Moisture Changes as viewed in the Cirrus Channel

March 14th, 2018 |

GOES-16 ABI Band 3 (0.86 µm) Reflectance, hourly from 1632-1932 UTC on 14 March 2018 (Click to enlarge)

Skies were clear over much of the southern Plains on 14 March 2018, as noted in the animation above that shows hourly GOES-16 ABI Channel 3 (0.86 µm) Imagery. Differences in absorption/reflectance between water and land yield excellent discrimination between lakes and land over Oklahoma and adjacent states.  GOES-16 ABI “Cirrus Channel” (Band 4, at 1.38 µm) shows little reflectance in the area over Oklahoma, except where cirrus clouds are present over western Oklahoma.  The rest of Oklahoma is dark because water vapor in the atmosphere is absorbing energy at 1.38 µm. An animation — also at hourly intervals — is shown below.  This uses the default enhancement in AWIPS, with reflectance values between 0 and 50 shown.

GOES-16 ABI Band 4 (1.37 µm) Reflectance, hourly from 1632-1932 UTC on 14 March 2018 with default AWIPS Enhancement (Click to enlarge)

If you alter the Band 4 enhancement to change the bounds from 0-50 (the default) to 0-2 (!), as was done in the animation below showing data every 5 minutes, a gradient in reflectance becomes apparent, and surface features — specifically lakes — over central Oklahoma that are initially present slowly become obscured as the gradient moves to the east. This gradient shows differences in moisture. The atmosphere that is moving into eastern Oklahoma from central Oklahoma is slightly more moist.  (Compare the morning sounding at Amarillo, for example, with a total precipitable water of 0.38″ to the morning sounding at Little Rock, with a total precipitable Water of 0.14″)

GOES-16 ABI Band 4 (1.37 µm) Reflectance, from 1632-1947 UTC on 14 March 2018 with default AWIPS Enhancement modified as described in text (Click to animate)

GOES-16 data includes channel differences and level 2 products that also confirm the slow increase in moisture. The Split Window Difference field, shown below with the default enhancement (Click here to see the same animation with the Grid MidRange Enhanced enhancement), and the Total Precipitable Water, at bottom, show a slow increase in moisture. These increases were above the surface: surface dewpoints in this region (source) were not increasing greatly.

Split Window Difference (10.3 µm – 12.3 µm) from 1632 – 1947 UTC on 14 March (Click to enlarge)

GOES-16 Total Precipitable Water Baseline Product, 1632-1947 UTC on 14 March 2018 (Click to enlarge)

Isolated cirrus cloud feature over Louisiana

November 10th, 2017 |

GOES-16 Visible (0.64 µm) images, with surface station identifiers plotted in yellow [click to play MP4 animation]

GOES-16 Visible (0.64 µm) images, with surface station identifiers plotted in yellow [click to play MP4 animation]

* GOES-16 data posted on this page are preliminary, non-operational and are undergoing testing *

An isolated cloud feature moving east-southeastward across Louisiana on 10 November 2017 caught the attention of several people on Twitter — GOES-16 “Red” Visible (0.64 µm) images (above) showed the motion of this cloud during the 1317-2052 UTC period.

In a 3-panel comparison of GOES-16 “Red” Visible (0.64 µm), Near-Infrared “Cirrus” (1.37 µm) and “Clean” Infrared Window (10.3 µm) images (below), the strong signature (bright white) on the 1.37 µm imagery indicated that this feature was a cirrus cloud. The uncharacteristically-warm Infrared brightness temperatures exhibited by this feature were due to the fact that the thin cirrus allowed warmer thermal radiation from the surface to pass through the cloud and reach the satellite detectors.

GOES-16 Visible (0.64 µm, top), Near-Infrared

GOES-16 Visible (0.64 µm, top), Near-Infrared “Cirrus” (1.37 µm, middle) and “Clean” Infrared Window (10.3 µm, bottom) images [click to play MP4 animation]

Rawinsonde profiles from Lake Charles and Slidell, Louisiana at 12 UTC (below) showed the presence of a moist layer aloft (at an altitude around 9.5 km or 31,100 feet) — the cirrus cloud feature likely resided within this moist layer, which would explain why the cloud was slow to dissipate. Air temperatures within this moist layer were in the -40 to -50ºC range, and winds were from the west-northwest at speeds of 30-35 knots (which was consistent with the cloud motion seen on satellite imagery).

Rawinsonde data for Lake Charles and Slidell, Louisiana at 12 UTC on 10 November [click to enlarge]

Rawinsonde data for Lake Charles and Slidell, Louisiana at 12 UTC on 10 November [click to enlarge]

Even with the higher spatial resolution Infrared Window imagery (1 km, vs 2 km at the satellite sub-point for GOES-16) of Terra MODIS (below), the minimum Infrared brightness temperature of the cirrus cloud feature was still a relatively warm -31ºC.

Terra MODIS Visible (0.65 µm), Cirrus (1.375 µm) and Infrared Window (11.0 µm) images [click to enlarge]

Terra MODIS Visible (0.65 µm), Cirrus (1.375 µm) and Infrared Window (11.0 µm) images [click to enlarge]

Another interesting aspect of this small cirrus cloud is that it was casting a shadow to the north (due to the low November sun angle) — and the Terra MODIS Land Surface Temperature product (below) indicated that LST values were about 10 degrees F cooler within the shadow (low to middle 60s F) compared to adjacent sunlit ground (low to middle 70s F). That particular area was not normally cooler in terms of LST values (because of varying vegetation, soil type, a deep lake, etc.), since it did not show up as a cooler feature on the following day.

Terra MODIS Visible (0.65 µm) image and Land Surface Temperature product [click to enlarge]

Terra MODIS Visible (0.65 µm) image and Land Surface Temperature product [click to enlarge]

Additional images and ground-based photos of the cirrus cloud feature can be found on this AccuWeather blog.

An RGB computed using the GOES-16 Cirrus Channel

November 3rd, 2017 |

Cloud Type RGB at 1502 UTC on 3 November 2017 (Click to enlarge)

GOES-16 data posted on this page are preliminary, non-operational and are undergoing testing

Red-Green-Blue (RGB) Composite Images are a handy way of showing information from multiple satellite bands (or band differences) at once. The image above shows an RGB created by NOAA Scientist Andy Heidinger that uses the GOES-16 Visible Band (0.64 µm) as the green component, Snow-Ice Band (1.61 µm) as the blue component and Cirrus Band (1.38 µm) as the red component to tease out information about Cloud Type.  The Cirrus Channel (unique to GOES-16 as far as Geostationary Satellites are concerned) is a handy channel to use in an RGB because it discriminates very well between high clouds and low clouds.  In a moist environment, low clouds are not apparent at all in the Cirrus Band.  The toggle below shows the Visible, Snow/Ice and Cirrus Channels at 1502 UTC.  Low clouds over Kansas have no signal in the Cirrus channel — there are other differences as well, of course.

In the RGB, Thin cirrus clouds (for example, the contrails over Illinois) are red, opaque ice clouds (over the western Atlantic) are yellow (having a contribution from both Red and Green Components), Low Clouds (over the southern Plains) are Cyan (having a contribution from Blue and Green), snow is Green, and lofted water clouds are white (having a contribution from all three). As the atmosphere dries, the amount of lofting necessary for the Cirrus channel to view a cloud composed of water droplets (and therefore white in the RGB) decreases.

GOES-16 Imagery at 1502 UTC on 3 November 2017: Snow/Ice (1.61 µm), Visible (0.64 µm) and Cirrus Channels (1.38 µm) (Click to enlarge)

The Day Land Cloud RGB (sometimes called ‘Natural Color’) can also be used to estimate cloud type. The toggle below shows how the Cloud Type RGB has more gradations between ice cloud type because of the use of the Cirrus Channel.  The Cloud Type RGB also highlights the contrails and thin cirrus more effectively, again because of the use of the Cirrus Channel

Cloud Type RGB (1.38 µm, 0.64 µm, 1.61 µm) and Day/Land/Cloud RGB (1.61 µm, 0.86 µm, 0.64 µm), 1502 UTC on 3 November 2017 (Click to enlarge)

 

Three toggles below show the Snow/Ice and Visible and Cirrus channels zoomed in over Illinois (where contrails are present), over the western Atlantic (where strong convection is occurring) and over the southwestern United States.

GOES-16 Imagery at 1502 UTC on 3 November 2017: Cirrus Channel (1.38 µm), Visible (0.64 µm) and Snow/Ice (1.61 µm) (Click to enlarge)

GOES-16 Imagery at 1502 UTC on 3 November 2017: Cirrus Channel (1.38 µm), Visible (0.64 µm) and Snow/Ice (1.61 µm) (Click to enlarge)

GOES-16 Imagery at 1502 UTC on 3 November 2017: Cirrus Channel (1.38 µm), Visible (0.64 µm) and Snow/Ice (1.61 µm) (Click to enlarge)

GOES-16 also has a Baseline Product that shows Cloud Type. That is shown below. The 1502 UTC Image was incomplete, so the 1507 UTC image is shown.

GOES-16 Cloud Phase, Baseline Product, 1507 UTC on 3 November 2017 (Click to enlarge)

Detection of low clouds on “Cirrus band” imagery

October 29th, 2017 |

GOES-16 Visible (0.64 µm, top), Cirrus (1.37 µm, middle) and Infrared Window (10.3 µm, bottom) images [click to play animation]

GOES-16 Visible (0.64 µm, top), Cirrus (1.37 µm, middle) and Infrared Window (10.3 µm, bottom) images [click to play animation]

* GOES-16 data posted on this page are preliminary, non-operational and are undergoing testing *

The ABI “Cirrus” (1.37 µm) band is centered in a strong water vapor absorption spectral region — therefore it does not routinely sense the lower troposphere, where there is usually substantial amounts of water vapor. Hence, its main application is the detection of higher-altitude cirrus cloud features.

However, in areas of the atmosphere characterized by low amounts of total precipitable water, the Cirrus band can sense clouds (and other features, such as blowing dust) in the lower troposphere. Such was the case on 29 October 2017, when a ribbon of dry air resided over the northern Gulf of Mexico in the wake of a strong cold frontal passage; low-level stratocumulus clouds were very apparent on GOES-16 Cirrus band images (above). Also of note: cloud features associated with Tropical Storm Philippe could be seen east of Florida.

The three GOES-16 Water Vapor bands (Upper-level 6.2 µm, Mid-level 6.9 µm and Lower-level 7.3 µm) highlighted the pocket of dry air that was moving across the northern Gulf of Mexico on that day (below).

GOES-16 Upper-level Water Vapor (6.2 µm, top), Mid-level Water Vapor (6.9 µm, middle) and Lower-level Water Vapor (7.3 µm, bottom) images [click to play animation]

GOES-16 Upper-level Water Vapor (6.2 µm, top), Mid-level Water Vapor (6.9 µm, middle) and Lower-level Water Vapor (7.3 µm, bottom) images [click to play animation]

The MODIS instrument on Terra and Aqua has a 1.37 µm Cirrus band as well; 1619 UTC Terra images (below) also revealed the stratocumulus clouds (especially those over the northeastern Gulf, where the driest air resided). Conversely, note how the low cloud features of Philippe were not seen on the Cirrus image, since abundant moisture within the tropical air mass east of Florida attenuated 1.37 µm wavelength radiation originating from the lower atmosphere.

In addition, the VIIRS instrument — on Suomi NPP, and the upcoming JPSS series — has a 1.37 µm Cirrus band.

Terra MODIS visible (0.65 µm), Cirrus (1.375 µm) and Infrared Window (11.0 µm) images [click to enlarge]

Terra MODIS visible (0.65 µm), Cirrus (1.375 µm) and Infrared Window (11.0 µm) images [click to enlarge]

Hourly images of the MIMIC Total Precipitable Water product (below) showed the ribbon of very dry air (TPW values less than 10 mm or 0.4 inch) sinking southward over the northern Gulf of Mexico. This TPW product uses microwave data from POES, Metop and Suomi NPP satellites (description).

http://cimss.ssec.wisc.edu/goes/blog/wp-content/uploads/2017/10/tpw_17z.png

MIMIC Total Precipitable Water images [click to play animation]