadcast site (below) indicated that the colder water temperatures coincided with areas of increased water turbidity (resulting from mixing by strong winds over those more shallow waters).

MODIS true color image + MODIS 11.0 µm IR image

MODIS 11.0 µm IR images from 02 January 2010 and 11 January 2010 (below) gave some indication about how much the waters off the west coast of Florida cooled during that 9-day period of record-setting cold temperatures. At Tampa in western Florida, temperatures during the 02 January to 10 January period were 10 to 30 degrees F below normal, and the length of the cold spell set a new record (with Tampa never reaching 60º F during that entire time). In addition, on the morning of 11 January, with northerly winds gusting to 24 mph:

RECORD EVENT REPORT
NATIONAL WEATHER SERVICE KEY WEST FL
740 AM EST MON JAN 11 2010

…RECORD LOW TEMPERATURE SET IN KEY WEST…

THE TEMPERATURE AT KEY WEST AIRPORT DROPPED TO 42 DEGREES AT 455 AM ON MONDAY JANUARY 11TH. THIS SET A NEW RECORD LOW TEMPERATURE FOR THE DATE. THE PREVIOUS RECORD WAS 48 DEGREES SET IN 1970.

THIS WAS ALSO THE SECOND COLDEST TEMPERATURE EVER RECORDED IN KEY WEST…WITH WEATHER RECORDS DATING BACK TO 1873. THE ALL TIME RECORD LOW TEMPERATURE IN KEY WEST IS 41 DEGREES…WHICH HAS OCCURRED TWICE…ON JANUARY 13TH 1981…AND JANUARY 12TH 1886.

MODIS 11.0 µm IR images (02 January 2010, 11 January 2010)

View only this post Read Less

GOES-11, GOES-14, and GOES-12 10.7 µm IR and 6.7/6.5 µm water vapor images

The Galeras volcano in Colombia (located in the Andes Mountains near Colombia’s border with Ecuador) experienced an explosive eruption around 00:43 UTC on 03 January 2010 (Washington VAAC advisory messages). McIDAS images of the GOES-11, GOES-14, and GOES-12 10.7 µm IR (top 3 panels) and 6.7/6.5 µm water vapor channel data (bottom 3 panels) showed the volcanic cloud as it spread outward and drifted to the west for several hours following the eruption.

GOES-11 (GOES-West, positioned at 135º West longitude) only imaged the region once every 3 hours during a full disk scan (in this example, at 00:00, 03:00, and 06:00 UTC). GOES-14 (positioned at 105º West longitude) was emulating GOES-West operations during the final days of its NOAA Science Test, and except for the 3-hourly full disk scans, its imaging area was terminated just to the east of the volcano. GOES-12 (GOES-East, positioned at 75º West longitude) had the best, most consistent view of the South American region.

Note how the signature of the volcanic cloud becomes more difficult to follow on the IR imagery, but is still faintly recognizable for an additional 1-2 hours on the water vapor imagery. The different appearance of the GOES-11 water vapor imagery is due to the fact that the 6.7 µm water vapor channel on GOES-11 is a much “narrower” channel (spectrally) compared to the 6.5 µm water vapor channel on GOES-12 and GOES-14; in addition, the large viewing angle from GOES-11 has shifted the water vapor weighting function to higher (colder) altitudes, making the features appear darker blue to white with this particular water vapor color enhancement.

GOES-11 IR difference product (10.7 µm - 12.0 µm)

A simple method for identifying and tracking volcanic ash clouds is the use of a “reverse absorption” or “split window” IR difference product, subtracting the brightness temperature values of the 12.0 µm IR channel from the 10.7 µm IR channel. The GOES-11 IR difference product at 03:00 and 06:00 UTC (above) showed that the volcanic ash cloud had drifted westward over Ipiales (station identifier SKIP) and Pasto (station identifier SKPS) in Colombia, and was moving just to the north of Quito, Ecuador (station identifier SEQU). Note that on the more recent GOES satellites — GOES-12 and beyond — the 12.0 µm IR channel on the Imager instrument has been replaced with a 13.3 µm IR channel, preventing the application of this type of IR difference volcanic ash identification on the more recently-launched GOES satellites.

A corresponding IR difference product using the MODIS 11.0 µm and 12.0 µm channels at 03:25 and 06:25 UTC is shown below. With finer spatial resolution than GOES (1 km, vs 4 km), the edges of the ash cloud feature appear with greater clarity on the MODIS images.

MODIS IR difference product (11.0 µm - 12.0 µm)

View only this post Read Less