With the very large satellite viewing angle (or “zenith angle”) associated with GOES-15 imagery over Alaska — which turns out to be 73.8 degrees for Fairbanks — the altitude of the peak of the Imager 6.5 µm water vapor weighting function (below) was shifted to higher altitudes (in this case, calculated using rawinsonde data from 12 UTC on 22 June, near the 300 hPa pressure level).The ABI instrument on GOES-R will have 3 water vapor bands, roughly comparable to the 3 water vapor bands on the GOES-15 Sounder — the weighting functions for those 3 GOES-15 Sounder water vapor bands (calculated using the same Fairbanks rawinsonde data) are shown below. Assuming a similar spatial resolution as the Imager, the GOES-15 Sounder bands 11 (7.0 µm, green) and 12 (7.4 µm, red) would have allowed better sampling and visualization of the lower-altitude portion of this particular storm system. The 3 ABI water vapor bands are nearly identical to those on the Himawari-8 AHI instrument; an example of AHI water vapor imagery over part of Alaska can be seen here. As the system departed and the clouds began to dissipate on 22 June, GOES-13 Visible (0.63 µm) images (below) did indeed show evidence of bright white snow-covered terrain on the northern slopes and highest elevations of the Brooks Range. A sequence of 1-km resolution POES AVHRR Visible (0.86 µm) images (below) showed a view of the storm during the 21-22 June period, along with the resultant snow cover on 22 June. However, the snow quickly began to melt as the surface air temperature rebounded into the 50’s and 60’s F at some locations. The increase in fresh snow cover along the northern slopes and the highest elevations of the central and northeastern Brooks Range — most notably from Anaktuvuk Pass to Fort Yukon to Sagwon — was evident in a comparison of Suomi NPP VIIRS true-color Red/Green/Blue (RGB) images from 17 June and 22 June, as viewed using RealEarth (below). The actual time of the satellite overpass on 22 June was 2134 UTC.
A nighttime comparison of Suomi NPP VIIRS Day/Night Band (0.7 µm), Shortwave Infrared (3.74 µm) and Infrared Window (11.45 µm) images at 0853 UTC on 20 June 2016 (above) revealed 2 key features of the large Cedar Fire that had been burning in eastern Arizona: (1) the fire “hot spot” signature (black to yellow to red pixels) on the Shortwave Infrared image, located about 20 miles southwest of Show Low (KSOW), and (2) an approximately 50-mile-wide pall of dense smoke aloft — illuminated by a nearly-full Moon — that had drifted westward then northwestward during the previous 24 hours and was centered northwest of Prescott (KPRC). Note that there was no signature of this smoke feature on the Infrared Window image, since smoke is effectively transparent to infrared radiation.
During the following afternoon hours, a toggle between 2117 UTC Aqua MODIS Near-Infrared “Cirrus detection” (1.61 µm), Visible (0.65 µm), Infrared Window (11.0 µm) and Topography images (below) showed that the smoke aloft had moved northward during the day and was over far northwestern Arizona and southwestern Utah. On the Visible image, the dense layer of smoke obscured the view of surface features that are normally seen on a cloud-free day, but the edges of the smoke feature were difficult or impossible to identify. However, the smoke feature was quite evident on the Near-Infrared “Cirrus detection” image — due to the fact that this spectral band (which will be on the GOES-R ABI instrument) is useful for detecting features composed of particles that are efficient scatterers of light (such as cirrus cloud ice crystals, airborne dust or volcanic ash, and in this case, smoke). As was seen in the VIIRS example above, there was no signature of the smoke on the Infrared Window image — the cooler (lighter gray) shades seen in that region were a result of higher terrain that exhibited cooler brightness temperatures due to more abundant vegetation.An animation of GOES-15 (GOES-West) Visible (0.63 µm) images (below) showed the aforementioned Cedar Fire smoke in northwestern Arizona early in the day (highlighted by a favorable forward scattering sun-satellite geometry), and also showed the smaller smoke plume from the Reservoir Fire that had just begun burning northeast of Los Angeles. In addition, the brief appearance of bright white flashes across Southern California and extreme southern Nevada (as seen on the 1800, 1830, 1841 and 1845 UTC images) were a result of reflection of sunlight from large solar panel farms.
The corresponding GOES-15 Visible (0.63 µm) and Infrared Window (10.7 µm) images (below) revealed cloud-top infrared brightness temperature values as cold as -58º C (darker red color enhancement) at 0030 and 0100 UTC on 04 May.A comparison of Suomi NPP VIIRS false-color “Snow vs cloud discrimination” Red/Green/Blue (RGB), Visible (0.64 µm), Shortwave Infrared (3.74 µm), and Infrared Window (11.45 µm) images at 1834 UTC (above) showed that while a large fire hot spot was apparent on the Shortwave Infrared image, there was no clear indication of any pyrocumulus cloud development at that time. However, a similar image comparison at 2018 UTC (below) revealed that a well-defined pyroCb cloud had formed (with a cloud-top infrared brightness temperature as cold as -60º C, dark red color enhancement) which was drifting just to the north of the Fort McMurray airport (whose cyan surface report is plotted near the center of the images). A 2104 UTC NOAA-19 AVHRR image provided by René Servranckx showed a minimum IR brightness temperature of -59.6º C. A closer look using Suomi NPP VIIRS true-color RGB and Shortwave Infrared (3.74 µm) images from the SSEC RealEarth site (below) showed the initial pyroCb cloud as it had drifted just east of Fort McMurray, with the early stages of a second pyroCb cloud just south of the city. A nighttime comparison of Suomi NPP VIIRS Day/Night Band (0.7 µm) and Shortwave Infrared (3.74 µm) images at 1015 UTC or 3:15 am local time (below; courtesy of William Straka, SSEC) showed the bright glow of the large Fort McMurray wildfire, as well as the lights associated with the nearby oil shale mining activity. A sequence of Suomi NPP VIIRS Shortwave Infrared (3.74 µm) images covering the 02 April – 04 April period (below) showed the diurnal changes as well as the overall growth of the fire hot spot (darker black pixels).
===== 05 May Update =====
The GOES-14 satellite was operating in Super Rapid Scan Operations for GOES-R (SRSOR) mode, providing images at 1-minute intervals — and the scan sector was positioned to monitor the Fort McMurray wildfire on 05 May. GOES-14 Visible (0.63 µm) and Shortwave Infrared (3.9 µm) images (below; also available as a large 133 Mbyte animated GIF) showed the growth of the smoke plume and fire hot spot signature (black to yellow to red pixels).
A 30-meter resolution Landsat-8 false-color Red/Green/Blue (RGB) image (below) showed the size of part of the fire burn scar (darker brown) as well as the active fires (bright pink) along the perimeter of the burn scar.
===== 06 May Update =====
The Fort McMurray fire continued to produce a great deal of smoke on 06 May, and the coverage and intensity of fire hot spots increased during the afternoon hours as seen on 1-minute GOES-14 Visible (0.63 µm) and Shortwave Infrared (3.9 µm) images (below; also available as a large 180 Mbyte animated GIF).
===== 13 May Update =====A comparison of Suomi NPP VIIRS Shortwave Infrared (3.74 µm) and Day/Night Band (0.7 µm) images at 0906 UTC or 3:06 am local time (above) showed the fire hot spots (dark gray to yellow to red pixels) and their nighttime glow.
A time series of VIIRS Shortwave Infrared (3.74 µm) images covering the 04-13 May period (below) revealed the rapid early growth of the fire, and the continued slow spread of the fire periphery toward the Alberta/Saskatchewan border. On 13 May the total size of the area burned by the Fort McMurray fire was estimated to be 241,000 hectares or 595,524 acres.
===== 16 May Update =====Strong southerly winds ahead of an approaching trough axis (surface analyses) created favorable conditions for rapid fire growth on 16 May — GOES-15 Visible (0.63 µm) and Shortwave Infrared (3.74 µm) images (above) showed the development of pyrocumulus clouds (first on the far western flank of the fire around 1930 UTC, then later in the eastern portion of the fire area). This new flare-up of fire activity prompted additional evacuations of some oil sands work camps and facilities north of Fort McMurray.
A comparison of Suomi NPP VIIRS Visible (0.64 µm), Shortwave Infrared (3.74 µm) and Infrared Window (11.45 µm) images at 1932 UTC (below) showed that a small pyroCb had developed, which exhibited a cloud-top IR brightness temperature of -41.48 C.A toggle between the corresponding VIIRS true-color RGB image and Shortwave Infrared images is shown below. A time series plot of surface weather conditions for Fort McMurray (below) shows that during prolonged periods of light winds, the surface visibility dropped below 1 mile at times. The air quality at Fort McMurray was rated as “extreme“, and deemed unsafe for residents to return to the city.
===== 17 May Update =====A shift to westerly winds followed the passage of a surface trough axis on 17 May (surface analyses), which slowed the northward progress of the fire. GOES-15 Visible (0.63 µm) and Shortwave Infrared (3.9 µm) images (above; also available as an MP4 animation) continued to show a great deal of thick smoke over the region, with hot spots from active fires.
However, during the afternoon hours multiple pyroCb clouds were seen to develop along the eastern flank of the fire. A comparison of Suomi NPP VIIRS Visible (0.64 µm), Shortwave Infrared (3.74 µm) and Infrared Window (11.45 µm) images at 2054 UTC (below) revealed the pyroCb clouds, which exhibited cloud-top IR Window brightness temperatures as cold as -57º C (darker orange color enhancement).A comparison of GOES-15 Shortwave Infrared (3.9 µm) and Infrared Window (10.7 µm) images (below; also available as an MP4 animation) showed the development of the pyroCb clouds around 2000 UTC, whose anvil debris moved rapidly southeastward; these pyroCb clouds exhibited a darker gray appearance on the shortwave IR images, along with cloud-top IR Window brightness temperatures as cold as -52º C (light orange color enhancement). Lightning strikes were detected during the early stages of pyroCb growth.
===== 18 May Update =====Daily Suomi NPP VIIRS Shortwave Infrared (3.74 µm) images covering the period 04 May to 18 May 2016 are shown above. The rapid growth of the perimeter of fire hot spots (yellow to red color enhancement) is quite evident during the first few days; patches of thick cloud cover tended to mask the fire hot spots during the middle of the period, but then another increase in hot spot growth is seen beginning on 16 May.
It is interesting to note on a comparison of the 0000 UTC Himawari-8 and GOES-15 Shortwave Infrared (3.9 um) images the large difference in the magnitude of the thermal anomaly — even though the viewing angle was larger for Himawari-8, the superior spatial resolution (2 km at nadir, compared to 4 km with GOES-15) detected a hot spot with an Infrared Brightness Temperature (IR BT) that was 36.6 K warmer (below). The Infrared channels on the GOES-R ABI instrument will also have a 2 km spatial resolution.
With the aid of reflected light from the Moon (in the Waxing Gibbous phase, at 75% of Full), a nighttime view using the Suomi NPP VIIRS Day/Night Band (0.7 µm) from the SSEC RealEarth site (below) revealed the bright glow of the eruption, along with the darker (compared to adjacent meteorological clouds) volcanic ash cloud streaming northeastward. The corresponding VIIRS Shortwave Infrared (3.74 µm) image showed the dark black hot spot of the volcano summit.The volcanic ash cloud continued moving in a northeastward direction, as seen in a sequence of GOES-15 Infrared Window (10.7 µm) and either Terra/Aqua MODIS or Suomi NPP VIIRS retrieved Volcanic Ash Height products from the NOAA/CIMSS Volcanic Could Monitoring site (below). Due to the oblique satellite view angle, the shadow cast by the tall volcanic ash cloud was easily seen on the following early morning (Alaska time) Himawari-8 AHI Visible (0.64 µm) images (below). A closer view (courtesy of Dan Lindsey, RAMMB/CIRA) revealed overshooting tops and gravity waves propagating downwind of the eruption site.
A few select Pilot reports (PIREPs) are shown below, plotted on GOES-15 Infrared Window (10.7 µm) and Aqua MODIS Ash Height derived products. Numerous flights were canceled as the ash cloud eventually began to drift over Western and Interior Alaska (media report).A comparison of Suomi NPP VIIRS Shortwave Infrared (3.74 µm), Day/Night Band (0.7 µm), and true-color Red/Green/Blue (RGB) images (below) showed the volcanic hot spot and the brown to tan colored ash cloud at 2141 UTC on 28 March. Significant ash fall (as much as 2/3 of an inch) was experienced at the village of Nelson Lagoon, located 55 miles northeast of Pavlof (media report). A comparison of the 3 Himawari-8 AHI Water Vapor bands (7.3 µm, 6.9 µm and 6.2 µm) covering the first 14 hours of the eruption from 0000 to 1400 UTC is shown below. Note that the volcanic plume was best seen on the 7.3 µm images, which indicated that it began to move over the coast of Western Alaska after around 0600 UTC; this is due to the fact that the 7.3 µm band is not only a “water vapor absorption” band, but is also sensitive to high levels of SO2 loading in the atmosphere (as was pointed out in this blog post).