Severe storms developed across the central high plains during the period of 9-11 May 2023 in association with an upper trough digging across the southwest US. Per NWS BOU on May 9: “RAP Mesoanalysis and water vapor satellite show the center of this upper trough just off central California coast this afternoon. Broad southwesterly flow aloft is present ahead of this trough axis, and the flow should strengthen and slowly back to the south-southwest through tomorrow afternoon. This trough should continue to progress eastward, eventually transitioning to a more negatively tilted trough axis by Wednesday evening.” Associated satellite iamgery with RAP 500 mb height is shown in Fig 1, during the period of early May 9 to mid May 10.

Two rounds of severe thunderstorms developed across the northern CO I-25 corridor on the 9th, resulting in large/damaging hail: the first during the late afternoon, and the second overnight. A single procedure that captures these storms in detail is a VIS/IR sandwich that transitions to IR only after sunset (Fig 2). During the day, the sandwich combo clearly reveals rapid storm strengthening per cooling in the IR and increase in texture in the VIS, including the development of OTs and AACPs. Overnight, the loss of texture with the loss of VIS makes storm top analysis slightly more difficult, but OTs and probable AACPs could still be diagnosed from the IR-Window imagery in this case.

VIIRS DNB/NCC Imagery can be leveraged alongside the GOES IR imagery for more detailed cloud analysis at night since it provides visible-like imagery at night. Four VIIRS scans (from NOAA-20, NOAA-21, and S-NPPx2) capture the evolution of thunderstorms and low clouds between 0800 UTC and 0941 UTC (Fig 3). The imagery shows a very active storm top with OTs crossing from Colorado into western Nebraska, with persistent lightning flashes. In the wake of the storm, low clouds are observed expanding across the urban I-25 corridor, which is denoted by the bright city lights.

The low clouds were tracked in space and time using GOES RGB imagery, including Nighttime Microphysics RGB at night and Day Cloud Phase Distinction RGB during the day (Fig 4). The low clouds quickly filled in across the region in the wake of the large thunderstorm complex amidst moist/upslope easterly flow. Low clouds persisted through the morning the following day.

From NWS Cheyenne prior to sunrise on 10 May: “Once these early morning storms have run their course, expect low stratus and perhaps even some patchy fog to develop along the Interstate-80 corridor from Cheyenne to Sidney, and as far north as Scottsbluff and Alliance. The HREF shows moderate to high probabilities of this occurring and model soundings do show a saturated profile just above the surface at Cheyenne and Sidney. Surface and satellite observations back this up with both showing an impressive low stratus deck over most of eastern Colorado. Low stratus looks to break up by late morning/early afternoon, which could have implications on convection in regards to when initiation begins, or (less likely) if it is limited by CIN and a capping inversion.”

After sunrise during the mid-morning hours, NWS BOU analysis included: “Quick update to the grids to account for the incredibly persistent low stratus deck this morning. Day Cloud Phase satellite imagery shows a blanket of low stratus across the plains and I-25 corridor. There are signs of this eroding up in Logan County, but expect this to stay socked into the urban corridor for at least another hour or two before it gradually scatters.”

The upper trough continued east into the Four-Corners area on the 10th, and forced another round of severe thunderstorms across the high plains. Another Sandwich product, this time using Geocolor instead of VIS, depicts the evolution of severe thunderstorms across the central high plains on the 10th (Fig 5).

Further south, thunderstorms developed along a dryline during the afternoon and into he evening. The GOES-East Split Window Difference with a simple grayscale colormap (and IR-Window overlay for cold BTs) receptively highlights the location of the dryline throughout the period, with brighter grays representing greater moisture, primarily in the low-levels (Fig 6). During the afternoon, clouds and deep convection developed along dryline bulges in southwest Texas and in Mexico to the south. Additional storms develop further north along and ahead of the dryline near the NM/TX border. After dark, interpretation of the SWD changes, but a signature of the dryline is observed accelerating east, and additional convection develops along it in southwest Texas near Midland. See this graphic for annotation of the scene.

The upper low continued to drift northeast across eastern Colorado on the 11th into the 12th. Abundant moisture pumping into the region along with forcing associated with the low resulted in a prolonged period of rainfall across the Colorado front range, resulting in 1-day precip totals of 1-4 inches+ from Pueblo to Fort Collins. From NWS PUB on the 11th: “Current water vapor imagery has upper low spinning across eastern Crowley and western Kiowa Counties, with deformation band showers
and storms moving south and west across the southeast mtns and portions of the southern I-25 Corridor with the precipitation associated with mid level trowel and wraparound precipitation across northeastern Colorado into the Front Range at this time.”

And from NWS BOU early on the 12th: “GOES-18 satellite imagery displays an upper level low feeding additional moisture into our region. The center of this low sits along the Colorado/Kansas border and is expected to shift northeast gradually this morning. With the persistence of this system bringing ample amounts of moisture, a Flood Watch remains until 12 PM MDT this afternoon.

The evolution of the low from the 10th through the 12th, including multiple rounds of strong thunderstorms and the persistent feed of deep moisture into the front range, all discussed above, is shown in GOES-East Water Vapor Imagery and GLM FED overlay (Fig 7).

Bill Line, NESDIS/STAR

A severe thunderstorm traversing almost the full length of Nebraska from west to east just south of I-80 produced landspout tornados and numerous instances of hail > 1″ in diameter. The evolution of the storm during the afternoon into the overnight hours is captured well in a VIS/IR combo product, which transitions from the VIS/IR sandwich during the day to IR at night (Fig 1). This procedure is simple to make in AWIPS (feel free to reach out if you’d like it!) Apparent in the imagery is a persistent above anvil cirrus plume, indicative of a particularly threatening storm.

A keen eye will detect a trail of high reflectance (lighter gray compared to surrounding) in the VIS, and cold BT (lighter gray compared to surrounding) in the IR Window. This signature in satellite imagery is that of a accumulating hail, or a hail swath. NWS Hastings NE, was on the case, and shared GOES-East IR Window channel imagery of the storm and associated hail swath, with the colormap range adjusted to better highlight the cool BT signature at the surface (see below).

Here's something we don't see every day:

Tonight's storm was such a prolific hail producer that the lingering hail on the ground can be seen on IR satellite! This is due to the temperature differences between the warm/dry ground and cooler hail-covered ground#scienceiscool 🤓 pic.twitter.com/5WxQT4smzZ

— NWS Hastings (@NWSHastings) May 7, 2023

Alternatively, one can load two instances of ch13 IR (or Ch13 and Ch14), and adjust the lower layer as a grayscale with the range focusing around that of the surface, and the upper layer as a colored enhancement to highlight the clouds. This AWIPS procedure, shown in Fig 2, captures the hail swath while also maintaining the ability to diagnose storm top features.

Even better, we can use RGBs to really isolate the hail swath feature. During the day, the Day Cloud Phase Distinction RGB does the trick, isolating the hail swath as green (similar to snow cover; Fig 3)). Since the time period is toward sunset, we can make simple adjustments to the recipe to draw out the feature (green max to 20, blue max to 12).

After sunset, one can switch to the Nighttime Microphysics RGB, which captures the hail swath as a red magenta (~pink) compared to the blue magenta (~violet) of the surrounding clear sky surface (Fig 4).

Combining the two animations (in this instance, outside of AWIPS), we get a seamless day/night transition between the two RGBs over the hail swath (Fig 5). Such a transition product is available in real-time on the STAR Image Viewer.

During the rest of the overnight, the hail swath slowly faded in IR imagery, becoming difficult to discern in ABI (Fig 6).

Viewing 375 m VIIRS IR window imagery (band I5), the location of the remaining hail swath becomes more apparent again given the higher spatial resolution (Fig 7). The scene is captured in four consecutive passes between 0716 UTC and 0901 UTC (two from S-NPP and one from NOAA-20 and NOAA-21 each). The hail swath is noticeable in the first VIIRS image, is masked by clouds in the next two, and reappears in the final image. VIIRS imagery is available to view on Polar SLIDER, which will soon include a CONUS sector.

See previous posts on this blog for more about hail swath detection using satellites.

Bill Line, NESDIS/STAR

Severe thunderstorms developing across central Texas on 26 April 2023 resulted in very large/damaging hail, including up to 4.5″ diameter near Waco. GOES Water Vapor Imagery captured the mid/upper level feature that aided in the development of thunderstorms (Fig 1). During the morning of the 26th, WV analysis from SPC included: “The primary mid/upper-level feature influencing this forecast is a compact cyclone — apparent in moisture-channel imagery over the southern Rockies and centered near SKX…” and from NWS Norman: “Water vapor imagery shows a closed upper low slowly translating southeastward over northeast New Mexico early this morning… expect activity to expand this morning as the upper low pivots eastward into the panhandles and lift increases over the area.” The animation in Fig 1 also included NWP 500 mb height, which aids in the depiction of the mid-level feature, 500 mb wind speed (shades of blue) to show the location of the mid-level flow in relation to the low, and surface Theta-e, to capture the moisture return ahead of the low. As the low drifts southeast during the period, thunderstorms develop just ahead of the vorticity maxima as the strongest mid-level flow overlays the increasing low-level moisture.

Forecast offices studied the GOES imagery during the late morning and early afternoon in there mesoscale analysis. The following NWS interpretations go along with the Day Cloud Phase Distinction RGB animation in Fig 2, which ends at 1831 UTC. As has been shown in numerous case studies on this blog, the DCPD RGB can be thought of as an enhanced version of visible imagery as it combines the high resolution detail of the VIS with phase information from the NIR and cloud-top-temperature information from the LWIR. The imagery shows the increasingly vertical growth of the cu field, especially near the boundaries. Throughout the period, the unstable region, with respect to surface-based convection, is obvious as denoted by the cumulus field south of the w-e boundary near Abilene, east of San Angelo, and well west of Waco. A stable environment is apparent over the eastern part of the scene, as denoted by widespread stratus and stable wave clouds, or billows. Elevated/mature convection is apparent in the northeast part of the scene. By the end of the period, glaciation is being observed in the tallest cu (color changing from cyan to green) within the unstable region near and southeast of Abilene, with numerous orphan anvils, indicating imminent CI. The imagery is annotated in Fig 2a.

From NWS FWD: “Visible imagery is depicting partial clearing is ongoing with temperatures approaching 80s degrees.” This location is in the western part of the FWD CWA.

From NWS SJT: “Visible satellite showing better cu bubbling up around and south of the outflow boundary and appears that convection will be developing before 3 PM. Again, roughly we are talking about an area from Abilene south to Ballinger and east towards Coleman, Brownwood, and Clyde line for initial development, with development spreading south into the afternoon.”

And from SPC: “Visible-satellite imagery shows a swelling cumulus field south of an outflow boundary and east of a dryline… Water-vapor imagery shows the leading edge of cirrus and implied ascent spreading quickly east across the Permian Basin and into the Big Country. As a result, convective initiation is expected in the next hour or so near Abilene.” The SPC mesoscale discussion was accompanied by a quick analysis of the visible imagery with relevant boundaries highlighted (Fig 3).

During the two hour period following the end of Fig 2, convection would initiate (vertical growth and and transition from green to yellow/red) and quickly strengthen into strong/severe thunderstorms (Fig 4).

At this time, switching to a product like the GOES VIS/IR Sandwich Imagery allows one to monitor storm top features and their trends as a proxy for updraft strength and potential storm severity, as a supplement to radar analysis. For example, many of these storms exhibited persistent and large Overshooting Tops and Above Anvil Cirrus Plumes, the latter which has been associated with the strongest thunderstorms. Other notable features include enhanced-V’s (or cold-U), and thermal-couplets. These features are annotated on one of the storms (near Waco) in Fig 6. Storms that are weakening exhibit a collapse of the OT and warming of the cloud top brightness temperatures, in addition to loss of texture in the VIS.

Bill Line, NESDIS/STAR

The Shiveluch Volcano on the Russian Peninsula of Kamchatka underwent a significant eruption during the overnight hours early on 10 April 2023. The CIMSS Satellite Blog discussed the eruption of a nearby Volcano a few days earlier. The JPSS constellation of satellites (S-NPP, NOAA-20, NOAA-21) provided overnight imagery of the eruption on the 10th shortly after it began. In particular, the VIIRS Day Night Band (DNB) Near Constant Contrast (NCC) product captured nighttime visible-like imagery of ash and clouds associated with the eruption, in addition to the light emitted from the eruption when not masked by clouds/ash (Fig 1). The volcano was included well within five consecutive VIIRS passes combined from the three satellites (25-min or 50-min apart).

The cloud shield associated with the eruption was captured in Himawari-9 infrared imagery from the overnight period into the next day (Fig 2). IR brightness temperatures to as cold as -60C were observed.

Visible imagery from Himiwari-9 from the next day shows several bursts of covnection above the volcano just after sunrise (Fig 3). The low reflectance of the ash cloud is apparent in the imagery compared to surrounding cloud cover.

The GOES-West satellite provided a unique, oblique view of the eruption during the morning. Visible imagery around sunrise shows convective bursts over the volcanic eruption (Fig 4).

Combining multiple IR channels into a single multispectral RGB imagery product, named the SO2 RGB, allows one to diagnose and track the SO2 plume associated with the eruption. In this case, viewing the Himawari-9 SO2 RGB, the SO2 plume (yellow) can be tracked from the point of the eruption on the 10th, all the way to the Aleutian Islands of Alaska two days later (Fig 5).

Bill Line, NESDIS/STAR