1-minute Mesoscale Domain Sector GOES-19 (GOES-East) Visible images (with/without an overlay of the FDCA Fire Mask derived product) and Shortwave Infrared images (above) showed the thermal anomaly associated with a fast-moving grass fire (the Sharpe Fire) that moved northeastward across the Oklahoma border into Baca County in far southeastern Colorado on 17 May 2026 — prompting the issuance of an Evacuation Immediate order for southern Baca County residents in the vicinity of Campo at 2045 UTC. RAWS sites upstream of and near the fire recorded southwesterly wind gusts as high as 45-47 mph.

This grass fire burned very hot, and at 2017 UTC it first exhibited a 3.9 µm shortwave infrared brightness temperature of 137.77ºC (below) — which is the saturation temperature of GOES-19 ABI Band 7 detectors.

1-minute GOES-19 GeoColor RGB images with Next Generation Fire System (NGFS) Fire Detection polygons (below) also showed the rapid northeastward spread of the wind-driven fire’s thermal signature. The surface observation site near Campo, Colorado recorded relative humidity (green) values as low as 4%, and southwesterly wind gusts (red) as high as 45 mph. Note how the wind direction at area observation sites fluctuated from a more northerly-northeasterly direction early in the day to southwesterly later in the day, as a segment of a quasi-stationary frontal boundary briefly surged northward (surface analyses).

View only this post Read Less

On the afternoon of Friday, 15 May 2026, a wildfire broke out near Two Harbors, Minnesota, along the north shore of Lake Superior. Minnesota State Highway 61 was shut down to traffic and evacuation orders were issued for residents between Two Harbors and Castle Danger. Relative humidity levels were very low, with Two Harbors reporting only 21% RH at the time the fire ignited. While winds were low at the time the fire began, since then speeds have picked up and gusts in excess of 20 mph have been reported. This makes it challenging to keep the fire contained and as of the morning of the 16th this fire, dubbed the Stewart Trail fire, has spread to encompass 375 acres.

Taking a look at the satellite perspective, let’s start with the true color loop from GOES-19 (GOES West). If you didn’t quite know what to look for, it might be a challenge to identify where a fire is in this loop.In the middle of the image, in the second half of the loop, you might see a small smoke plume originating right next to the shore and streaming off to the east-northeast. However, the cumulus clouds are somewhat obscuring this plume and overall it’s a challenge to see where a fire might be when it’s young like this.

Fortunately, we have other bands available to us. GOES Band 7 (3.9 microns) is an excellent tool for detecting and monitoring fires, and this case is no exception. The 3.9 micron band is very sensitive to objects with fire-like temperatures, so it is able to note the presence of fire when longer-wavelength bands don’t show a signal. You can watch this loop from 1900-2100 UTC (2:00 PM to 4:00 PM CDT) and easily identify the location of the fire as the large dark spot next to Lake Superior that appears midway through. Note that during the day this time of year, Superior appears cooler than the surrounding land thanks to its massive thermal inertia.

The Fire RGB product also helps us detect fires by combining fire signals from three different shortwave channels. As a fire grows more intense, its signal will appear on shorter wavelength channels. Here we see the fire clearly show up as a set of red pixels. This means that the fire is on the lower end of intensity, even though it is still a significant fire.

Of course, while geostationary satellites excel in temporal resolution, their spatial resolution is not as good as that of the polar orbiting systems. Taking a look at the true color image from NOAA-21 at 1951 UTC (2:51 PM local), it’s a little easier to pick up the smoke plume than it was with the geostationary satellite.

But what happens when we look at a band sensitive to fire temperatures? VIIRS offers the I4 Band, with 375 m resolution at 3.74 microns. There is a noticeable black spot in the center of the image corresponding to the location of the fire. Note how this is much smaller than the black spot seen on the geostationary. This VIIRS view is more likely to be an accurate assessment of the true size of the fire at the image time, which means it can be a valuable tool for forecasters, emergency managers, and first responders.

In fact, we can compare the GOES and VIIRS directly. Use the slider to se how the apparent extent of the fire changes depending on the resolution of the instrument used to monitor it. This time, 1951 UTC, is early in the fire’s development and so it barely appears as a signal in GOES. However, it is readily apparent as a small but intense patch in VIIRS.

View only this post Read Less

1-minute Mesoscale Domain Sector GOES-19 (GOES-East) Visible images (with/without an overlay of the FDCA Fire Mask derived product), Shortwave Infrared images and Infrared Window images (above) showed the thermal anomalies associated with several grass fires south and southwest of Dodge City (KDDC) in southwestern Kansas — and two of these grass fires produced pyrocumulonimbus (pyroCb) clouds late in the day on 15 May 2026. The initial (and largest) pyroCb exhibited a 24-minute period of GLM-detected lightning activity (beginning at 2234 UTC, and ending at 2258 UTC).

Each pyroCb initially exhibited cloud-top infrared brightness temperatures in the -40s C (shades of blue to cyan), which is a necessary trait to be classified as a pyrocumulonimbus (since those temperatures assure that heterogeneous glaciation has occurred at the cloud top) — and the coldest cloud-top infrared brightness temperature of the larger pyroCb eventually reached -50.61ºC at 2356 UTC on 15 May (above). According to a plot of rawinsonde data from Dodge City (below), that -50.61ºC temperature represented an altitude not far below the Equilibrium Level (EL) of an ascending Forecast Surface air parcel.

The coldest cloud-top infrared brightness temperature of the second (smaller) pyroCb was -42.31ºC at 0014 UTC on 16 May (below). Note that the cloud material of this second pyroCb had more of a southeastward drift than the first pyroCb — rawinsonde data depicted light northwest winds near the altitude of the -42ºC temperature level (in contrast to westerly winds at slightly higher altitudes near the Equilibrium Level).

1-minute GOES-19 True Color RGB images from the CSPP GeoSphere site (below) better highlighted the dark burn scars of the grass fires, as well as the hazy pall of smoke that was spreading northward and westward (at Dodge City, this smoke reduced the surface visibility to 3-4 miles at times).

The grass fires that produced the 2 pyroCb clouds were burning in Clark County and Meade County — both of which were experiencing Severe to Extreme Drought conditions.

View only this post Read Less

On the afternoon of 14 May 2026, a strong midlatitude cyclone centered over the prairie provinces of Canada forced very strong winds in the Northern Great Plains of the continental United States. This in turn lofted an extreme amount of dust into the lower atmosphere, lowering visibilities to less than a mile and creating hazardous conditions across parts of the Dakotas. KELO-TV in Sioux Falls, South Dakota, has images and video of this event. Satellites, of course, are an excellent tool to diagnose the origin and extent of this kind of event.

But first, let’s start with the surface chart. The following image shows the 1200 UTC surface analysis on 14 May 2026 from NOAA’s Weather Prediction Center Surface Map Archive. It depicts a strong cyclone centered over western Saskatchewan with a central pressure of 985 hPa and a modestly high isobar gradient. An occluded font extends to the Saskatchewan/Manitoba/North Dakota triple point, with a cold front stretching southward and a warm front extending to the southeast.

The satellite view of this same time, as captured by the GOES-19 (East) Band 13 (10.3 micron) channel, clearly shows the presence of a mature cyclone. Since this is 1200 UTC, the sun hasn’t quite risen yet over western Canada and Montana, so the infrared band remains the go-to tool for a couple of hours yet. We can see some deeper cloud development ahead of the occluded front and along the cold front as well.

As the day progressed, this cyclone intensified. The central pressure dropped and the pressure gradient tightened. This resulted in a substantial increase in wind speeds across the region. This next loop depicts the GOES-19 True Color view of the cyclone throughout the daytime hours of the 14th on a 30 min time scale. Watch as the dust originates over regions ob both north-central and western Montana and then is advected eastward.

There are going to be two significant ingredients for a dust storm: strong winds, and bare soil. We’ll start with the first and look at the winds at the time. Here’s a look at the winds from across the northern tier of the continental United States as captured by the NOAA Weather Prediction Center. This is from 1800 UTC, so at the time the dust was really starting to pick up according to our satellite loop. Note the presence of 30 and 40 kt winds across much of the region. A 40 kt wind is 46 mph, and it’s important to remember that these are sustained winds, so gusts are even higher.

The other important ingredient, of course, is having bare soil. Strong winds can blow across vegetated regions and not contribute to significant dust lofting as the vegetation anchors the soil and prevents it from being blown around. However, if there are bare fields then winds can easily pick up the dirt and carry it. We can use satellites to assess how much vegetation is present. First, let’s start with the Normalized Difference Vegetation Index (NDVI) from the JPSS Direct Broadcast. Remember, plants absorb visible light to make food via photosynthesis, so their visible reflectance is quite low. But vegetation reflects a lot of shortwave infrared light. If we compare those two channels, we can identify what parts of the land is vegetated versus bare, snow, or water as they all have different spectral signatures. Here’s a plot of NDVI from 12 May as captured by VIIRS from NOAA-21 and processed by SSEC’s own Community Satellite Processing Package (CSPP). This plot is shaded with greener regions corresponding to higher NDVIs and browner representing lower ones. While there was some cloud cover in western North Dakota obscuring the signal somewhat, we can see that the regions where the dust originated in Montana correspond to low NDVI values.

We can even track NDVI values over time. Here is a plot from the European Union’s Sentinel 2 satellite, as recorded in the Copernicus Browser. A region of eastern Montana was selected, and the mean NDVI for that region for every satellite overpass over the course of a year was calculated. The rightmost part of this time series represents the most recent observation. Note that the NDVI is higher now than it was in the dead of winter when the region was covered in snow, but it is still much less than it is in the summer peak when crops (mostly wheat and sunflowers) are at their maximum greenness before being harvested. This is a good indicator that there’s a substantial amount of bare ground in this region.

But we can even look at the spectral signature of this area. Sentinel 2 observes at several different channels in the visible and near infrared, and we can plot up the reflectance as a function of wavelength for all of those channels. The most recent spectral signal for western Montana is plotted below in light green. The typical spectral reflectance for grass (dark green) and soil (brown) is also shown. Note how closely the observed region matches the soil plot while having very little in common with the grass plot. Clearly, this is a region that is dominated by bare soil.

So, we clearly have our two main ingredients of strong winds and bare soil. When we put them together, we definitely get an intense dust storm. That was born out by the surface observations at Minot, in western North Dakota. Here’s a meteogram from Minot in western North Dakota courtesy of the Plymouth State Weather Archive. The “vis” line is the visibility in statute miles, and the wgst line shows the wind gusts in knots. Note how the gusts at 2000 UTC are at 54 kts (62 mph)! No wonder the visibility dropped to just a mile. Conditions were even more extreme further west in Williston, ND, where gusts as high as 67 mph were recorded.

One of the most useful tools for identifying and tracking dust storms is the Dust RGB product. Here’s an animation of the Dust RGB from GOES-19 throughout the daytime hours on the 14th. The magenta regions denote high dust concentrations and easily identify where the dust originated and clearly show it being drawn northward into the center of rotation of the low pressure system. The recipe for the Dust RGB only relies on longwave bands, so while we’re only showing the daytime hours in this loop, it’s available and useful at all hours of the day.

The dust also clearly affected the optical depth of the atmosphere. Here is the aerosol optical depth (AOD) as recorded by the polar orbiting NOAA-21 and retrieved by CSPP as it passed overhead at 2010 UTC. Note the intense band of high AODs stretching from north-central Montana to central North Dakota.

____________________________

A longer animation combining GOES-19 daytime True Color RGB and nighttime Dust RGB images created using Geo2Grid (above) showed that the airborne dust progressed as far eastward as Ontario, Minnesota and far northwest Wisconsin by the morning of 15 May. Peak wind gusts on 14 May included 81 mph in Montana and 76 mph in North Dakota.

____________________________

Dust storms aren’t uncommon in the central United States this time of year. The Blog covered a less intense event in eastern North Dakota and Minnesota a few days earlier while it was almost exactly one year to the day since a major dust storm impacted Chicago, an event we also covered. Here in the early part of the growing season, when fields are being tilled for planting, the time is ripe for these kinds of events to take place.

View only this post Read Less