Light winds and clear skies with High Pressure over the western Great Lakes fostered the development of a warm(ish) patch of water over central Lake Michigan on 6 April 2025. (This has happened before!) Light winds means little vertical mixing in the surface layer of the lake, and solar insolation can heat the water. The warmest temperature diagnosed by the ACPSO algorithm was 43^oF, green in the enhancement used above.

GOES-16 Lake Surface Temperatures, below, show the temporal evolution of the warm eddy, with warmest temperatures between 1700 and 1900 UTC. A mid-lake temperature sensor warms 3 degrees during the day. In addition, lake breezes develop over coastal Wisconsin, Illinois, Indiana and Michigan, just what one might expect in a light wind regime as the land becomes warmer relative to the still-cold lakes.

A timely Metop-C overpass on 6 April produced the ASCAT wind plot shown below (from this source). Very light winds were present over southern Lake Michigan.

View only this post Read Less

Brandon Aydlett, the Science and Operations Officer (SOO) at the NWS forecast office in Guam (where the National Weather Service day begins) relays the following message:

Yesterday we received a call from a local radio noting something suspicious in IR imagery over the Philippines. Of course, it looked eerily like a volcanic eruption, of course, in a volcanically-active area

What steps might one take see if something is/is not a volcano? Consider the image below that the radio personality saw, from 1820 UTC on 2 April 2025. What is that isolated feature over the island of Samar in the central Philippines? If you saw that very cold cloud top, how might you decide if it’s a thunderstorm vs. a volcano?

---

NOAA-20 overflew the region shortly after 1700 UTC on 2 April, and a toggle between the Day Night Band imagery and the I05 infrared (11.45) imagery, below (from the NASA Worldview site), shows a very faint cloud signature (the waxing crescent moon was providing no illumination!), but a faint circulation can (just barely) be discerned in the Day Night band imagery.

MIMIC Total Precipitable Water fields, below, from 0000 UTC 1 April through 0000 UTC 3 April show a concentrated area of moisture moving towards/over Samar, arriving at about 1200 UTC on 2 April.

ASCAT winds from Metop-B, shown below (source), also show a circulation moving towards the island of Samar.

Himawari-9 provides animations (and multispectral observations) that might help differentiate between a thunderstorm and a volcano. The animations (clean window infrared and Ash RGB) below are from 1000 to 2000 UTC; at 2000 UTC the cold cloud tops are starting to warm. Structures in the warmest clouds in the Band 13 imagery show arcs that are consistent with the circulation detected by Metop-B ASCAT. The Ash RGB lacks the pink hues associated with volcanic ash (Ash RGB Quick Guide is here). So far, a lot of satellite indicators — from both geostationary and polar orbiters — argue against a volcanic event.

It’s not unheard of that tropical convection and volcanic eruptions coincide in this part of the Philippines. (Here’s the ash cloud from Pinatubo in 1991 mingling with the clouds of Typhoon Yunya, from this blog post; in Pinatubo’s case, the daytime imagery allowed for an easy visual distinction between the darker gray ash cloud and the brighter white cumulonimbus. This event on 2 April was at night.) However, neither the Tokyo VAAC nor Seismic monitoring nor imagery at the CIMSS VOLCAT site suggest any kind of volcanism.

Answer: This is a thunderstorm!

---

I am indebted to Brandon Aydlett, SOO at WFO GUM, for alerting me to this interesting case. He also sent along this pdf outlining what happened when he got the request. You’ll note that the blog post includes much of what he said.

View only this post Read Less

A large severe weather outbreak took place across the south central United States on Wednesday 2 April 2025. The NWS Storm Prediction Center (SPC) anticipated a significant weather event in the region, having predicted an enhanced risk of severe weather as early as Monday 31 March and a rare high risk the night before the event. The environmental conditions ahead of storm formation largely behaved as predicted, with elevated dew points caused by significant moisture advection from the Gulf coast and an advancing cold front stretching from the Midwest to the lower Mississippi valley creating the necessary instability for deep moist convection. When combined with ample shear and jet-streak based ageostrophic lifting, the ingredients were all present for multiple long-lived tornadoes. This can be seen in the surface map, where there’s substantial northerly flow from the gulf, dew points in the Mississippi valley between the mid-60s F and 70 F, and a dominant cold front.

It was a busy day for NWS forecasters throughout much of the continental United States (CONUS). Well over 700 tornado, severe thunderstorm, and flash flood warnings were issued across much of the eastern half of CONUS. According to The Weather Channel, this day marked the 3rd highest total of NWS-issued warnings ever, trailing only the April 2011 Super Outbreak and the Memorial Day Weekend 2004 Outbreak. The SPC Storm Reports map for the day shows just how widespread this event was.

Of course, NOAA satellite infrastructure was well-situated to capture the build-up and execution of this outbreak, and provided a vast amount of valuable data to forecasters, emergency manager, mass media, and the general public. Here are a few highlights:

NUCAPS soundings may prove to be an increasingly valuable part of the global observation network due to ongoing and anticipated further reductions in the US radiosonde network. By capitalizing on both American and EUMETSAT polar orbiting hyperspectral sounders, NUCAPS can provide vertical thermodynamic profiles multiple times per day. Here is an animation of three NUCAPS soundings from near Memphis, Tennessee.

Two of these soundings are overnight while the third is in the early afternoon. The daytime heating is clearly evident, and there is also substantial moistening in the lower levels. Midlevel lapse rates remain basically unchanged between these three soundings, but the influx of near-surface sensible and latent heat results in substantial instability increases. The Microwave + infrared retrieval shows CAPE going from 0 J/kg all the way to over 2200 J/kg.

Of course, the trustworthy GOES-16 geostationary satellite also provided excellent views of the evolving atmosphere. The mesoscale sector was focused on the area and providing very finely detailed views of the storms.

This animation of 1 minute GOES-16 data from Band 2 (0.64 microns) does a wonderful job of showing the evolution of these clouds as the cold front approaches southern Illinois and the lower Mississippi River. Since this animation is near sunset and this is a visible wavelength channel, the light is fading. However, the low solar angle really helps the texture pop, highlighting numerous overshooting tops throughout the domain.

At the same time, the Band 13 imagery allows for different characteristics of the storm to be identified. A string of enhanced v structures, indicative of very strong updrafts, is readily apparent. Wavelike structures over Illinois and Ohio indicate significant turbulence in the anvils downstream of the convective cores. Large storms over central Tennessee likely initiated from cold pools advancing ahead of the cold front and tapping all of that moist, unstable air.

Numerous tornadoes were reported on this day, including several rated at EF-2 or greater. One of the most devastating tornadoes of this event hit the small down of Selmer, Tennessee in the overnight hours. The tornado, with a preliminary rating of EF-3, destroyed entire neighborhoods in this small town of 4500 people. The CIMSS ProbSevere Product was able to capture and correctly identify the approaching storm cell as a storm of concern well before it hit the town.

Selmer is located in the upper right of this animation loop, at the intersection of US Routes 64 and 45. Note how ProbSevere has encircled the cell that eventually hits Selmer 90 minutes ahead of its arrival there. A full half-hour before its arrival, ProbSevere identified its tornado probability at 87%. With fast moving storms and nocturnal storms common, tools like ProbSevere are critical for keeping people safe, all the way from Mississippi to the Québec/Ontario border.

View only this post Read Less

On a rare High Risk Convective Outlook day, a widespread outbreak of severe thunderstorms produced tornadoes, large hail and damaging winds (SPC Storm Reports) across a large swath of the central and eastern US during the afternoon and subsequent nighttime hours on 02 April 2025 — along and ahead of a cold front and pre-frontal squall line (surface analyses). As mentioned in this blog post, GOES-19 had arrived at the GOES-East position (over the Equator at 75.2ºW Longitude) and had begun broadcasting data — which presented an opportunity for a GOES-16 (the current operational GOES-East) vs. GOES-19 comparison during this severe weather outbreak.

There was some overlap of Mesoscale Domain Sector 2 from both satellites (below) — the 1-minute GOES images included plots of time-matched (+/- 3 minutes) SPC Storm Reports (T=Tornado, TEF2=Tornado rated at EF2 damage, H275=Hail of 2.75″ in diameter, W=Wind damage, W90=Wind gust of 90 mph).

Closer views of a few areas that received notable severe weather are shown below:

Central Illinois Tornadoes/Hail/Damaging Winds

GOES-16 vs. GOES-19 Visible images (above) and Infrared images (below) showed convection that produced an EF2-rated tornado, hail as large as 2.00″ in diameter and wind gusts as high as 90 mph in central Illinois. In the Visible imagery, small-scale features such as thunderstorm overshooting tops were resolved equally well by both satellites; in the Infrared images, cloud-top infrared brightness temperatures of cold features such as overshooting tops were generally in very good agreement (separated by 0.1 K or less).

Missouri/Arkansas/Kentucky Tornadoes

GOES-16 and GOES-19 Infrared images (above) showed thunderstorms that produced several tornadoes across southeast Missouri and southwest Kentucky, including an EF3-rated tornado in southeast Missouri around 2013 UTC and a fatal EF2-rated tornado in southeast Missouri around 2332 UTC. These storms also produced hail to 2.75″ in diameter and wind gusts to 70 mph.

Indiana Tornadoes/Damaging Winds

GOES-16 and GOES-19 Infrared images (above) displayed thunderstorms that produced EF1-rated tornadoes and wind gusts to 100 mph in parts of Indiana.

Tennessee/Mississippi Tornadoes

GOES-16 and GOES-19 Infrared images (above) depicted thunderstorms that produced tornadoes across northern Mississippi and southwest Tennessee — including an EF3-rated tornado that impacted Selmer, Tennessee around 0542 UTC, and a fatal tornado near La Grange, Tennessee around 0631 UTC. Well-defined Enhanced-V storm top signatures were associated with some of these tornado-producing thunderstorms (numerous examples of Enhanced-V storm top signatures have been shown on this blog).

View only this post Read Less