2025 has proven to be an atypically wet year for American Samoa. The cumulative precipitation chart backs that up: going back to 1 January, the total rainfall at Pago Pago has been the highest on record up to this date, and total rainfall is 70% higher than a normal year.

July has been no exception to this trend. With nearly 9 inches of rain as of 8 July, the Pago Pago airport has already exceeded the normal value for the entire month. These trends are going to continue for quite some time, however. The CIMSS MIMIC product quantifies the total precipitable water present in the atmosphere, and it indicates an atmospheric river stretching from New Guinea to the Samoan archipelago and beyond into the southeastern Pacific.

MIMIC indicates TPW values of approximately 2 inches, which is confirmed by the radiosonde launches from Pago Pago. The calculated precipitable water is 49.22 mm, which is a little over 1.9 inches. While this is slightly reduced compared to earlier in the week (values as high as 2.2 inches were observed), this can still result in a substantial amount of rain. Given the significant amounts of rain that have already occurred, the ground is saturated and flash flooding is a concern. The lowest levels of the atmospheric profile are quite saturated, in fact, helping to bolster the idea of substantial rain.

Given the continued flow of moisture from the equatorial western Pacific as seen from MIMIC, the chances for rain in American Samoa remain significant as long as appropriate lifting mechanisms are present. The right amount of low-level convergence, for example, can be enough to force enough upward motion to initiate convective rain. Unfortunately, the most recent ASCAT overpasses just missed American Samoa, makinng it a challenge to directly observe low-level convergence over the islands.

View only this post Read Less

5-minute CONUS Sector GOES-19 (GOES-East) Infrared images displayed using RealEarth (above) showed a series of southward-moving thunderstorms that produced 1.91″ of rainfall near Ruidoso (and 3.29″ over the nearby South Fork Fire burn scar) on 08 July 2025. The resulting flash flooding was responsible for 3 fatalities, and damaged dozens of homes.

The same sequence of 5-minute GOES-19 Infrared images is shown below — including plots of the Flood Advisories and Flash Flood Warnings that were issued during that 5-hour period. A  Flood Watch was issued for the general area at 1805 UTC, and the first Flash Flood Warning for the immediate Ruidoso vicinity was issued at 2012 UTC (which was upgraded to a Flash Flood Emergency at 2047 UTC).

The hazard of heavy rainfall and flash flooding was compounded by enhanced runoff from the South Fork Fire burn scar (that wildfire occurred just west-northwest of Ruidoso about a year earlier, in June 2024) — a Landsat-8 “Natural Color” RGB image about 2 weeks prior this flash flooding event (below) displayed the large South Fork Fire burn scar (lighter shades of reddish-brown), as well as the smaller Salt Fire burn scar to the south-southeast.

With much of the rainfall occurring in a relatively short period of time (particularly over the runoff-prone South Fork Fire burn scar), the Rio Ruidoso at Hollywood rapidly rose to 20.24 feet. The Flash Flood Emergency that had been issued at 2047 UTC (2:47 PM local time) — which included Ruidoso, Hollywood, Ruidoso Downs and the Rio Ruidoso as far downstream as Glencoe — was still in effect when the Rio Ruidoso water level began to rise at Hollywood (below).

View only this post Read Less

Within 24 hours, a volcano on the Indonesian island of Flores erupted three times, with reports of an ash plume penetrating 18 km into the atmosphere. Mount Lewotobi is a twin volcano consisting of a pair of peaks, Perempuam and Laki-laki. It is this later peak that has been active lately, with numerous eruptions since November. On July 7 and 8, 2025, some of the most intense eruptions yet took place, with ash clouds being reported up to 18 km high and debris ejected up to 8 km away. Flights were cancelled out of Bali.

The first eruption was at 11:05 AM local time (0305 UTC), and produced the tallest ash plume. This was well-captured by the Advanced Himawari Imager, as it shows the brown-gray ash plume being transported west by the predominant easterly winds at that latitude.

The Ash RGB provides some interesting insights into the makeup of the plume. The plume is clearly multicolored, indicating that it is far from homogeneous. An animation of the plume is seen here:

To better analyze the makeup of the plume, here is a single still frame of the plume at 0420 UTC (12:20 PM local time). Here, it is possible to take a closer look at the components of the plume as observed by satellite. The Ash RGB recipe can be seen in the quick guide for this product. When this recipe is applied to the Laki-laki plume, notable features pop out. The eastern edge of the plume, closest to the volcano itself, is pink to red in color, indicating thinner quantities of ash. Further west, the plume is brown which is indicative of thicker ash. The bright green area represents a region of high SO2 concentration. At the western tip of the plume the predominant color is yellow; since yellow is a mixture of both red and blue, this represents a region of high concentrations of both ash and SO2.

Of course, volcanoes represent regions of extreme heat, and so they are well observed by satellite fire detection algorithms. Here, teh Fire RGB clearly indicates the eruption with a bright red spot on the eastern edge of the island.

Geostationary satellites serve a very important role for volcano monitoring as their consistent temporal resolution ensures a ready view of the environment in which the eruption is taking place. Sometimes, however, you get lucky with a polar orbiting satellite that passes overhead when something interesting is going on, allowing one to investigate the plume in higher spatial and/or spectral detail. Here is the view of the SO2 cloud as captured by TROPOMI aboard Sentinel-5P on 7 July. This overpass was at approximately 0550 UTC, nearly three hours after the initial eruption) and shows extraordinarily high SO2 concentrations with the highest regions in the center possibly obscured by the optically thick ash.

Note how this compares to the true color and Ash RGB views from the AHI for roughly the same time. The AHI views are plotted independently from the TROPOMI image, so the projection and scale are not identical. Still, it is clear that the different methods of viewing SO2 concentration each has some benefits. The TROPOMI view offers quantitative SO2 retrievals, and appears to identify regions in the center of the plume where SO2 is still high but largely absent from the RGB. By contrast, the large regions of green and yellow in the Ash RGB show that SO2 is present where TROPOMI’s quantitative retrievals are limited by the thick ash cloud.

This was just the first eruption of Laki-laki during this cycle. A second eruption took place at 1130 UTC, and can be seen in the Ash RGB animation as a pulse of pink (indicating ash) moving to the northwest beneath the high cirrus.

A third, less intense eruption occurred at 2153 UTC. Overall, 4000 people have been evacuated. Fortunately, no fatalities have been reported and injuries appear to be minor.

A major eruption of Lewotobi Laki-laki also occurred 3 weeks earlier, as discussed in this blog post.

View only this post Read Less

5-minute CONUS Sector GOES-19 (GOES-East) Infrared images displayed using RealEarth (above) showed the clusters of thunderstorms that produced up to 9.40″ of rainfall near Ingram (located between Hunt and Kerrville) in Kerr County, Texas during the 24-hour period ending at 1700 UTC (Noon local time) on 04 July 2025. The resulting flash flooding was responsible for at least 130 fatalities (including up to 96 in Kerr County, with 27 of those at the Camp Mystic summer camp located just northwest of Hunt) and over 850 high-water rescues. With rainfall rates as high as 3-4″ per hour at some upstream locations, the Guadalupe River at Comfort rapidly rose to 34.76 feet.

The same sequence of 5-minute GOES-19 Infrared images is shown below — but on a map of County outlines, and including plots of the numerous Flood Advisories, Flood Warnings and Flash Flood Warnings that were issued during that 18-hour period.

An animation of the MIMIC Total Precipitable Water product (below) revealed a plume of tropical moisture that was moving northward across south-central Texas during the 02-04 July period, helping to create an environment that was favorable for thunderstorm development (a longer animation beginning on 29 June indicated that this plume of tropical moisture was likely enhanced by the remnants of Tropical Storm Barry, after it made landfall in eastern Mexico late in the day on 29 June).

The Precipitable Water (PW) value derived from rawinsonde data at Del Rio, Texas at 0000 UTC on 04 July was 2.34″ — just 0.09″ below the climatological maximum of 2.43″ for all Del Rio soundings on 04 July at 0000 UTC (below).

View only this post Read Less