During the early morning hours of 14 March 2022, a plume of moisture from the Gulf of Mexico was advected northward prior to a severe weather setup later that day. Along with surface observations and RAP surface analysis data, imagery from the GOES-16 Nighttime Microphysics RGB provided conformation of this moisture advection with stratus clouds developing across eastern Texas and southern Oklahoma (green-yellow) from Figure 1. Strong contributions in the red and green bands signify thick clouds that mostly contain water, helping to determine that these are low level stratus clouds driven by the synoptic scale advection of low level moisture across the region.
The NWS Storm Prediction Center issued a Slight Risk for northeast Texas and the Ark-La-Tex region, with all hazards (tornadoes, large hail, and damaging winds) possible (see slideshow below). Use of the Nighttime Microphysics RGB in this scenario may provide conformation of the moisture advection, along with its current spatial extent in regions where few surface observations exist. Monitoring the extent of these stratus clouds also provides a ‘first look’ at which areas will receive more or less solar heating during the morning, which may impact the initiation time, coverage, and maximum strength of convection later in the day.
A broad trough and embedded shortwaves digging east across the southern US brought severe weather, including tornadoes, to parts of the south and southeast US on 21-22 March 2022. The evolution of the trough across the country is shown well in 6.2 um GOES-East Water Vapor Imagery in Figure 1. GLM FED is also included in the animation as progressively more opaque yellow atop the green cold clouds.
Adding RAP sfc and upper level analysis fields onto the water help one to better understand features in the imagery by associating them with familiar fields, such as 500 mb height and wind speed, and sfc pressure (Figure 2).
Partially overlapped GOES-East 1-min mesoscale sectors resulted in a corridor of 30-second imagery across central Texas. The difference between 30-second imagery and 1-min imagery may not sound like a lot, put processes that are occurring on such small timescales do appear notably smoother to the human eye in the side-by-side comparison. This is exemplified in an animation of visible imagery over a tornado-producing severe thunderstorm on the border of 30-sec and 1-min imagery (Fig 3).
Further south near San Antonio, the evolution of the cu field leading up to eventual convective initiation is captured in 30-second Day Cloud Phase Distinction imagery. The 2.5 hour animation (300 images) reveals a cumulus field becoming increasingly agitated with the growth of cumulus clouds into towering cu, eventual glaciation with the colors changing from blue/cyan to green, to convective initiation diagnosed by vertical growth and transition of colors from green to yellow/red (Fig 4). The growth of the eventual first severe-warned storm occurs under high cirrus (red), but can be followed in the very high temporal resolution imagery.
Post convective initiation, 30-second imagery of storm maturation captured the evolution of storm top features, such as overshooting tops and an above anvil cirrus plume, in much detail (Fig 5). The imagery is extremely fluid, and ensures forecasters are receiving updates about the storm faster than ever. The animation is 240 images, or 2 hours long.
During the same period of 30-second imagery, adding a semitransparent 10.3 um overlay, resulting in the VIS/IR Sandwich, helps to capture the storm top features a little better by including the quantitative brightness temperature information (Fig 6).
After viewing the 30-sec animation a few times, take a peak at the 1-min animation of the same scene. It is fascinating how 1-min imagery appears relatively “choppy” (Fig 7)!
The most impressive storm developing near San Antonio exhibited an exposed updraft in the GOES-East 30-second imagery. Rocking a 30-sec visible imagery animation of this storm over a 25 minute period reveals counterclockwise rotation of the updraft (Fig 8). Extending west of the updraft is the more horizontally oriented flanking line. Inflow feeder clouds are also analyzed southeast of the updraft, as well as an overshooting top and above anvil cirrus plume at the storm top. Given the presence of high clouds, and rapid evolution, some of these features can easily be missed in coarser temporal resolution imagery.
As was shown in Figure 1+2, the convective threat shifted east to the southeast on the 22nd as the broad upper trough shifted east and another shortwave and associated strong jet streak spread across the region. Two strong thunderstorms passed through the New Orleans area just after sundown, and were captured in GOES-East 1-min ABI and GLM imagery. The northern storm produced an EF1 tornado, while the thunderstorm produced an EF3 tornado. The 1-min IR imagery revealed rapidly cooling cloud tops just prior to the initial tornado reports between 0024 UTC and 0029 UTC (Fig 9).
GLM FED associated with the storms included rapid upticks in total lightning activity leading up to tornado development (Fig 10). The FED data highlights the location and movement of the most intense storm updrafts, as well as the presence of lightning flashes and resulting lightning danger well removed of the storm core.
Finally, combining the GLM Flash Extent Density with Minimum Flash Area into a single RGB reveals where an abundance of small flashes (indicative of strengthening updraft) were occurring, as bright shades of yellow (Fig 11). Red represents low FED and small flashes, so a transition from Red to Orange to yellow in this RGB indicates increasing numbers of small flashes. Shades of blue represent long flashes, which are often present with the anvil regions of the thunderstorms, as well as with decaying updrafts.
A significant shortwave trough digging across the Four Corners brought strong winds to a large swath of the southern High Plains on 17 Mar 2022, resulting in widespread blowing dust and critical fire weather conditions. GOES-East Water Vapor Imagery combined with RAP Analysis fields allows one to connect features observed in the imagery with those present in SFC/UL charts (Fig 1). In this case, a surface low is analyzed deepening out ahead (east) of the main mid-level cyclonic circulation associated with the mid-level trough. A mid-level (70+ knot) jet streak rounds the base of the trough and is associated with a meridional temperature gradient in the water vapor imagery. The warming in the water vapor imagery here also represents a region of drying/descending air. Deep vertical mixing during the day across the dry/warm region helped to mix down the increasing mid-level winds, resulting in strong surface wind gusts.
As is typically shown, basic split window difference imagery captures lofted dust quite well, and is represented here as darkest shades of gray (Fig 2). The area of lofted dust is coincident with the downward momentum transport and drying observed in water vapor imagery, circling the mid-level circulation.
The SWD is combined with IR and SWIR imagery to yield a Dust-Fire RGB, which shows the blowing dust as bright green, wildfire hot spots as red, and clouds as varying shades of blue (Fig 3).
DEBRA Dust imagery results from a more advanced algorithm that highlights suspected dust, based on the IR techniques, onto visible imagery as shades of yellow (Fig 4). DEBRA Dust is available on CIRA Slider.
Two, GOES-East mesoscale sectors of 1-min imagery each were positioned over the region to capture the widespread wildfire threat. A 3-hr long 1-min animation of Geocolor imagery combined with a hot SWIR range provides a nice depiction of the blowing dust and wildfire development (Fig 5). The Geocolor imagery from GOES-East, especially later in the afternoon (increasing forward scattering), depicts aerosols quite well, including blowing dust as shades of brown/tan, and wildfire smoke as gray. By including only hot (40+C) shortwave IR pixels as an overlay, we are able to capture wildfire hot spots as well (yellow). Click the following 1-min and 30-sec animations for better resolution views. Geocolor is available on CIRA Slider, STAR Image Viewer, and coming soon to every AWIPS.
Because the two GOES-East mesoscale sectors overlapped slightly, 30-second imagery was available (in the overlap region). This partially overlapped meso sector 30-sec imagery can be accessed in AWIPS with some simple configuration additions. Thirty-second Geocolor imagery of early upper-level cloud cover evolution is shown in in Fig 6, while a wildfire hot spot, smoke plume, and blowing dust are shown in Fig 7.
Strong southerly winds within a dry antecedent atmosphere a few days later on 20 March resulted in another day of critical fire weather conditions and blowing dust. A Geocolor + Fire Power derived product + SWD Dust Enhancement AWIPS display captures the evolution of the wildfire hot spots and smoke plumes across Texas, and broad plume of blowing dust across west Texas (Fig 8). By employing the Fire Power product in this type of display, as opposed to the SWIR imagery, one needs not worry about setting thresholds for hot spot detection.
An early March shortwave trough brought a wide range of active weather to much of the central United States on 04-05 March 2022, including snowfall, blowing dust, low clouds and fog, wildfires, and severe thunderstorms. GOES Satellite imagery was leveraged by forecasters in operations to detect and track the varying hazards, some of which are documented below.
Water vapor imagery is analyzed by forecasters to help gain an understanding of the recent and current observed synoptic scale setup in the atmosphere. How are these features influencing our weather now, and how might they influence our weather in the future? Analyzing water vapor imagery from the evening of the 4th through the morning of the 5th, key features to note include a broad toughing over the western US while a ridge builds to the east (Fig. 1). Within the broad trough, a shortwave lifts northeast across the Four Corners, and another closed circulation dips south across N California. A mid/ul jet max is analyzed in the imagery rounding the base of the Four Corners trough, denoted by the eastward spread of drying (warming BTs).
NWS forecast discussions focus on details relevant to their local weather, and how features in WV imagery influence their forecast decision-making.
From PUB at 2226 UTC on the 4th: “Current water vapor imagery is indicating deep southwest flow aloft across the state as a strong jet core is rounding the base of an upper low spinning across the southern Great Basin. Clouds and isolated showers/virga associated with an embedded wave this morning across the plains have cleared with associated lee troughing on the plains leading to gusty south to southwest winds developing across the plains this afternoon, with a few areas hitting red flag criteria attm. Further west, satellite imagery and regional radars indicating cloud top cooling and scattered showers and isolated thunderstorms moving across eastern Utah and into western Colorado attm.”
And from OUN at 0915 UTC on the 5th: “… but we did go a little drier given upstream observations and significant drying noted on the lower layer water vapor imagery as ejecting mid-level jet spreads northeastward from southern New Mexico.”
And from Topeka at 1000 UTC on the 5th: “As of 3 AM, water vapor imagery notes a negatively-tilted upper trough axis over the Colorado Rockies and is continuing to eject northeastward across the central plains. An induced lee cyclone has continued to deepen in southeast Colorado and is beginning to move northeast across westcentral Kansas and soon towards southcentral Nebraska… The lee cyclone will continue to move northeast across northcentral Kansas and into southeast Nebraska throughout the day today bringing a variety of weather concerns….”
During the morning of the 5th, less snow had fallen across northeast CO and southeast WY than guidance had indicated leading up to the event. The culprit could be observed in satellite imagery, as CYS mentions by 1137 UTC: “Radar and satellite both show relatively little signal of any significant precipitation much further south of the WY/CO border. GOES-16 imagery in the water vapor channel shows much drier air pushing northward against the cloud shield associated with the strengthening surface low. Diffluent motion in the cloud cover can also be discerned from satellite imagery over our area. This trend, combined with recent hires guidance, has guided a slight northward shift in our targeted area of heaviest snow, resulting in reduced snow totals along I-80 from Cheyenne to Sidney, but increased totals further north, especially in the Wheatland/Chugwater area.” This activity is apparent by the end of Fig 1.
Playing the water vapor animation through Saturday evening, the eastward progression of the dry/descending air and associated southwesterly jet is observed across the AZ/NM/Mexico border and into the central/southern plains as the mid/upper low shifts ene across NE/IA .
I could go one with additional water vapor imagery applications for this case, but there are other great applications to share! Along the southern periphery of the broad trough under the associated jet, gusty westerly to southwesterly winds mixed to the surface from southern CA to W Texas during the afternoon on the 4th and again on the 5th. This led to areas of blowing dust, reducing visibility, and resulting in the issuance of warnings and advisories
From ABQ on the 4th at 2120 UTC: “Blowing dust is evident on the latest Dust RGB loops (Fig 3)and lower vsbys have been reported at a few obs sites.” This area of blowing dust in northwest NM prompted the issuance of several Dust Storm Warnings.
Looking south in the EPZ forecast area: “Looking at split window GOES satellite imagery (Fig 4)… streamers of blowing dust are moving in from the Chihuahuan dust sources, so that small amount of precip isn’t doing much for dust inhibition.” A Dust Storm Warning was also issued for one of these streamers. The SWD is the key ingredient to satellite based dust detection RGBs and algorithms. Used alone, it is a great source for dust detection. Combining it with other data sources allows for the creation of more advanced products that can further isolate dust and provide information about additional features such as clouds, wildfires, etc.
Well north during the overnight hours of the 4th – 5th, BIS leveraged satellite imagery and fog probability products to assess fog/freezing drizzle potential: “Currently, surface low pressure was situated over the central Plains states with a weak reflection of an inverted trough northward into western North Dakota. Weak surface convergence was occurring here with surface observations and satellite fog probability product (Fig 6) indicating stratus/fog over western and south central ND and this is likely where any significant freezing drizzle remains. Farther east, dewpoint depressions increase quite a bit as you get into eastern portions of central ND with MVFR rather than IFR ceilings. Think the freezing drizzle potential here is very low. Will likely take a last look before issuing the products to see where we can trim off portions of the current advisory.” Viewing the new (with RPM v22, in Local Menu Items) Nighttime Microphysics + L2 product Readout, one can sample the RGB to view the Fog Probability Product readout information in the context of the RGB image, all in a single display (Fig 6, left). Learn more about this here.
To the northeast, TOP was providing IDSS to partners by alerting them to new wildfire starts per GOES Imagery: “Several fire have already been detected via GOES satellite imagery (Fig 7) and have been reported by local officials this afternoon.” At least five hot spots can be confidently diagnosed in the Fig 7 animation within the TOP forecast area. A sixth and seventh may be present to the southeast under the cloud cover. Early alerts of new wildfire starts (or significant developments of already established wildfires) may help emergency crews locate the fire, and allow them to arrive at the scene, sooner than otherwise. Other such examples have been documented on this blog (see here and here).
Two mesoscale sectors of 1-min imagery each were requested by NWS to support forecast operations on the 5th. One meso sector was requested by NWS Tulsa for Critical Fire Weather over the Southern Plains, and the other by SPC for the midwest severe weather threat. These sectors overlapped considerably across E OK/KS/KE and W MO/IA. As a result, 30-second imagery was available over these areas. When the center point of two meso sectors is the same (30-second imagery is requested), the meso overlap imagery can be viewed as meso-1 in AWIPS. This is not the case for a partial overlap. A few simple AWIPS modifications allows one to view this 30-second partial meso overlap imagery (hopefully the subject of a future AWIPS update). The thunderstorm associated with the EF4 Winterset/Newton Tornado developed within the overlap area, and is shown in Figs 8 and 9. Severe storm indicators present in the imagery include inflow feeder clouds, persistent overshooting tops, and above anvil cirrus plumes. Rotation of the exposed updraft may also be apparent at times.