5-minute CONUS Sector GOES-19 (GOES-East) GeoColor RGB images with an overlay of Next Generation Fire System (NGFS) Fire Detection polygons (above) showed the dense smoke plumes and thermal signatures associated with the ongoing Nopiming Provincial Park wildfire on 22 May 2025 (this wildfire exhibited extreme behavior and also produced numerous pyrocumulonimbus clouds on 13 May).

GOES-19 Visible images with an overlay of the FDCA Fire Mask derived product (below) also displayed the wildfire smoke plumes and thermal anomalies.

10-minute Full Disk scan GOES-19 True Color RGB and Aerosol Optical Depth (AOD) derived product images from the CSPP GeoSphere site (below) depicted AOD values as high as 1.0 (dark red) within the dense smoke plumes. Smoke from the previous day of Nopiming wildfire activity was also seen drifting across western Manitoba and crossing the Manitoba/Saskatchewan border.

Toggles between VIIRS True Color RGB and False Color RGB images from Suomi-NPP and NOAA-20 (as visualized using RealEarth) are shown below. Active fires along both the western and eastern periphery of the large Nopiming Fire burn scar showed up as brighter shades of pink in the False Color RGB images — and although none of the individual fires were particularly large, because of the dry fuels being being burned the resulting smoke plumes were quite dense.

As the smoke drifted southwestward across the US/Canada border a few hours after sunset, a plot of surface observation data from Cavalier (K2C8) in far northeast North Dakota (below) showed that the surface visibility at that site was reduced to 6 miles with haze at 0435 UTC and 0515 UTC on 23 May (the base of the smoke layer at those 2 times was 300 ft above ground level).

A NOAA-20 VIIRS Day/Night Band image valid at 0912 UTC (4:12 AM local time) on 23 May revealed the bright nighttime glow of fires that continued to burn overnight along/near the edges of the darker-gray Nopiming Fire burn scar. Faint smoke plumes (light shades of gray) could be seen drifting W-NW away from several of the active fires.

View only this post Read Less

Iceland enjoyed a remarkable string of clement weather in mid-May. The true-color animation below, taken from the NASA Worldview site, shows almost a fortnight of clear skies!

VIIRS gives a high-resolution view of the surface and cloud features as shown above. JPSS satellites carry sounder instruments used to create NUCAPS data; NUCAPS profiles allow a meteorologist to view what’s going on within the troposphere. Both VIIRS and NUCAPS fields can be viewed using CSPP‘s polar2grid software using data downloaded from the cloud (here, for example, from Amazon Web Services). For the purposes of this blog post, I downloaded M03/M04/M05 SDR files (0.488 µm/0.555 µm/0.672 µm, i.e., blue, green, and red) and also I01 (0.64 µm) so the image could be sharpened, and the GITCO/GMTCO VIIRS files that contain georeferencing for I- and M-bands, respectively. (VIIRS Bands are defined here.) The data were downloaded for a daytime NOAA-20 overpass on 19 May 2025 that viewed Iceland between 1332 and 1336 UTC 19 May 2025. I also downloaded NUCAPS EDR files from the same AWS site. Then I ran the following commands in polar2grid (I’m using version 3.1) in a directory I created just under the bin directory in the polar2grid code distribution:

../p2g_grid_helper.sh Iceland -19.0 65.0 1500 -1500 1080 840 > Iceland.yaml
../polar2grid.sh -r nucaps -w geotiff -p Temperature_853mb -g Iceland --grid-configs ./Iceland.yaml -f /path/to/storedfiles/Iceland/NUCAP*
../polar2grid.sh -r viirs_sdr -w geotiff -p true_color_raw -g Iceland --grid-configs ./Iceland.yaml -f /path/to/storedfiles/Iceland/SV*
../add_colormap.sh ../../colormaps/p2g_sst_palette.txt j01_atms-cris_Temperature_853mb_20250519_133218_Iceland.tif       
../add_coastlines.sh --add-coastlines --add-colorbar --colorbar-text-size 18 --colorbar-height 40 j01_atms-cris_Temperature_853mb_20250519_133218_Iceland.tif       
../add_coastlines.sh --add-coastlines noaa20_viirs_true_color_raw_20250519_133136_Iceland.tif

The first command creates a grid (‘Iceland’) and stored grid parameters in a file (‘Iceland.yaml’). Next, polar2grid calls using first the nucaps reader and then the viirs_sdr reader, to create imagery of 853 mb Temperature, and of ‘raw’ True-color imagery on that pre-defined ‘Iceland’ grid. I added a predefined colormap (‘add_colormap.sh’) to the 853-mb data, and then a final command to draw coastlines and add a colorbar (for the temperature field). The annotation was added with ImageMagick commands. The toggle below shows the true-color imagery and the 853-mb temperatures.

The warmest temperatures at 853 mb are just offshore of Greenland. That suggests to this blogger that there is wind downsloping off the Greenland ice sheet and warming dry adiabatically. That plume of warm air then moved out over the N. Atlantic to the north of Iceland. The coolest 853-mb temperatures might be linked to the glaciers of Iceland. The swirl patterns in the ocean ice off the coast of Greenland are also interesting to view.

View only this post Read Less

Night Microphysics RGB imagery over the Samoan Islands, above, saved from the CSPP Geosphere site, shows the characteristic color changes in the Night Microphysics RGB imagery from pink/salmon to dark red as lines of cumulus grow to thunderstorms. GREMLIN fields from the CIRA Slider, below, show the satellite-estimated MRMS fields. Rain starts developing southeast of American Samoa, getting fairly intense after 1500 UTC.

The shower development occurred within a moist airmass, as diagnosed by the MIMIC Total Precipitable Water field shown below at 1400 UTC on 21 May 2025. (Source) The Samoan Island chain was near the southern edge of the moisture associated with the South Pacific Convergence Zone; to date in May, Pago Pago has received for than 18″ of rain, an amount that almost places May 2025 in the top ten of wet Mays at that station (the wettest May, with 29.1″, was 1999).

What kind of products might help a forecaster anticipate the convective development? GFS forecasts for the Galvez-Davison Index (source), below, show values consistently around 30, suggesting scattered showers and maybe a thunderstorm. (Values tomorrow there are forecast to be substantially higher; check back to see if it’s raining!)

---

For nearcasting/nowcasting convective development, consider the animation below. GOES-18 Derived Lifted Index values show the region (Values around -2) on the shoulder of an unstable area to the north, where Lifted Indices are close to -5. Low-level derived motion vectors from GOES-18 (in dark blue) show confluence over the Samoan Islands; winds are primarily easterly to the north of the islands, and more southeasterly to the south of the islands. Low-level winds are conducive to convective development and the airmass is moist. Upper-level winds are also plotted (in red) ; but an obvious divergent signal is lacking.

Legacy Profiles from GOES-18 augment model information with the temperature and moisture information within ABI’s 16 channels. The (clear-sky) locations at 1300 UTC are shown below, and a comparison between a profile in the deep moisture (where convection is not occurring) — at 11.4^oS, 170^oW — and a profile at the southwest edge of the domain shown — at 16^oS, 173^oW — shows the differences in moisture/stability (especially the level of the LFC) in the two regions.

LightningCast Probability fields (available here) give the probability of when a GLM observation of lightning is likely in the next 60 minutes. When the geographic distribution of the probability fields starts to change, that can mean something. The animation below stretches from 1330-1510 UTC on 21 May 2025. The probability contours expand south of American Samoa at 1450 UTC. This suggests development of some kind is happening there. By 1510 UTC, obvious convection is developing there!

For more information on Pago Pago’s weather, refer to the National Weather Service.

View only this post Read Less

A shortwave impulse spawned numerous supercell thunderstorms in a highly unstable and well-sheared environment in the Southeast U.S. ProbSevere v3 readily tracked numerous severe and tornadic storms across the region.

One storm in particular produced tornadoes in Madison, Alabama, and prompted a tornado emergency warning. In this example, the ProbTor v3 was much higher than ProbTor v2, due to some unrepresentative environmental data in the RAP (used in ProbTor v2). The 52% value is very high for ProbTor, which uses MRMS, GOES, Earth Networks, and HRRR inputs.

Looking at the time series of ProbSevere v2 vs. v3 (Figure 3), we notice a couple things:

1. ProbSevere v3 was higher early in the storm’s lifetime, which has been a consistent finding. The ML models of v3 are more skillful when combining the environmental and observational parameters in nascent storms.
2. ProbTor v3 does a much better job maintaining elevated probabilities for the majority of the storm’s lifetime. As implied previously, we believe this is due to more accurate environmental data in the HRRR model.

The long-lived supercell exhibited several lightning jumps. Shortly before the first tornado report at 23:20 UTC, there was a very large lightning jump (see the yellow bars in the image below). However, the large increase in flash rate was partially due to a storm merger, and not a meteorological intensification in updraft strength.

I say partially, because the change in maximum GLM flash-extent density also increased, and before the storm merger at 23:10 UTC. The GLM flash-extent density (yellow line in Figure 5) shows increases around 23:00 UTC. This field is not affected by a storm-object’s boundaries. Therefore, it seems that intensification was already occurring in the tornadic storm when the storm merger happened. The merger did seem to either further intensify the flash rate in our supercell or at least did not impede the intensification that was already occurring.

Lastly, the meteogram below shows values for some satellite-based storm attributes. The satellite growth rate (blue) is an important predictor for growing storms, and can help boost probabilities before radar predictors are nearing severe thresholds.

Once a storm is nearing or at maturity, the “intense convection probability” from IntenseStormNet provides a fused ABI+GLM-based perspective on the “intensity” of the storm. In this storm, the probability was over 95% for much its lifetime. The probability jumped up 8 minutes prior to the first local storm report at 20:52 UTC.

These “intense” probabilities can also be plotted on a 2D view (Figure 7). This product works well for many developing and mature supercells, and could be useful on its own in lieu of radar data. Even with good radar coverage, we’ve found that it skillfully contributes to ProbSevere v3 models.

View only this post Read Less