After an unusually quiet February tornado-wise in the United States (only one tornado was reported, an EF-0 in California that lasted for 3 minutes late in the day on February 27th), two tornadoes touched down near sunset in western Oklahoma, near the town of Hammon. The above animation from NAWIPS shows 1-km GOES-12 visible imagery with superimposed contours of cloud-top cooling (and lightning strikes) from about 2000 UTC on 9 March to 0015 UTC on the 10th, shortly after tornado touchdown (the tornado location is indicated by the red triangle in the animation). The growing cell that will ultimately spawn the tornado is correctly identified. Such cloud-top cooling is computed routinely as part of a University of Wisconsin Cooperative Institute for Meteorological Satellite Studies (CIMSS) project on convective initiation. The algorithm locks on to the convective cell that spawns the tornado, and not adjacent cells. The strongest cooling was associated with the storm that produced the tornado.

If the detected cloud-top cooling is co-located with changes in cloud phase (for example, from supercooled water droplets to ice crystals, or from all water droplets to supercooled water droplets) as inferred from GOES-12 radiances for different infrared channels, then convective initiation is deemed to be ongoing. In this animation of 11-micron cloud temperatures overlain with Convective Initiation, note the CI Likely at 2132 UTC along the Texas/Oklahoma border. Note that only the nascent cell that subsequently produces a tornado is identified, as only that cell contains the required strong cooling and the phase changes. That region of cloud top cooling and phase change then moves north-northeastward as the convective cell that spawned the tornado develops. Convective Initiation is no longer indicated, however, because the initiation stage of convection is over by 2200 UTC.

(Added: CNN has Andy Gabrielson’s too-close-for-comfort video).

(Added: Another video, from YouTube).

(One more Addition: Write up on the tornadic event from the Norman, OK office of the National Weather Service).

View only this post Read Less

Meteosat-9 10.8 µm IR images

EUMETSAT Metetosat-9 10.8 µm IR images (above; also available as a QuickTime movie) showed the evolution of a disturbance that had all the appearances of being another example of a rather rare event: a subtropical cyclone in the South Atlantic Ocean (off the southeast coast of Brazil) during the 09 March 2010 – 10 March 2010 time period.

This system was eventually declared to be tropical cyclone on 10 March, according to this HPC discussion:

SOUTH AMERICA SYNOPTIC DISCUSSION
INTERNATIONAL DESKS
NWS HYDROMETEOROLOGICAL PREDICTION CENTER
CAMP SPRINGS MD
847 AM EST WED MAR 10 2010

GFS DATA AT FTPPRD.NCEP.NOAA.GOV/PUB/DATA/NCCF/COM/GFS/PROD/

SYNOPSIS (VALID FROM 00Z MAR 10). THE UPPER LEVEL ANALYSIS SHOWS A CLOSED LOW NEAR 33S 45W EXTENDING A SHORT WAVE TROUGH TO THE NORTHWEST INTO BRASIL ALONG 20S 50W. THIS FEATURE IS DECOUPLING FROM A WARM CORE SURFACE LOW OFF THE COAST OF BRASIL…WITH CLOSED CIRCULATION ESTIMATED NEAR 29.6S 48.2W. ALTHOUGH A TIGHT/COMPACT STORM…IT IS NOW CLASSIFIED AS A TROPICAL CYCLONE RATHER THAN SUBTROPICAL.

On 13 March this storm was given the name “Anita” by the Brazilian MetSul weather center . Note that Brazil has only had one documented case of a land-falling tropical cyclone that had reached hurricane intensity — “Catarina” in March 2004.

A false-color NOAA-19 Red/Green/Blue (RGB) image using channels 01/02/04 (below) displayed a nice view of the tropical cyclone on 10 March. The low-level circulation (clouds with a slightly yellow hue) was becoming partially exposed, with a large burst of convection occurring in the southwest quadrant of the cyclone.

NOAA-19 false-color Red/Green/Blue (RGB) image

On a closer view using the corresponding NOAA-19 10.8 µm IR image (below), note the presence of a packet of gravity waves which was propagating southwestward away from the region of coldest overshooting tops (which were around -70º C, darker black color enhancement).

NOAA-19 10.8 µm IR image

A later animation of IR imagery from the CIMSS Tropical Cyclones site (below) showed the development of additional convective bursts within the southern portion of the cyclone. Even well to the northwest of the center of the circulation, there was a ship report showing wind speeds of 35 knots.

IR images + ship reports

.

IR image + deep layer wind shear

The tropical cyclone formed in an environment characterized by a moderate amount of deep layer wind shear (above), over a region of sea surface temperatures that were near 25º C (below).

Sea surface temperature analysis

Additional details and images of this South Atlantic tropical cyclone can be found at the Weather Underground, AccuWeather, and NASA.

View only this post Read Less

AVHRR false color RGB image + map of land boundaries (green)

Andy Heidinger (NOAA/NESDIS/Advanced Satellite Products Branch) pointed out a very long and thin cloud feature, which can be seen near the center of the false color Red/Green/Blue (RGB) image created using AVHRR imagery (above). In this particular RGB image, low clouds appear dark blue, while cirrus clouds are white. The cloud feature of interest (which stretched from the North Slope region of Alaska westward across the Arctic Ocean to the north of Siberia on 23 February 2010) appeared to be over 1000 km long and less than 10 km wide — a perfect candidate for the “What the heck is this?” blog category!

With a strong high pressure cell in place over the North Pole, it is possible that this thin cloud arc marked the leading edge of a relatively weak cold frontal boundary. The southward progress of this cloud feature could be followed on a sequence of AWIPS images of AVHRR 12.0 µm IR channel data (below) — the cloud arc was highlighted with a yellow to cyan color enhancement, representing IR brightness temperatures of -10º to -20º C. In addition, well offshore of the northeastern coast of Alaska you could also see the warmer thermal signature (denoted by the yellow color enhancement) of large thin spots and cracks forming in the the sea ice covering the Arctic Ocean.

It may be pure coincidence, but when this thin cloud arc passed southward across northern Alaska coastal station PAWI (Wainwright), they briefly reported freezing fog and a drop in visibility to 0.5 mile.

AVHRR 12.0 µm IR images

The progression of this cloud band could also be seen on a sequence of grayscale AVHRR composite IR images (below, courtesy of Matthew Lazzara, AMRC). The darker appearance of the cloud arc on the grayscale images supported the idea that this was indeed a relatively warm low cloud feature.

AVHRR composite IR images

AVHRR Cloud Type product

An AWIPS image of the AVHRR Cloud Type product (above) indicated that the cloud arc feature was composed primarily of supercooled water droplets (green color enhancement). The time of the AVHRR Cloud Type image corresponds to the time when Wainwright (station identifier PAWI) reported a brief period of freezing fog as the cloud arc passed southward through the area.

The corresponding AVHRR Cloud Top Height product (below) indicated that the top of the thin cloud band was in the 2-3 km range (darker purple color enhancement).

AVHRR Cloud Top Height product

View only this post Read Less

GOES-12 (top panels) vs GOES-13 (bottom panels) visible channel images

McIDAS images of GOES-12 and GOES-13 visible channel data (above) showed the slow eastward movement of ice across Lake Erie on 19 February 2010. Though not particularly strong, the westerly winds across the region were likely a factor in the ice movement. The improvement in GOES-13 Image Navigation and Registration (INR) is immediately obvious, with significantly less image-to-image “wobble” compared to GOES-12 — this enables the ice motion to be tracked more accurately. Note: GOES-13 is scheduled to replace GOES-12 as the operational GOES-East satellite on 14 April 2010.

A closer view using 250-meter resolution Terra and Aqua MODIS true color images from the SSEC MODIS Today site (below, viewed using Google Earth) revealed that there was still some land-fast ice along the far southern shore of Lake Erie, but most of the ice field was indeed moving eastward during the 103 minutes separating the times of the Terra satellite overpass (16:34 UTC) and the Aqua satellite overpass (18:17 UTC).

Terra and Aqua MODIS true color images (viewed using Google Earth)

An AWIPS image of the MODIS Sea Surface Temperature (SST) product (below) indicated that the water temperatures in the ice-free portions of Lake Erie were in the 32-33º F range (violet color enhancement).

MODIS Sea Surface Temperature product

View only this post Read Less