It was another clear night over the Great Lakes on 12 May 2025 and NOAA-20 had a descending orbit over western lower Michigan around 0740 UTC that allowed a view of all five Great Lakes. CSPP software running at CIMSS produces both SDR (Sensor Data Record) and EDR (Environmental Data Record) files that can be then manipulated by polar2grid to create imagery. There are sites where imagery is routinely available. The image below, for example, shows NOAA-20 ACSPO SSTs from the CIMSS DBPS website, as discussed here, but that Madison-centered image does not include Lake Ontario, and the colorbar used is a little too warm for mid-May! (Note: RealEarth also used the data)

The direct broadcast website (https://ftp.ssec.wisc.edu/pub/eosdb/) does include an image of the entire Great Lakes produced by NOAA-20 data (at this ephemeral url). Again, the colorbar is a little too warm.

The data to create (using polar2grid) more customized imagery is available at the CIMSS ftp site mentioned above. ACSPO SSTs can be created using an EDR file (that is, the file 20250512073000-CSPP-L2P_GHRSST-SSTsubskin-VIIRS_N20-ACSPO_V2.80-v02.0-fv01.0.nc shown in this file list at https://ftp.ssec.wisc.edu/eosdb/j01/viirs/2025_05_12_132_0733/edr/. I then followed the directions in the polar2grid documentation here to produce a Lake Surface Temperature scaled from 273.15 K – 293.15 K. The commands I used are listed below. Note that ‘p2g_sst_palette.txt’ is pre-loaded within the polar2grid directories. The ‘rescale.yaml’ file is something I created following the documentation.

../p2g_grid_helper.sh greatlakes -83.5 45.1 750 -750 1800 1200 > GreatLakes.yaml
../polar2grid.sh -r acspo -w geotiff -p sst -g greatlakes --grid-configs ./GreatLakes.yaml --extra-config-path ./rescale.yaml -f /pathToVIIRS_SSTfile/20250512073000-CSPP-L2P_GHRSST-SSTsubskin-VIIRS_N20-ACSPO_V2.80-v02.0-fv01.0.nc
../add_colormap.sh ../../colormaps/p2g_sst_palette.txt noaa20_viirs_sst_20250512_073317_greatlakes.tif
../add_coastlines.sh --add-coastlines --coastlines-resolution f --add-colorbar --colorbar-height 42 --colorbar-text-size 24  --colorbar-min 0.0 --colorbar-max 20.0 noaa20_viirs_sst_20250512_073317_greatlakes.tif

The SST image created with the commands above has no value — is transparent — where there is no water. Let’s put the SSTs on top of Day Night Band imagery, and for that I needed SDR files that can be found at https://ftp.ssec.wisc.edu/eosdb/j01/viirs/2025_05_12_132_0733/sdr/ ; I downloaded all the SVDNB files (containing the data) and the GDNB0 files (containing georeferencing) to an otherwise empty directory, and created the the DNB imagery. Of the three varieties of Day Night Band imagery created (hncc, dynamic and adaptive), I decided for this day that adaptive looked the most acceptable.

../p2g_grid_helper.sh greatlakes -83.5 45.1 750 -750 1800 1200 > GreatLakes.yaml
../polar2grid.sh -r viirs_sdr -w geotiff -p hncc_dnb dynamic_dnb adaptive_dnb -g greatlakes --grid-configs ./GreatLakes.yaml -f /pathToFiles/DNB/*
../add_coastlines.sh --add-coastlines --coastlines-resolution f *dnb*.tif

I then used ImageMagick, shown below, to combine the two images.

convert -composite -gravity center noaa20_viirs_adaptive_dnb_20250512_073317_greatlakes.png noaa20_viirs_sst_20250512_073317_greatlakes.png NOAA20_VIIRS_DNB_ACSPO_SST_20250512_0733UTC.png

Western Lake Erie, as is typical, has the warmest waters — almost 60^o F! Saginaw and Green Bays also have relatively warm water. In contrast, Lake Superior and much of Lakes Huron and Ontario remain very cold — 40^o F or cooler.

View only this post Read Less

From the mailbag early on Friday 9 May 2025

The contrail show over the Lake MI region is pretty crazy this morning with the rotated flow aloft. Wondering if these are originating from a bunch of intercontinental flights that touch high latitudes in "great circle" fashion.

What did the mid-level water vapor imagery show? The 10-hour animation below from 0636 to 1631 UTC does indeed show multiple contrails, colder than their surroundings, rotating around an upper-level anticyclone over Iowa/Minnesota. (Click here for a view from the ground).

What other GOES-R bands/products allow one to view cirrus clouds? ABI’s Band 4 detects radiation at 1.38 µm, a radiation wavelength that is strongly absorbed by water vapor. Thus, in regions of water vapor, any reflective signal is strongly attenuated if there is water vapor above the reflective surface. Contrails high in the atmosphere typically are not overlain by much water vapor, so Band 4 will highlight these features during the daytime. The signal from any low-level cloud, however, is attenuated by water vapor in the atmosphere.

The Split Window Difference (SWD) fields, 10.3 µm – 12.3 µm, can also be used to detect the presence of cirrus clouds. Contrails show positive SWD values because of stronger absorption by water vapor of radiation with wavelengths of 12.3 µm than at 10.3 µm. The sub-pixel effect is also important. The satellite sensor at 12.3 µm is more sensitive to the colder part of the pixel in a partly cloudy scene that includes contrails that will not cover the entire satellite pixel. The 10.3 µm sensor will be a bit more sensitive to the warmer parts of the pixel than the 12.3 µm sensor. The difference field will therefore highlight contrails.

The Split Window Difference value shown above changes as the sun comes up and the boundary layer warms. Eventually, the gradient in Total Precipitable Water (TPW) that is evident in the animation below, also becomes apparent in the animation above. A strong inversion as was present on this day (here is the Quad Cities IA sounding, and here is the Chanhassen MN sounding). Once the inversion breaks, the SWD signal that describes moisture becomes more apparent. The hourly Land Surface Temperature estimates that might be used to detemined when the inversion is breaking are shown at the bottom.

This 2014 journal article by Lindsey et al. describes in more depth the evolution of SWD values as the boundary layer warms. Thanks to TJ Turnage, SOO in Grand Rapids, for the email!

View only this post Read Less

10-minute Full Disk scan GOES-19 (GOES-East) “Red” Visible (0.64 µm), Shortwave Infrared (3.9 µm) and “Clean” Infrared Window (10.3 µm) images (above) showed that a wildfire in central Saskatchewan produced a pyrocumulonimbus (pyroCb) cloud on 08 May 2025 (this was the first documented pyroCb of the 2025 North America wildfire season). The pyroCb exhibited cloud-top 10.3 µm infrared brightness temperatures (IRBTs) in the -40s C (denoted by shades of blue to cyan), a necessary condition to be classified as a pyroCb.

The coldest pyroCb cloud-top 10.3 µm IRBT was -47.7ºC — which was slightly colder than the air temperature of the Maximum Parcel Level (MPL) analyzed from rawinsonde data at The Pas, Manitoba (below).

GOES-19 Visible + Fire Mask and Infrared images (below) showed that the pyroCb developed as a cold front was passing through a small cluster of wildfires; surface air temperatures at nearby METAR sites were as warm as 86ºF, with wind gusts as high as 37 kts (43 mph).

A plot of surface observation data from Nipawin (station identifier CYBU) showed that after the cold frontal passage, smoke from the nearby pyroCb-producing wildfire reduced the surface visibility to 2-3miles at times (below).

View only this post Read Less

Due to a lack of radar coverage over American Samoa, WSO Pago Pago requested 1-minute Mesoscale Domain Sector coverage over the islands to monitor convective development and the potential for flash flooding. GOES-18 (GOES-West) Clean Infrared Window (10.3 µm) images (above) showed showers and thunderstorms that developed in the general vicinity of the American Samoa island of Tutuila (where Pago Pago International Airport NSTU is located) on 08 May 2025. The coldest cloud-top infrared brightness temperatures associated with some of these thunderstorms were in the -80 to -85ºC range (shades of violet to purple embedded within brighter white regions). GLM Flash Points indicated that there was only intermittent lightning activity with this convection (and the lightning occurred north of Tutuila — no thunderstorms or lightning was reported at Pago Pago).

A plot of rawinsonde data from Pago Pago at 0000 UTC (below) displayed a Total Precipitable Water (PW) value of 2.35″ — and a GOES-18 derived Total Precipitable Water value near Tutuila around that time was somewhat higher at 2.52″ (GOES-18 derived Total Precipitable Water values in the vicinity of the island were as high as 2.61″ at 0451 UTC).

The MIMIC Total Precipitable Water product (below) showed that American Samoa (centered at 14.3ºS, 170.7ºW) was situated along the southern edge of a broad east-to-west oriented band of high moisture just north of the islands.

_____________

East-southeast surface winds at Pago Pago during this time period were gusting as high as 36 kts or 41 mph (METAR list | Decoded surface reports) — and these winds were creating Significant Wave Heights of 15-16 feet just east and southeast of the island of Tutuila at 0931 UTC (above). Widespread moderate southeasterly winds across the Samoan Islands region were caused by the pressure gradient between a trough of low pressure just north of the islands (along which the more pronounced convective activity was focused) and high pressure to the south (mean sea level analyses: 0000 UTC | 0300 UTC | 0600 UTC | 1230 UTC | 1500 UTC | 1800 UTC | 2100 UTC) — and the broad areal coverage of these winds was depicted in OSCAT-3 scatterometer data at 1158 UTC (below).

The combination of these winds and periodic heavy rainfall was responsible for some flooding and wind damage across parts of Tutuila (Local Storm Reports).

---

GREMLIN (GOES Radar Estimation via Machine Learning to Inform NWP) fields use ABI and GLM data to estimate what radar reflectivity might look like. From 0700-1200 UTC on 8 May (below), GREMLIN showed heavier rains northwest of Samoa, but coverage is slowly increasing.

Between 1400 and 1700 UTC (below), rains have overspread American Samoa, although the bulk of the rains are to the north. (Note: GREMLIN fields are not parallax-corrected).

Between 1720 and 1920 UTC (below), rains continue, and they are widespread.

Widespread rains start to dissipate over Tutuila between 2100 and 2330 UTC, as shown below.

GREMLIN fields are available at the CIRA Slider.

View only this post Read Less