The possibility of heavy rain over the Samoan islands led the Pacific Region to request 1-minute imagery over American Samoa, imagery that ended at 2107 UTC on 18 December when the domain was moved to cover the southwestern US (where a significant fire risk was occurring on 18 December). GREMLIN observations from 13-19 UTC on 18 December, shown below with Total Precipitable Water (TPW), suggest the heaviest rains were over the Manu’a Islands to the east of Tutuila Island. Note the region of slightly drier air — TPW < 2″ — (orange/rust in the enhancement) south of the Samoan Islands, where TPWs are closer to 2.3″ (magenta in the enhancement).

Day Cloud Phase Distinction fields as sun rose over Samoa on 18 December show convection developing over the main Samoan Islands as active convection continues to the east. (Click here for the Day Cloud Phase Distinction overlain by GREMLIN).

One of the ABI fields used in the Machine Learning algorithm that estimates radar echoes is Band 9 (Mid-level water vapor at 6.95 µm). The animations below show Band 9 and also Band 9 overlain with GREMLIN.

---

What kind of winds accompanied this convection? ASCAT observations can give a hint, as the observations below from Metop-C show. However, large regions are unsampled.

Derived Motion Winds calculated from GOES-18 imagery (Bands 2, 7, 8, 9, 10, 14) can give wind information, and values are shown below. Low-level winds (violet and dark blue in the imagery below) are from the northwest whereas upper-level winds (shades or red) are from the southwest, so there is considerable shear over the Islands (Here is a still image from 1940 UTC). There is a noticeable increase in the number of derived wind vectors as the sun rises and visible imagery becomes available! Low level winds are not moving the dry air to the southwest of Samoa over the islands.

Microwave estimates of TPW, below, taken from the MIMIC website, also show Samoa deep within the moisture of the South Pacific Convergence Zone. Heavy rain chances will likely continue there for the next few days.

GREMLIN fields for GOES-West are also available here on the CIRA SLIDER. Finally, maybe you’re wondering why, if the Mesoscale sector was over Samoa, didn’t I show 1-minute imagery!? I matched the time increment to that of the GREMLIN product, which is a full-disk field, produced every 10 minutes.

View only this post Read Less

This short tutorial will explain how Polar2Grid can show individual isobaric levels, and also lapse rates (dT/dp) between two isobaric levels.

Polar2Grid is CSPP software designed to process files either from the NOAA NODD (EDR files for NOAA-20 are here; EDR files for NOAA-21 are here) or from Direct Broadcast antenna data streams. Polar2grid creates reprojected imagery. In this example, I decided to work with data over the western United States. The NOAA-20 orbits on 18 December (here, from this website), show a descending pass over the Rockies, from the USA-Canada border at 0938 UTC to Los Angeles at about 0943 UTC. The data for this were downloaded from the AWS Cloud for NOAA-20, and a list of the files is shown below. Times for the granules listed start at 09:37:22.9 and end at 09:43:44.7.

(base) [scottl@machine NUCAPSData]$ ls -1
NUCAPS-EDR_v3r2_j01_s202412180937229_e202412180937527_c202412181021100.nc
NUCAPS-EDR_v3r2_j01_s202412180937549_e202412180938247_c202412181022030.nc
NUCAPS-EDR_v3r2_j01_s202412180938269_e202412180938567_c202412181021170.nc
NUCAPS-EDR_v3r2_j01_s202412180938589_e202412180939287_c202412181022310.nc
NUCAPS-EDR_v3r2_j01_s202412180939309_e202412180940007_c202412181021180.nc
NUCAPS-EDR_v3r2_j01_s202412180940029_e202412180940327_c202412181021580.nc
NUCAPS-EDR_v3r2_j01_s202412180940349_e202412180941047_c202412181023480.nc
NUCAPS-EDR_v3r2_j01_s202412180941069_e202412180941367_c202412181023040.nc
NUCAPS-EDR_v3r2_j01_s202412180941389_e202412180942087_c202412181023430.nc
NUCAPS-EDR_v3r2_j01_s202412180942109_e202412180942407_c202412181023470.nc
NUCAPS-EDR_v3r2_j01_s202412180942429_e202412180943127_c202412181023120.nc
NUCAPS-EDR_v3r2_j01_s202412180943149_e202412180943447_c202412181023480.nc
(base) [scottl@machine NUCAPSData]$ 

The variables that can be displayed using polar2grid and those files is quite long. If you were to download the files, and run the following command: ./polar2grid.sh -r nucaps -w geotiff --list-products-all -f /path_to_files/NUCAPS*.nc , you would see the many possibilities. For this blog post, I’m choosing Temperature_802mb and Temperature_596mb.

By default, polar2grid will rescale the plotted temperatures based on the range it finds. I want to have the same scale used for both levels here, and to do that, I have to tell polar2grid. This is done by adding a cris.yaml file into the directory polar2grid_v_3_1/etc/polar2grid/enhancements/, which directory is created as part of the polar2grid install process. The contents of the cris.yaml file that I created are below. I’ve mandated that both 802-mb temperatures and 596-mb temperatures be scaled from 240 to 300 K.

I also want to display these data on a map of my choosing (rather than the native NOAA-20 orbital map). Polar2grid allows a user to define a map with the p2g_grid_helper.sh shell script below; the script expects a grid name (CANucaps), a center longitude and latitude (115 W, 35 N), grid spacings in m (4000/-4000) and grid sizes (960×720). The output here is put into a yaml file.

./p2g_grid_helper.sh CANucaps -115.0 35.0 4000 -4000 960 720 > CANucaps.yaml

Next, I call polar2grid to make imagery, color-enhance it, and add coastlines and a colorbar, shown below.

../polar2grid.sh -r nucaps -w geotiff -p Temperature_802mb Temperature_596mb -g CANucaps --grid-configs ./CANucaps.yaml -f /path_to_NUCAPSData/*
./add_colormap.sh ../../colormaps/p2g_sst_palette.txt j01_atms-cris_Temperature_802mb_20241218_093722_CANucaps.tif
./add_colormap.sh ../../colormaps/p2g_sst_palette.txt j01_atms-cris_Temperature_596mb_20241218_093722_CANucaps.tif
./add_coastlines.sh --add-coastlines --add-colorbar --colorbar-text-size 16 j01_atms-cris_Temperature_802mb_20241218_093722_CANucaps.tif
./add_coastlines.sh --add-coastlines --add-colorbar --colorbar-text-size 16 j01_atms-cris_Temperature_596mb_20241218_093722_CANucaps.tif

The resultant imagery is shown in a toggle below. I didn’t label the imagery, but I hope you can tell which field is 802mb, and which is 596mb! The colorbar automatically scales to the values specified in the cris.yaml file.

---

Next, I decided I wanted to compute and plot the temperature difference between 802 and 596 mb. That requires additions to two files within the polar2grid distribution. A cris.yaml file is needed in (polar2grid_v_3_1/etc/polar2grid/composites/), with contents as shown below. I defined the parameter (lapse802_596) as using the DifferenceCompositor that requires two fields. Then, in the enhancements directory (polar2grid_v_3_1/etc/polar2grid/enhancements/) I scaled the field to be between 4 and 25K.

If I ran polar2grid now, and pointed to the NUCAPS data directory, and asked it to –list-products-all — I would see a new possibility! lapse802_596. The three polar2grid calls below,

./polar2grid.sh -r nucaps -w geotiff -p lapse802_596 -g CANucaps –grid-config ./CANucaps.yaml -f /path_to_NUCAPSData/*

./add_colormap.sh ../../colormaps/p2g_sst_palette.txt j01_atms-cris_lapse802_596_20241218_093722_CANucaps.tif

./add_coastlines.sh –add-coastlines –add-colorbar –colorbar-text-size 16 j01_atms-cris_lapse802_596_20241218_093722_CANucaps.tif

The result of those three calls is shown below. The steepest lapse rates are off the coast of California and in northwestern Mexico.

---

Note: An earlier version of this blog post had a different cris.yaml file in it; in that one, none of the named temperatures (that were mispelled!!) included a name: identifier/match criterion. As a result, all cris variables (including derived ones!) had the same stretched values in computed color bars.

Thanks to Dave Hoese, CIMSS, for helping me figure this out!

View only this post Read Less

10-minute Full Disk scan GOES-16 (GOES-East) daytime True Color RGB + Nighttime Microphysics RGB images from the CSPP GeoSphere site (above) depicted the cyclonic (clockwise in the Southern Hemisphere) circulation associated with the formation of Subtropical Storm Biguá just off the coast of Rio Grande do Sul state in far southern Brazil during the 14th-16th December 2024 period.

12-hourly surface analyses shown below (source) indicated that this subtropical storm developed as an area of low pressure (denoted by a red “B”) moved southeast across Rio Grande do Sul on 14th December, becoming classified as Subtropical Storm (“Tempestade Subtropical”) Biguá shortly after it moved offshore at 0000 UTC on 15th December. 24 hours later, Biguá was downgraded to a Subtropical Depression (“Depressao Subtropical”) at 0000 UTC on 16th December.

A sequence of ASCAT surface scatterometer winds (source) from Metop-B and Metop-C (below) showed that the compact low-level circulation center of Subtropical Storm Biguá remained just off the Brazilian coast on 15th December. The strongest winds were generally within the southern semicircle of the storm.

View only this post Read Less

EUMETSAT Meteosat-9 Infrared Window (10.8 µm) images (above) showed Category 4 Cyclone Chido as its eye moved across the small island of Mayotte (airport identifier FMCZ) in the Mozambique Channel around 0730 UTC on 14th December 2024 — and went on to make landfall just south of Penba, Mozambique (airport identifier MQPB) around 0400 UTC on 15th December. Chido traversed increasingly warmer sea surface temperatures (source) as it approached Mozambique.

As Cyclone Chido passed over Mayotte, the airport reported wind gusts of 92 kts (106 mph) as the eye approached and 91 kts (105 mph) as the eye departed (below).

Shortly before Chido made landfall in Mozambique, a Synthetic Aperture Radar (SAR) image at 0253 UTC (below) indicated that a derived maximum wind speed of 123.84 knots was present in the SE quadrant of the eyewall (source).

View only this post Read Less