Deadly fires on the island of Maui occurred in an environment favorable to fire development: dry and windy. What kind of satellite data could be used to assess those conditions? Sentinel-1A overflew the western Hawai’ian islands near sunset on 8 August 2023 (that is, at 0440 UTC on 9 August), as shown above. Widespread 30-knot winds (yellow in the enhancement used) are indicated. The island of Maui was just missed by this scan. The SAR Winds are shown below in a toggle with GOES-18 Visible imagery. A smoke plume that extends west of Maui is faintly apparent. Low clouds are present east and north of the Hawai’ian islands, but are absent to the south and west. The motion of these low clouds can be tracked to infer winds.

GOES-18 Derived Motion Wind vectors, shown below, hourly from 0201-0701 on 9 August, show widespread values of 30-35 knots to the east and north of the Hawai’ian island chain. The features tracked to infer the winds were from near-surface to 900 mb.

Metop-C overflew the Hawai’ian islands shortly after 0800 UTC on 9 August, and Advanced Scatterometer (ASCAT) winds from that pass are shown below, overlain on top of GOES-18 Band 7 (Shortwave Infrared, 3.9 µm) imagery that shows the signatures of the two fires (yellow and black) on Maui. Winds of 25-30 knots are widespread near the Hawai’ian islands, with somewhat weaker winds to the north and east. Stronger winds — gales — are detected in the Alenuihaha channel between Hawai’i and Maui.

Metop-B also sampled this region, but a bit earlier than Metop-C. Wind plots for both satellites (taken from this site) are shown below. Winds are strongest surrounding the Hawai’ian island chain.

Three different wind data sources above highlight the strong environmental winds surrounding the Hawai’ian island chain. How dry was the environment? The MIMIC Total Precipitable Water animation below, shows significant drying occurring between 0000 UTC 7 August and 0000 UTC 9 August as the moisture associated with Hurricane Dora moves south of the islands.

The percent-of-normal TPW values surrounding Hawai’i (source) around 0000 UTC on 9 August,shown below, were 40-50%.

The 0000 UTC Sounding from Hilo (here, from this source), shows a computed TPW of 1.36″. This SPC site shows that value to be dryer than normal.

NOAA-20 NUCAPS data can be used to assess moisture at various levels. NOAA-20 overflies Hawai’i at around 0000 and 1200 UTC daily, and the toggle below shows gridded NUCAPS estimates of relative humidity in the 925-700 mb layer at 2338 UTC on 8 August and 1205 UTC on 9 August. Sounding availability points are also shown, and the NUCAPS retrievals were converging (green points) over most of the domain shown.

At 2338 UTC on 08 August, a successful retrieval is indicated just west of Maui; at 1205 UTC on 09 August, a successful retrieval is apparent near the northern shore of Maui. The two NUCAPS soundings at those points are shown below.

Gridded NUCAPS data are also available online for Hawaii at this website. The animation below toggles through relative humidity estimates at 850, 700, 500 and 300 mb. Dry air is indicated.

---

Many satellite products can help a forecaster understand the environment when dangerous weather is expected. This snippet from a Forecast Discussion from late on 3 August shows that Fire Weather was anticipated to occur over the Hawai’ian Islands long before the fires occurred.

View only this post Read Less

GOES-18 (GOES-West) Day Land Cloud Fire RGB, Shortwave Infrared (3.9 µm), “Clean” Infrared Window (10.3 µm) and “Red” Visible (0.64 µm) images with an overlay of the Fire Power derived product — Fire Power is a component of the GOES Fire Detection and Characterization Algorithm FDCA — (above) showed signatures associated with a wildfire (located about 20 miles WNW of Mayo, Yukon CYMA) that produced a pyrocumulonimbus (pyroCb) cloud late in the day on 06 August 2023.

At 2130 UTC, the wildfire exhibited a maximum 3.9 µm infrared brightness temperature of 106C, with a Fire Power value of 5257 MW (below).

The pyroCb cloud exhibited a cloud-top 10.3 µm infrared brightness temperature as cold as -48.15C at 2350 UTC (below).

View only this post Read Less

GOES-18 (GOES-West) Ash RGB images (above) showed the northeastward drift of volcanic clouds produced by an eruption of Mount Shishaldin that began just before 1300 UTC on 04 August 2023 (a Mesoscale Domain Sector was positioned over that region at 1603 UTC, providing 1-minute imagery after that time). Those northeast-moving volcanic clouds contained moderate concentrations of ash (denoted by shades of pink in the Ash RGB images) — then after 1700 UTC, SO2 RGB images revealed the formation of a southeast-moving volcanic cloud that contained modest concentrations of SO2 (shades of yellow) that drifted just to the south of False Pass (in the SO2 RGB images, the northeast-moving ash clouds exhibited darker shades blue). High clouds began to overspread the area from the west after 1830 UTC, which tended to mask the volcanic signatures.

1-minute GOES-18 True Color RGB images from the CSPP GeoSphere site (below) helped to highlight the ash-rich volcanic cloud (shades of tan to brown) moving northeast from the summit of Shishaldin, and also showed the higher-altitude volcanic cloud drifting southeast (which contained SO2).

A plot of 0000 UTC rawinsonde data from Cold Bay, Alaska (below) indicated that northwesterly winds existed at altitudes of 28000 ft (9 km) and higher, with southwesterly winds below that level (down to altitudes around 20000 ft or 6 km).

Radiometrically-retrieved GOES-18 Ash Cloud Height from the NOAA/CIMSS Volcanic Cloud Monitoring site (below) showed that maximum height values were generally in the 6-8 km (20000-26000 ft) range.

Given that volcanic ash presents a significant hazard to aviation, Volcanic Ash Advisories and Forecasts were issued (below).

GOES-18 Ash RGB images with a few Pilot Reports that mentioned the altitude of volcanic ash are shown below.

View only this post Read Less

Hourly estimates of Total Precipitable Water (TPW) from MIMIC, shown above in an mp4 animation (click here for an animated gif), show the Intertropical Convergence Zone (ITCZ, as measured by TPW) over the western Pacific evolving from a linear feature at the start of the animation to one with a significant break near 140^oE – 150^oE longitude. This has occurred as Typhoons Doksuri (moving between the Philippines and Taiwan at the start of the animation) and Khanun (moving south of Japan at the end of the animation) have developed (here is a blog post on the formation of Khanun). Formation of tropical cyclones during the breakdown of the ITCZ is not uncommon (as noted, for example, here, and here). The mp4 animation below (click here for an animated gif) tracks the Upper- and Lower-level water vapor infrared imagery (Himawari-9 Bands 8 and 10 at 6.24 µm and 7.3 µm, respectively) over the same time period.

Himawari-9 Clean Window imagery, below, at 0000 UTC on 24 July (left) and 2 August (right) show little in the way of ITCZ convection between the Equator and 20^oN on 2 August, especially compared to 24 July, over the western Pacific. ITCZ convection does persist to the east of 160^oE longitude. MIMIC TPW fields at those two times (here) also show a similar rearrangement of convective centers.

View only this post Read Less