1-minute Mesoscale Domain Sector GOES-16 (GOES-East) “Red” Visible (0.64 µm) images (above) showed a supercell thunderstorm that produced a tornado, hail as large as 5.00 inches in diameter and wind gusts to 100 mph (SPC Storm Reports | NWS Aberdeen summary) in central South Dakota during the late afternoon and early evening hours on 18 July 2023. Several pulses of overshooting tops were evident in the visible imagery, an indicator of vigorous thunderstorm updrafts. It bears mention that 10 minutes prior to the 100 mph wind gust (at 0115 UTC), there were reports of semi trucks overturned by strong winds along Interstate 90 (in eastern Lyman County at 0105 UTC). The peak wind gust at nearby Chamberlain (K9V9) was 53 knots (61 mph) at 0122 UTC.

In the corresponding 1-minute GOES-16 “Clean” Infrared Window (10.3 µm) images (below), the coldest thunderstorm overshooting tops exhibited infrared brightness temperatures around -70ºC (brighter white pixels embedded within dark black regions).

A closer view of Infrared imagery at 2347 UTC (below) revealed that a report of 4.50″ diameter hail occurred within a cluster of 2 other reports of 2.75″ diameter hail just southwest of Pierre (KPIR) — in fact, with further evidence the size of that largest hail was later upgraded to 5.00 inches (which was a record hail size for Hughes County in South Dakota: media report).

A cursor sample of GOES-16 infrared (10.3 µm) brightness temperature, Cloud Top Temperature and Cloud Top Height derived products at 0021 UTC (above) displayed values of -72.61ºC, -77.64ºC and 47031.21 feet, respectively — these Cloud Top Temperature and Cloud Top Height values were similar to those of a Most Unstable (MU) air parcel’s Maximum Parcel Level (MPL), calculated using 0000 UTC rawinsonde data (source) from Aberdeen SD (below).

1-minute GOES-16 Visible images with an overlay of GLM Flash Extent Density (below) showed modest lightning activity with this thunderstorm, with a number of minor lightning jumps during that time period.

In a larger-scale view of GOES-16 Infrared and Visible images (below), the signature of an Above-Anvil Cirrus Plume (reference | VISIT training) was evident in the Visible imagery (but was more subtle in the Infrared imagery).

The supercell thunderstorm developed within a corridor of instability and moisture that existed in the warm sector between a cold front and a warm front (surface analyses) — as seen in GOES-16 Convective Available Potential Energy (CAPE), Lifted Index and Total Precipitable Water derived products (below). CAPE values were around 1900 J/kg, Lifted Index values were around -7ºC and Total Precipitable Water values were around 1.6 inches.

View only this post Read Less

Suomi NPP overflew Tropical Storm Calvin to the east of Hawai’i at around 1136 UTC on 18 July. Data downloaded from the NODD, and processed by Polar2Grid (version 3.0), show the storm in very low light conditions, above. Calvin is moving rapidly westward towards Hawai’i, and interests in that state — especially those on the Big Island — should closely monitor the storm’s progress. For more information refer to the Central Pacific Hurricane Center and the National Weather Service Forecast office in Honolulu.

Editor’s note: Part of the impetus for this post was to see just how promptly an image could be uploaded from the NODD. The Suomi NPP data showed up at about 1315 UTC, and the image above was published at 1342 UTC, about two hours after the satellite overflew the storm.

---

Day Night Band imagery can often be used to view the center of a sheared storm overnight. Is that the case in the low-light situation (New Moon was on 17 July!) above? The animation below, from the CSPP Geo site, with views from every three hours, shows a center displaced from the main convection before sunset (0301 UTC on 17 September) and after sunrise (1801 UTC on 18 September). It’s not difficult then to use the Day Night band image above — even with the very low contrast — to infer a center location, as noted.

__________

The Suomi-NPP VIIRS Day/Night Band image valid at 1138 UTC — which later became available in AWIPS — is shown above, with/without an overlay of the 1200 UTC Tropical Surface Analysis. As previously noted, Calvin’s exposed low-level circulation center can faintly be seen just south of the deep convection (even in such low-light conditions, with the Moon in its Waxing Crescent phase just 1 day after a New Moon).  

 

View only this post Read Less

GOES-18 data above show persistent convection around Samoa and American Samoa between 0400 and 0730 UTC. Sentinel-1A overflew the region just after 0550 UTC, yielding information about wind features. At that time, a convective storm was developing in between Tutuila and Upolu. The animation below steps through the Band 13 (Clean window infrared, 10.3 µm) imagery, with a toggle that includes the SAR winds.

---

SAR detected strong winds under the convection — up to about 30 knots. The toggle below compares winds and Normalized Radar Cross Section fields taken from this website. Past SAR blog posts at this site have suggested that SAR winds are diagnosed too high in regions where significant ice is present in a convective cloud – because of reflection of the SAR Radar signal off that ice back to the satellite. A recent paper (link) suggests that the enhanced backscatter arises not from the ice, but from scattering from wobbling, non-spherical, oblate hydrometeors within the melting layer. That is, the cloud ice is necessary, but the signal arises only when the ice enters the melting layer.

---

Wave observations from the Aunuu buoy (available at this website), below, show waveheights decreasing during the time of the convective development above (which was, after all, far removed from the buoy site).

View only this post Read Less

The presence of three JPSS satellites — Suomi NPP, NOAA-20 and NOAA-21 — allows for animations at high resolution (I01 imagery on VIIRS has 375-m resolution) at even low latitudes. The animation over the island of Guam (at 13.4^oN) shows the evolution of clouds between 0246 and 0358 UTC on 17 July 2023. These data were downloaded from the NOAA NODD and processed with Polar2Grid v 3.0 as described here. This requires downloading granules from the VIIRS-I1-SDR directory and geolocation data from the VIIRS-IMG-GEO directory at the NODD websites. Whether or not good coverage is available (and when it happens!) can be shown a day to three in advance at the SSEC Polar Orbit Tracking website, which includes graphics over the western Pacific for Suomi-NPP, NOAA-20 and NOAA-21. Knowing when the data will be available means a timely image can be made. Note that the resolution of the images above has been reduced to 1 km!

---

Geostationary satellites also show imagery over the same region. Himawari-9 visible (band 3, 0.64 µm) imagery, below, has been processed using Geo2Grid software operating on Himawari-9 HSD files (that, like JPSS data files, are available online). The animation below shows Band 3 imagery over the same domain as above, from 0250-0400 at 10-minute intervals. Whether or not a user needs high spatial resolution (far better with JPSS data, especially in the infrared) or better temporal resolution is a question best answered by the phenomena that are being investigated. If there were a small fire on Guam, for example, JPSS data with a 375-km resolution at 3.74 µm is more likely to detect it than Himawari-9 Band 7 imagery (3.8 µm) with 2-km nadir resolution.

View only this post Read Less