Convective cloud top properties: temperature and ice particle size

June 19th, 2008

MODIS IR images (Animated GIF)

Numerous severe thunderstorms developed along the foothills of the Rockies and the High Plains during the afternoon hours on 19 June 2008. AWIPS images of the MODIS 11.0 µm “IR window” channel, Cloud Top Temperature product, and 3.7 µm “shortwave IR” channel (above) showed some of these areas of convection around 20:00 UTC (2 PM local time). Shortly after the time of these MODIS images, hail up to 2.75 inches in diameter was reported near Pueblo, Colorado (SPC storm reports). About 90 minutes after the time of the MODIS images, hail of 4.50 inches in diameter and a wind gust to 81 mph was produced by the storm that was seen developing in the Texas panhandle region. Although these storms exhibited similar cold cloud top temperatures on the IR window image (-60 to -77º C, red to black to gray color enhancement) and the Cloud Top Temperature product (-60 to -69º C, darker blue to light violet color enhancement), note the darker gray appearance of many of the westernmost storms on the 3.7 µm shortwave IR image — this darker signal is due to a dominance of smaller anvil top ice particles (in contrast with the brighter white appearance due to larger anvil top ice particles associated with the storms farther to the east). Anvils composed of smaller ice particles are more reflective of incident solar radiation, which results in significantly warmer shortwave IR brightness temperatures (in this case, +5 to +20º C, compared to -10 to -15º C in the brighter white anvil features composed of larger ice particles). According to Lindsey et al. (2006), regions such as the high plains and mountains (having environments with relatively dry boundary layers, steep lapse rates, and large vertical shear values) tend to favor thunderstorms with enhanced 3.9-μm reflectivity.

With cloud top temperatures in the -60 to -70º C range, no supercooled water droplets could have been present at the cirrus anvil top — this was confirmed by the MODIS “cirrus detection channel” image in tandem with the MODIS Cloud Phase product (below), which indicated ice phase (salmon color enhancement) for all the convective storm cirrus anvil features.

MODIS cirrus channel + cloud phase product (Animated GIF)

 

Additional reference:

  Cloud-top Structure of Northeast Colorado Thunderstorms May 24, 2005

GOES-11 and GOES-12 in Rapid Scan Operations

June 18th, 2008

GOES-11 + GOES-12 visible images (Animated GIF)

Both GOES-11 (GOES-West) and GOES-12 (GOES-East) were placed into Rapid Scan Operations (RSO) during the afternoon and early evening on 18 June 2008, to monitor the development of severe thunderstorms across the central US (SPC storm reports). In RSO mode, images are available at 5-7 minute intervals (instead of the usual 15 minute intervals for standard operations) — however, the RSO image times are not exactly the same for GOES-11 and GOES-12. A side-by-side comparison of GOES-11 and GOES-12 visible channel images centered on Bismarck, North Dakota (above) showed the formation and intensification of thunderstorms that produced hail as large as 2.75 inches in diameter at 22:48 UTC and 4.25 inches in diameter at 00:22 UTC (both times at a location between Bismarck KBIS and Garrison KN60), and a tornado around 00:25 UTC (just west of Bismarck KBIS). Both the GOES-11 and GOES-12 images are displayed in their native satellite projections, so the cloud features (and the area shown) appear a bit different due to the differing satellite viewing angles.

Continued flooding in the Upper Midwest

June 17th, 2008

MODIS false color images (Animated GIF)

The historic Central US river flooding event of 2008 continued into mid-June across parts of the Upper Midwest states (Google Earth map of river gauges). A sequence of MODIS false-color images from the SSEC MODIS Today site (above) covers the period from 01 June to 17 June, and shows the increase in flooded and inundated rivers across eastern Iowa, northwestern Illinois, and southwestern Wisconsin. Water has a much darker appearance in these false color images, so the swollen rivers tend to stand out in contrast to the densely-vegetated areas (green colors) which were characterized by a high concentration of trees, or the sparsely-vegetated agricultural fields (tan to light brown colors) where crops had only recently been planted.In particular, the MODIS images show the flooding of a large area of Iowa farmland adjacent to the Missisippi River (near the bottom center of the images), after the Iowa River broke through a levee near Oakville, Iowa on 16 June.

MODIS true color image (Google Earth)

Flooding also forced the closure of the Great River Bridge that connects Iowa and Illinois along Highway 34 near Burlington, Iowa. On the MODIS true color image from 17 June (above; displayed using Google Earth), note the well-defined southern edge of the flooded area (darker brown colors) between Oakfield and Kingston, Iowa — this west-to-east oriented water boundary was likely due to a massive sandbagging effort along a diversion channel that runs eastward into the Mississippi River. This narrow diversion channel can be seen in more detail using Google Maps satellite imagery (below).

Google Maps satellite image

Farther to the west, a June 2007 vs. June 2008 comparison of MODIS true color images centered over southern Minnesota and northern Iowa (below; interactive image fader) revealed how delayed the “green-up” of cultivated crops (primarily corn and soybeans) was across that particular region. Cooler than normal temperatures in May 2008 followed by the unusually heavy rainfall and flooding in June 2008 conspired to delay the planting of crops across much of that area.

MODIS true color images (Animated GIF)

An AWIPS image of the MODIS Normalized Difference Vegetation Index (NDVI) product from 16 June (below) showed that NDVI values were as low as 0.2 to 0.3 over some portions of Minnesota and Iowa.

MODIS Normalized Difference Vegatation Index (NDVI)

The “normal” Vegetation Fraction over that same region for the month of June (below) should be in the 0.5 to 0.7 range (based upon a 5-year monthly climatology).

June mean vegetation fraction