Cincinnati’s 25th running of the Flying Pig Marathon commenced on the morning of Sunday, May 7th, despite deteriorating weather conditions. After a round of storms at about 09:30 UTC, which subsequently passed through, race officials began the marathon as scheduled at 10:30 UTC. Figure 1 shows the composite reflectivity and LightningCast contours 0-30 minutes prior to the start time. The point marked “H” in the center of the animations below is the Cincinnati Municipal Airport — Lunken Field, which is about 3 miles east of downtown Cincinnati, where the marathon took place. In Figure 1, we see some storm cells intensifying and heading east toward downtown Cincinnati. The LightingCast probability of lighting in the next hour was > 50% at 10:30 UTC, when the race began.

ProbSevere LightningCast uses artificial intelligence and GOES-R ABI data to predict the probability of lightning in the next 60 minutes, as observed by the GOES-R Geostationary Lightning Mapper (GLM). Figures 2 and 3 show animations of LightningCast probabilities with GOES-16 10.3 µm brightness temperature and GLM flash-extent density (Figure 3). As the convection approaches downtown Cincinnati, we see cooling cloud tops and intensifying lightning flash rates. Furthermore, in Figure 3, we see a number of anvil flashes “overhead” of point H (i.e., within the GLM pixel containing point H) just prior to the start of the race. With thick ice from the anvil of the departing round of storms and the anvil of the approaching line of storms, LightningCast probabilities remained high despite no flash centroids within 10 miles of downtown Cincinnati during the inter-storm period (~10:15 to 11:00 UTC).

Using the LightningCast time series capability at Cincinnati Municipal Airport — Lunken Field (Figure 4), we see that the probability of lightning remained elevated (40-60%) in the inter-storm period, then jumped up to 71% by 10:36 UTC. The probability steadily increased to 96% at 10:51 UTC, 28 minutes before a shelter-in-place order was issued to runners. The “order” was more of a recommendation, according to race officials, as many runners were either confused by it or ignored it. The race began at 10:30 UTC, which was 15 minutes after the last flashes (within 10 miles) from the first round of storms.

Fortunately, there were no reported lightning-related injuries to racers or spectators, despite dozens of observed flashes every 5 minutes, within 5 miles of downtown Cincinnati. We aim to make LightningCast usable by forecasters and event managers to help make informed decisions such as postponements and sheltering at events with high vulnerability to lightning.

h/t to Kevin Thiel for providing information for this event.

View only this post Read Less

GOES-16 “Red” Visible (0.64 µm) and “Clean” Infrared Window (10.3 µm) images [click to play animated GIF | MP4]

1-minute Mesoscale Domain Sector GOES-16 (GOES-East) “Red” Visible (0.64 µm) and “Clean” Infrared Window (10.3 µm) images (above) showed an isolated supercell thunderstorm that produced a few tornadoes and hail as large as 4.00″ in diameter (SPC Storm Reports) across northern Missouri late in the day on 06 May 2023. Pulses of overshooting tops exhibited infrared brightness temperatures of -70ºC or colder (brighter shades of white).

1-minute GOES-16 Infrared images (below) include an overlay of GLM Flash Extent Density — one feature of note was the anvil lightning that eventually extended about 100 miles northeast of the thunderstorm core (which was producing a tornado and 4.00-inch diameter hail at 0042 UTC), stretching into southern Iowa.

GOES-16 “Clean” Infrared Window (10.3 µm) images, with an overlay of GLM Flash Extent Density [click to play animated GIF | MP4]

GOES-16 Visible images (below) include overlays of Total Precipitable Water, Lifted Index and Convective Available Potential Energy (CAPE) derived products — showing that the thunderstorm was moving into an environment of moisture and instability, helping to sustain its intensity.  

GOES-16 “Red” Visible (0.64 µm) images, with overlays of Total Precipitable Water, Lifted Index and Convective Available Potential Energy (CAPE) derived products [click to play animated GIF | MP4]

Farther to the west, thunderstorms produced hail and damaging winds across parts of north-central and southeast Nebraska, as seen in 1-minute GOES-16 Visible and Infrared images (below). One notable feature of interest was the cooler (lighter shades of gray) west-to-east oriented swath of hail on the ground in the wake of the southern Nebraska storm — even though that storm was initially producing primarily small-diameter hail (1.00 inch or less), that hail fell at a high enough rate to accumulate and remain on the ground for several hours. Another similar (but more subtle) southwest-to-northeast oriented swath of accumulating hail was evident in the wake of the storm over north-central Nebraska.

GOES-16 “Red” Visible (0.64 µm) and “Clean” Infrared Window (10.3 µm) images, with an overlay of GLM Flash Extent Density [click to play animated GIF | MP4]

A closer view of the southern Nebraska storm is shown below, using 1-minute GOES-16 Infrared images with a color enhancement tailored to highlight the colder swath of hail accumulation (brighter shades of cyan to darker shades of blue). Surface 10.3 µm infrared brightness temperatures within the narrow hail swath were in the 5-8ºC range, in contrast to 10-13ºC over adjacent bare ground. A few Local Storm Reports (2328 UTC | 0004 UTC | 0102 UTC | 0157 UTC) mentioned hail accumulation or a long duration of hail.

GOES-16 “Clean” Infrared Window (10.3 µm) images, with Local Storm Reports plotted in red [click to play animated GIF | MP4]

1-minute GOES-16 True Color RGB and Nighttime Microphysics RGB images from the CSPP GeoSphere site (below) provided another close-up view of the southern Nebraska storm — its hail swath showed up as pale shades white in the Nighttime Microphysics RGB imagery.

GOES-16 True Color RGB and Nighttime Microphysics RGB images [click to play MP4 animation]

In a toggle between the GOES-16 Land Surface Temperature (LST) derived product at 0000 and 0100 UTC (below), LST values within the narrow hail swath were in the low-middle 50s F (darker shades of blue), compared to the low-middle 60s F (shades of green) over adjacent bare ground. Another cold LST hail swath was apparent in the wake of the thunderstorm over north-central Nebraska.

GOES-16 Land Surface Temperature derived product at 0000 and 0100 UTC [click to enlarge]

Thanks to NWS Hastings for bringing this interesting feature to our attention!

View only this post Read Less

Joint Polar Satellite System (JPSS) data are now available online as part of the NOAA Open Data Dissemination (NODD) Program; at present, the data are available via Amazon Web Services (with future capabilities planned for Azure and Google). Global Suomi-NPP, NOAA-20 and NOAA-21 data (including Sensor Data Records (SDRs) and Environmental Data Records (EDRs) from ATMS, VIIRS and OMPS — and also geolocation files) are all available for free download. How can you create a display of the imagery, as shown above? The CSPP Software package Polar2Grid (v 3.0, available for download here) produced the image you see above.

After downloading Polar2Grid, you’ll need to determine the times of the data you wish to display; the data on the Cloud are saved off in roughly 90-second intervals. I consulted the SSEC Polar Orbit page (link) to find a (random) orbit from NOAA-21, where the orbit chosen is an ascending pass over Indonesia between 0320 and 0330 UTC on 5 May. The next step is to go to the Amazon Web Service data repository and find the I01 SDRs for the imagery to be created, and the geolocation information. Those webpages are shown below, with the files with data from 0320 to 0330 highlighted in the toggle. Note that the timestamps of the eight I01 and geolocation files are the same: 03:20:13.5, 03:21:37.7, 03:23:03.6, . . . 03:30:11.3. Download these files to a directory on the machine on which Polar2Grid is also installed. Note that these ‘GIMGO’ geolocation files are not terrain-corrected; the AWS site does include a directory (VIIRS-IMG-GEO-TC) that includes terrain-corrected geolocation files (‘GITCO’) that would be appropriate to use in regions of high terrain.

The Polar2Grid calls (remember that the environment variable $POLAR2GRID_HOME must be set) to create the imagery are straightforward:

$POLAR2GRID_HOME/bin/polar2grid.sh -r viirs_sdr -w geotiff -p i01 -f ./Cloud/*I01*.h5

(the ‘GIMGO’ files downloaded are also in that ./Cloud/ directory) reads the viirs_sdr I01 files and creates a geotiff image, and

$POLAR2GRID_HOME/bin/add_coastlines.sh --add-coastlines --add-grid --grid-D 10.0 10.0 --grid-d 10.0 10.0 --grid-text-size 20 noaa21_viirs_i01_20230505_032013_wgs84_fit.tif

adds coastlines and a latitude/longitude grid to the geotiff file created, and creates a png file (shown here at full resolution; VIIRS Image Files at native resolution are very large — this 10-minute one is 6500×7600 pixels! — because they’re at 375-m resolution; note also that the flag --coastlines-resolution f was added to the ./add_coastlines.sh call for the full-resolution image). The reduced-size image above was reduced in size and annotated using ImageMagick.

View only this post Read Less

Sentinel-1A overflew the Hawai’ian Islands near sunset on 4 May, as shown above (Sentinel-1A SAR Imagery is available online here and here). The toggle above compares the derived winds with GOES-18 Visible imagery (Band 2, 0.64 µm); the visible data enhancement has been changed (that is, brightened) from the default range of 0 to 130 to just 0 to 5 for this post-sunset scene. Winds approaching 30 knots (orange/red in the enhancement used) are common in the bands of wind between the islands. The regions of relatively calm winds (purple and blue in the enhancement used) in between the strong wind bands are where cloud bands exist, as shown both in the toggle above and the side-by-side image below.

The presence of the cloud bands suggests that surface convergence is occurring in between the bands of strong winds.

View only this post Read Less