GOES-12 3.9 µm shortwave IR channel images (click image to play animation)

As a result of prolonged drought and a mid-summer heat wave across southern Chile, a number of wildfires were burning in parts of the region on 01 January – 02 January 2012 (surface analysis). GOES-12 3.9 µm shortwave IR images (above; click image to play animation) showed a number of fire “hot spots” (yellow to red color enhancement) between Concepcion (station identifier SCIE) and Chillan (station identifier SCCH) from the late afternoon on 01 January until the early morning hours on 02 January.

During the subsequent daytime hours, GOES-12 0.63 µm visible channel images (below; click image to play animation) revealed a long hazy smoke plume that was drifting northwestward out over the adjacent Pacific Ocean. As daytime heating increased, cumulus clouds with a few thunderstorms could also be seen developing farther inland over the higher terrain of the Andes Mountains.

GOES-12 0.63 µm visible channel images (click image to play animation)

Rawinsonde data from Santo Domingo (station identifier SCSN) at 12 UTC indicated that southeasterly winds existed near the top of the deep temperature inversion, between 741 hPa (2.6 km) and 700 hPa (3.1 km) — so this is likely the approximate altitude of the smoke plume seen drifting toward the northwest on the GOES-12 visible satellite imagery.

Santo Domingo, Chile rawinsonde data plot

A high-resolution MODIS true color image of the fire smoke plume can be seen on the NASA Earth Observatory site.

GOES-12 0.63 µm visible channel images

Hurricane Adrian developed into a Category 3 hurricane early in the day on 09 June 2011. McIDAS images of GOES-12 0.63 µm visible channel data (above) initially showed a well-defined eye before it began to get partially obscured by the high clouds of a central dense overcast (CDO).

DMSP SSMIS 85 GHz microwave images from the CIMSS Tropical Cyclones site (below) revealed a distinct eye at 12:09 UTC and 14:56 UTC.

DMSP SSMIS 85 GHz microwave images

GOES 10.7 µm IR images (below) also briefly showed a well-defined eye early in the day, which later filled in a bit beneath the CDO as a curved band of cold high clouds began to wrap around the eastern and northern quadrants of the hurricane.

GOES 10.7 µm IR images

The circulation of Hurricane Adrian could be clearly seen on an AWIPS image of ASCAT scatterometer winds overlaid on a GOES IR image (below).

ASCAT scatterometer winds (overlaid on GOES IR image)

AWIPS images of the MIMIC Total Precipitable Water (TPW) product (below) showed that Adrian was tapping moisture from the Inter-Tropical Convergence Zone (ITCZ) / “Monsoon Trough”, which was located at approximately 10º North latitude over the eastern Pacific Ocean.

MIMIC Total Precipitable Water product

===== 10 June Update =====

Hurricane Adrian intensified to a Category 4 storm on 10 June 2011. 4-km resolution GOES 10.7 µm IR channel images (below; click image to play animation) continued to show a well-defined eye structure.

GOES 10.7 µm IR images (click image to play animation)

A closer view of the eye could be seen using 1-km resolution GOES visible channel images (below; click image to play animation).

GOES visible channel images (click image to play animation)

The intensity of Hurricane Adrian was expected to decrease as the storm began to move over colder waters, as seen on an image of the Sea Surface Temperature (SST) analysis (below).

Sea Surface Temperature (SST) analysis

GOES-12 0.65 µm visible channel images

An explosive eruption of the Puyehue-Cordón Caulle volcano in Chile occurred on 04 June 2011. GOES-12 0.65 µm visible channel images (above) showed a darker gray ash cloud punching above the meteorological cloud deck around 18:15 UTC, with the ash cloud quickly spreading southeastward and moving over Bariloche, Argentina (station identifier SAZS).

A comparison of GOES-12 3.9 µm shortwave IR and 10.7 µm IR window channel images (below) revealed a pronounced and persistent “hot spot” signature (dark black pixels) at the summit of the volcano on the shortwave IR images — while the long and narrow cold high-altitude volcanic cloud (exhibiting IR brightness temperatures around -60º C, darker red color enhancement) could be seen spreading southeastward for a great distance on the IR window images.

GOES-12 3.9 µm shortwave IR (top) and 10.7 µm IR window (bottom) images

CIMSS activities in the GOES-R Proving Ground include the generation of real-time volcanic ash retrieval products (using Meteosat SEVIRI data as a proxy for GOES-R data), which showed a significant volcanic ash plume emerging over the Atlantic Ocean (below).

SEVIRI volcanic ash retrieval products

MODIS 0.65 and 0.87 µm visible images (centered on Tuscaloosa, Alabama)

A comparison of 250-meter resolution Aqua MODIS 0.65 µm and 0.87 µm visible channel images centered on Tuscaloosa, Alabama on 28 April 2011 (above) showed signatures of a few of the larger and longer tornado damage paths from the historic tornado outbreak (SPC storm reports) that occurred on 27 April 2011. A collection of GOES, POES AVHRR, and MODIS images of the tornado outbreak are available on a separate CIMSS Satellite Blog post.

A comparison of before (17 April 2011) and after (28 April 2011) 250-meter resolution MODIS true color Red/Green/Blue (RGB) images from the SSEC MODIS Today site (below) also showed a few of the tornado damage paths — though some of the damage paths were not as evident as they were on the single-channel visible images above. In addition, another long tornado track can be seen across western Georgia.

MODIS true color images before (17 April) and after (28 April) the tornado outbreak

Below is the same before/after MODIS true color image comparison, viewed using Google Earth.

Before (17 April) and after (28 April) MODIS true color images (viewed using Google Earth)

A comparison of AWIPS images of 1-km resolution MODIS 0.65 µm visible channel data and the corresponding 1-km resolution Normalized Difference Vegetation Index (NDVI) product (below) revealed that a few of the larger tornado damage paths were characterized by a slightly lower NDVI value (lighter green color), due to the downed trees and damaged vegetation.

MODIS 0.65 µm visible image + MODIS Normalized Difference Vegetation Index product

One of the larger tornado damage tracks across northwestern Alabama was also apparent on GOES-13 0.63 µm visible channel imagery (below). An animation helps to confirm that the feature is not a contrail or some other type of linear cloud feature (which would be moving in an animation, rather than stationary). While the north-south spatial resolution of the GOES imager visible detectors is 1.0 km at the satellite sub-point (over the Equator), with the larger geostationary satellite viewing angle the north-south spatial resolution over northern Alabama is about 1.38 km.

GOES-13 visible images (centered along Mississippi/Alabama border region)

It is interesting to note that the larger tornado damage path which can been seen across northwestern Alabama on the GOES-13 visible image is very difficult to identify on the corresponding GOES-12 0.65 µm visible image (below). This is due to a change is the spectral response function on the visible channel of the GOES-13 and later series of imagery instruments — for more details, see this CIMSS Satellite Blog post.

GOES-13 0.63 µm visible image (left) + GOES-12 0.65 µm visible image (right)

Update 29 April 2011: A Terra MODIS pass from late morning on April 29th shows the damage paths more clearly through central Alabama. As the churned-up vegetation within the damage path slowly browns, the contrast to undamaged vegetation outside the damage path should increase, allowing for a clearer picture of the damage path. The bowtie-corrected MODIS imagery for visible channel 1 (0.65 microns, below) and visible Channel 2 show three lines of damage through central Alabama. (Here is an animation of the two visible channels).

Since this day was less cloudy than the previous, here is a better before (17 April) versus after (29 April) comparison of 250-meter resolution MODIS true color images from the SSEC MODIS Today site (below, viewed using Google Earth).

MODIS true color images from before (17 April) and after (29 April)

Update 30 April 2011: A few of the tornado damage paths across northern Alabama were also evident on an AWIPS image of the 1-km resolution MODIS Land Surface Temperature (LST) product (below). The LST values within the damage paths were in the low to middle 80s F (darker red color enhancement), compared to surrounding LST values in the upper 70s F (orange color enhancement), indicating that the damage paths with destroyed vegetation and tornado debris were able to heat up a few degrees more than the adjacent undisturbed vegetation. The urban areas of the cities of Tuscaloosa (KTCL) and Birmingham (KBHM) also exhibited warmer LST values (darker red) than the surrounding less urbanized, more densely forested areas.

MODIS Land Surface Temperature product

CIMSS participation in GOES-R Proving Ground activities includes making a variety of MODIS images and products available for National Weather Service offices to add to their local AWIPS workstations.

GOES-12 0.65 µm visible images (click image to play animation)

Subtropical Storm “Ariani” developed off the southeast coast of Brazil on 15 March 2011. GOES-12 0.65 µm visible channel images (above; click image to play animation) displayed a number of overshooting tops within the large cloud shield of the disturbance. Some hints of a cyclonic (clockwise in the Southern Hemisphere) circulation could be seen both within the southern flank of the cloud shield, and also just to the southwest of the cloud shield over the open water.

GOES-13 10.7 µm IR images from the CIMSS Tropical Cyclones site (below; click image to play animation) showed a broad area of old cloud tops associated with this feature.

GOES-13 10.7 µm IR images (click image to play animation)

A GOES-13  IR/Water Vapor difference product (below; click image to play animation) did indicate that there was a large area of overshooting tops (red to violet color enhanced areas). For more information on this product and its application to tropical cyclone intensity analysis, see Olander and Velden, 2009.

GOES-13 IR/Water Vapor difference product (click image to play animation)

GOES-13 (left) and GOES-12 (right) 3.9 µm shortwave IR images

McIDAS images of GOES-13 and GOES-12 3.9 µm shortwave IR (IR channel 2) data (above) revealed the “hot spot” (yellow to red color enhancement) due to the eruption of the Tungurahua volcano in the South American country of Ecuador on 04 December 2010. The summit of the volcano is located south-southeast of the city of Latacunga (station identifier SELT). According to an ash advisory issued by the Washington DC Volcanic Ash Advisory Center (VAAC), ash was estimated to be extending upward to altitudes about 26,000 feet around this time.

Note that at times there are sight differences in the size and intensity of the volcano hot spot, due to the different viewing angles from the GOES-13 satellite (located at 75º West longitude) and the GOES-12 satellite (located at 60º West longitude). Also note the improved image navigation and registration (INR) with GOES-13, which keeps the volcano hot spot centered at the same location during the image animation.

Tropical Storm Igor has formed in the far eastern Atlantic, just south of the Cape Verde Islands. GOES-12 imagery from today (above) shows a sheared system with persistent and abundant convection southwest of a low-level swirl of clouds. (That low-level swirl is most evident in the 1145 UTC image). Igor formed just north of a region of significant shear (shown here) and the shear may slow intensification of the storm.

Other environmental factors favor strengthening. For example, MIMIC total precipitable water (from this website) shows abundant moisture surrounding the developing storm. An analysis of the Saharan Air Layer shows little signal in the region of the storm. Igor’s forecast path is over warm water. Forecasts from the National Hurricane Center suggest slow strengthening as the system moves westward across the tropical Atlantic.

For more information on Igor, visit the CIMSS Tropical Weather Website and the National Hurricane Center Website.

(Added: Click here for views centered on Igor from GOES-12, GOES-13 and GOES-15, satellites over the Equator at 60 W, 75 W and 89 W, respectively)

The parade of tropical impulses moving westward off of Africa into the tropical Atlantic that has produced Hurricanes Danielle and Earl, and Tropical Storm Fiona, has now yielded a new Tropical Depression, Number 9, that will become Gaston if it achieves Tropical Storm status. Some environmental conditions favor intensification, and some work against it.

The tropical depression is moving over warm sea surface temperatures, and is in an environment of low shear, two factors that argue for slow intensification of the system. However, an analysis of Saharan Air using Meteosat data (diagnosed as a split-window technique using 10.8 micron and 12.0 micron data) shows very dry air surrounding the storm. (See image below). Saharan Air Layers greatly impede the development of tropical cyclones. MIMIC Total Precipitable Water (from this site) also shows very dry air surrounding half of the developing storm. Water vapor imagery from GOES-East and from GOES-12 both show very dry air surrounding the tropical depression.

Visible and Infrared imagery from the storm this morning (below) show the impact of forward scattering and back-scattering on the detection of thin clouds. The visible imagery at 0815 UTC (left), when the sun is low in the sky and forward scattering dominates, suggests far more cloudiness than at 1745 UTC (right) when the sun is high in the sky and backscattering dominates. Infrared imagery, however, shows little change in the amount of detected cloudiness. Thin cirrus detection by visible satellite is easiest for very low sun angles; as the sun rises higher in the sky, cirrus clouds become less distinct in visible imagery. Note that GOES-R will include a detector sensitive to radiation at 1.3 microns to highlight cirrus clouds regardless of the Sun’s position (for example, see this comparison of MODIS visible, 1.3 µm near-IR, and 11.0 µm IR images).

For up-to-date information on the tropical systems in the Atlantic, visit the CIMSS tropical website, or the National Hurricane Center website.

(Added: TD #9 became Tropical Storm Gaston as of 5 PM EST on 1 September)

The infrared imagery from Friday morning, 27 August, shows three separate tropical systems — in various stages of development — over the Atlantic Ocean, with a fourth system poised to move out into the tropical Atlantic from Africa. Meteosat imagery shows a fifth system moving westward over central Africa (These images are available at the University of Wisconsin SSEC‘s Geostationary Image Browser).

Hurricane Danielle, above, as seen by GOES-15 (Click here for a longer loop from GOES-13) is a mature Category IV storm over the central Atlantic, with well developed outflow in an environment characterized by warm sea-surface temperatures and small vertical wind shear. The cold cloud tops surrounding the mostly cloud-free eye have brightness temperatures in the 195-200 K range. GOES-15 IR Imagery shows similar temperatures.

Tropical Storm Earl, above, is in an environment not as conducive to development as Danielle. Analyses from the CIMSS Tropical Weather Web Page show a storm track that has recently passed over cooler ocean water. More importantly, Earl is surrounded by dry Saharan Air and that dry air is suppressing some of the convective activity needed to fuel system development. Analyses for Total Precipitable Water (taken from this site) also show Earl entraining dry air. However, the projected path of Earl is towards a warmer sea surface in a more moisture-rich environment, so intensification is forecast.

In contrast the Earl, the disturbance off the coast of Africa, above, as viewed from GOES-12, is over a region of warm water, and is south of dry air with origins over the Sahara. Shear values over the storm are low, so intensification should occur.

The next names in the list for the Atlantic are Fiona, Gaston and Hermine. For the latest on these storms, visit the National Hurricane Center website.

(Added: Dan Lindsey at CIRA notes the similarity to 1995. Here is an image from 2345 UTC on 29 August 1995 showing a similar line-up of storms across the tropical Atlantic).

GOES-15 full disk 6.5 µm water vapor images (at 30-minute intervals)

The normal operational GOES scan schedule provides one full disk image every 3 hours — but as part of the GOES-15 NOAA Science Test, the GOES-15 satellite was placed into a mode which allowed one full disk image to be scanned every 30 minutes. McIDAS images of GOES-15 6.5 µm “water vapor channel” data (above) showed a series of these 30-minute interval full disk images during a 24-hour period during 21 August – 22 August 2010. It should be pointed out that the ABI instrument on the GOES-R satellite will provide a full disk image every 5 minutes!

The image artifacts seen moving from west to east across the northern portion of the images (from 05:15 UTC to 06:45 UTC) were due to stray light contamination. This stray light problem affects the image quality to varying degrees during the Autumn and Spring season “eclipse periods”, since the newer GOES satellites (GOES-13 through GOES-15) have larger batteries that allow them to continue to operate through the eclipse periods when the satellite is in the Earth’s shadow.

Note that some of warmest water vapor brightness temperatures (yellow to orange color enhancement) — which normally indicate areas of very dry air in the middle to upper troposphere — were found over the central US during this particular period. This region of warm/dry air showed up well on GOES-11, GOES-12, GOES-13, and GOES-15 water vapor images (below), although the signal was less obvious on the 6.7 µm GOES-11 imagery (which is a spectrally narrow water vapor channel at an 8 km spatial resolution, compared to the spectrally broad 4-km resolution channels on GOES-12 and later). Each of the water vapor images is displayed in the native projection of the particular satellite.

GOES-11, GOES-12, GOES-13, and GOES-15 water vapor images

A closer view using AWIPS images of GOES-13 6.5 µm water vapor channel data (below) showed the evolution of this feature, which was becoming warmer/drier as a middle-tropospheric ridge of high pressure was building across the region. Water vapor brightness temperatures were as warm as -9.5º C (orange color enhancement), which is very unusual to see covering such a large area over the central US.

GOES-13 6.5 µm water vapor images (with rawinsonde locations)

4-panel images displaying the 3 GOES Sounder water vapor channels (6.5 µm, 7.0 µm, and 7.4 µm) along with the standard GOES-13 imager 6.5 µm water vapor channel are shown below. The water vapor channel weighting function of each of these channels peaks at different altitudes, which is obvious by the difference in water vapor brightness temperatures on each of the images (warmer/drier areas are enhanced with yellow to red colors).

GOES Sounder and GOES Imager water vapor channel images

The GOES-13 sounder and GOES-13 imager water vapor weighting functions for Topeka, Kansas (below) indicated that the altitudes of the weighting function peaks did indeed descend from 12 UTC on 21 August to 00 UTC on 22 August as the middle troposphere became warmer/drier — but the altitudes of the various weighting function peaks were not significantly lower than those computed using the US Standard Atmosphere.

Topeka, Kansas water vapor weighting function plots (compared to US Standard Atmosphere)

The rawinsonde date from Topeka, Kansas at 12 UTC on 21 August and at 00 UTC on 22 August are shown below. While the air aloft was certainly dry, it was also quite warm at those altitudes — and the warm temperature of this mid-tropospheric air was likely contributing to the unusually warm/dry appearance on the water vapor imagery. This case helps to highlight the fact that the water vapor channels are also IR channels, so they are sensitive to temperature as well — and some of the signal of the features seen on the imagery may be due to temperature as well as moisture aloft.

Topeka, Kansas rawinsonde plots (12 UTC 21 August, 00 UTC 22 August)