Category Archives: Northern Plains

Methods of Fog Detection in the GOES-16 Era

GOES-16 ‘Fog Detection’ Channel Difference (10.3 µm – 3.9 µm), 0912 – 1132 UTC, 29 June 2017 (Click to enlarge)

GOES-16 data posted on this page are (still!) preliminary, non-operational data and are undergoing testing

The 16 channels on the GOES-R Series Advanced Baseline Imager (ABI) allow for many different channel combinations that can be used to detect atmospheric phenomena. The animation above shows the traditional method for detecting low stratus: the brightness temperature difference between the shortwave infrared (3.9 µm) and the cleanest longwave infrared (10.3 µm) windows. Cloudtops composed of water droplets are highlighted in the animation because they do not emit 3.9 µm radiation as a blackbody, but do emit 10.3 µm radiation as a blackbody; thus, the brightness temperature difference at night (when no reflected solar radiation at 3.9 µm is present) is positive. The range of the colorbar in the above animation is from -50 to +50 C; stratus appears as green over much of northern Wisconsin, Minnesota and North Dakota.  Higher cirrus clouds are cyan, and they interfere with the satellite detection of low clouds, especially over eastern North Dakota where IFR conditions were widespread (source), and where a Dense Fog Advisory existed.  Note the apparent disappearance of the fog signal — in green — as the sun rises.  Increasing amounts of reflected solar radiation are causing the brightness temperature difference value to switch sign from (10.3 µm – 3.9 µm) > 0 at night (because of emissivity differences) to (10.3 µm – 3.9 µm) < 0 during the day (because of solar reflectance).

The ‘Nighttime Microphysics Advanced RGB’ is also used as a fog detection device. In the animation below, low stratus (and by inference, fog) is highlighted in cyan, a signal that comes mostly from the ‘green’ part of the RGB, namely the Brightness Temperature Difference as shown above. Because the two products are linked by the 10.3 – 3.9 brightness temperature difference, shortcomings in that product as far as fog detection affect the RGB. Note how the fog signal erodes over Minnesota/Wisconsin as the sun rises, and how it is obscured by high clouds (dark purple/magenta) over North Dakota.


Note: GOES-R IFR Probabilities are computed using Legacy GOES (GOES-13 and GOES-15) and Rapid Refresh model information; GOES-16 data will be incorporated into the IFR Probability algorithm in late 2017

GOES-R IFR Probability fields, shown for this event below, were designed to mitigate detection issues noted above.  Where high clouds are present, meaning the satellite cannot detect low clouds, information about low-level saturation from the Rapid Refresh is used to assess whether or not fog is likely.    That low-level information from the model also can be used to distinguish between fog and elevated stratus that can look very similar from the top, as a satellite views it.  The fusing of model and satellite data makes for a product that has better statistics in detecting low ceilings and reduced visibilities.

At the end of the two animations above, for example, how confident will a satellite analyst or forecaster be that there is dense fog over eastern North Dakota?  How about the analyst/forecaster using IFR Probability fields? IFR Probabilities maintain a signal for fog over the entire region from North Dakota to Wisconsin, even through sunrise and under high clouds.

GOES-R IFR Probabilities, 0915 to 1115 UTC on 29 June 2017 (Click to enlarge)

IFR Conditions over North Dakota

GOES-13 Brightness Temperature Difference and GOES-R IFR Probability at 0615 UTC on 24 April 2017, along with surface observations (Click to enlarge)

Note: GOES-R IFR Probabilities are computed using Legacy GOES (GOES-13 and GOES-15) and Rapid Refresh model information; GOES-16 data will be incorporated into the IFR Probability algorithm in late 2017.

A late-season snow storm was affecting North Dakota on Monday 24 April. Snow amounts were modest and IFR Conditions were widespread.   The multiple clouds layers associated with the storm meant that satellite detection of low clouds/fog was difficult.  The toggle above shows Brightness Temperature Difference (3.9 µm – 10.7 µm) and GOES-R IFR Probabilities.  The IFR Probability field distinctly outlines the region where visibilities and ceilings are restricted by the storm.  It is very difficult to discern from the Brightness Temperature Difference product where low clouds/fog exist.  Because the IFR Probability field incorporates surface information (that is, low-level saturation as predicted by the Rapid Refresh Model), it is better able to alert a forecaster to the presence of IFR or near-IFR conditions.

Low IFR and IFR Conditions over the central United States

Note: GOES-R IFR Probabities continue to be computed with GOES-13 and GOES-15 data only. Incorporation of GOES-16 data will occur near the end of 2017.

Front Page from the Aviation Weather Center webpage ( at 1155 UTC 3 April 2017. Stations with IFR / LIFR conditions are indicated by red / magenta. (Click to enlarge)

A wide swath of Dense Fog Advisories were hoisted over the central part of the United States on Monday 3 April 2017 in association with southerly flow and a variety of fronts. The Aviation Weather Center front page (screen-captured, as shown above), showed Low IFR and IFR conditions throughout the central Plains.

The development of the Low IFR Conditions coincided with an increase in Low IFR Probabilities during the night, as shown by the stepped animation below showing data from 0200, 0500 and 0915 UTC. Low IFR Probabilities over Iowa, Nebraska and Kansas, part of a smooth yellow field, are driven solely by Rapid Refresh Data in a region where high and mid-level clouds are preventing the satellite from observing low clouds.

GOES-R Low IFR Probabilities at 0200, 0500, 0915 UTC on 3 April 2017 (Click to enlarge)

GOES-R IFR Probabilities, below, generally cover the same region as Low IFR Probabilities shown above, but have larger values. As with the Low IFR Probabilities, model data only are determining the probabilities over much of Iowa, Nebraska and Kansas, a region where the probability field is uniform and flat, especially at the end of the animation.

GOES-R IFR Probabilities at 0200, 0500, 0915 UTC on 3 April 2017 (Click to enlarge)

GOES-R Low IFR Probability is shown below with surface observations superimposed. There is a good relationship between high probabilities and observed IFR and Low IFR conditions.

GOES-R IFR Probability fields, 0945 UTC on 3 April 2017 along with surface observations of ceilings and visibilities at 1000 UTC (Click to enlarge)

Note: GOES-R IFR Probabities continue to be computed with GOES-13 and GOES-15 data only. Incorporation of GOES-16 data will occur near the end of 2017.

Fast-moving Fog over northeast Montana

GOES-R IFR Probability fields computed with GOES-13 and Rapid Refresh Data, 1400-1500 UTC on 24 March 2017 (Click to enlarge)

Rains over Montana earlier this month (data from this site) (along with snowmelt) caused substantial flooding on Big Muddy Creek in the extreme northeast part of the state. Saturated soils in that region have increased the likelihood of fog, and fog was indicated by IFR Probability in that region on the morning of 24 March as shown above.

GOES-16 Visible Imagery showed the fog speedily moving down Big Muddy Creek. An animation using GOES-13 Visible imagery (0.64 µm) is shown below. The GOES-16 CONUS cadence is every five minutes; it is every 15 minutes with GOES-13, except when Full Disks are being scanned (14:45 UTC) or when housekeeping is occurring (15:30 UTC).

GOES-13 and GOES-16 visible data both show quick movement of the fog. For this case, it was harder to judge motion from the IFR Probability fields. This could be related to Infrared and model resolution; the creek valley might be too narrow for the satellite infrared data and for the model.

GOES-13 VIsible (0.64 µm) imagery, 1415-1615 UTC. Sheridan County Montana is outlined in the first image. Fog advancing down Big Muddy Creek is apparent

For over the Northern Plains


GOES-R IFR Probability Fields, computed from GOES-13 and Rapid Refresh output, 0215-1115 UTC on 1 June 2016 (Click to enlarge)

Dense Fog Advisories were issued over parts of Iowa and Minnesota early on 1 June 2016 (see map below). The fog developed over wet ground left in the wake of convection that moved through the region late in the day on 30 May/early on 1 June (Precipitation totals available here).  GOES-R IFR Probability fields, above, show the two areas of dense fog developing.  The region over Minnesota was characterized a lack of high clouds — the satellite could view the developing fog, and satellite parameters were included in the computation of IFR Probability.  Consequently, the IFR Probability values were larger.

Fog over Iowa initially developed under mid-level clouds behind departing convection. IFR Probability fields in that case show a flatter distribution because horizontal variability is controlled mostly by model fields that are smooth; additionally, IFR Probability values are somewhat reduced because satellite predictors cannot be used. By 0815 UTC, however, mid- and high-level clouds have dissipated, and the satellite has a unobstructed view of the fog/stratus. Satellite predictors could then be used and IFR Probabilities increased, and the field itself shows more horizontal variability as might be expected from the use of nominal 4-km resolution satellite pixels.

Screen Capture of website at 1129 UTC on 1 June. Dense Fog Advisories are indicated over eastern Iowa and northeast Minnesota (click to enlarge)

Fog over the Great Lakes


GOES-R IFR Probability fields computed with GOES-13 and Rapid Refresh Data, 1215 UTC on 26 May 2016 along with surface reports of Ceilings and visibilities (Click to enlarge)

High Dewpoint air (upper 50s and low- to mid-60s) has overrun the western Great Lakes, where water temperatures are closer to the mid 40s.  (Water Temperature from Buoy 45007 in southern Lake Michigan).  Advection fog is a result, and that fog can penetrate inland at night, or join up with fog that develops over night.  The image above shows the extent of low visibilities over the upper Midwest and the IFR Probability field early morning on the 26th of May. Lakes Michigan and Superior are diagnosed as socked in with fog. A similar field from 1945 UTC on 25 May similarly shows very high Probabilities over the cold Lakes. Expect high IFR Probabilities to persist over the western Great Lakes until the current weather pattern shifts.

Brightness Temperature Difference Fields can also show stratus over the Great Lakes, of course, but only if multiple cloud layers between the top of the stratus and the satellite do not exist. Convection over the upper Midwest overnight on 25-26 May frequently blocked the satellite’s view of the advection fog. The toggle below, from 0515 UTC on 26 May, shows how model data from the Rapid Refresh is able to supply guidance on IFR probability even in the absence of satellite information about low stratus over the Lakes.


GOES-13 Brightness Temperature Difference Fields and GOES-R IFR Probability fields, 0515 UTC on 26 May 2016 (Click to enlarge)

Dense Fog and Freezing Fog in Montana and North Dakota


GOES-R IFR Probability Fields, 0300-1300 UTC on 20 January 2015 (Click to enlarge)

Dense Fog over North Dakota and Freezing Fog over eastern Montana prompted the issuance of Dense and Freezing Fog advisories early Wednesday Morning.    GOES-R IFR Probabilities, above, computed from data from the GOES-13 Imager and from Rapid Refresh Model output, capture the growth/evolution of the fog field. In general, the IFR Probability field captures the area of reduced visibilities (with some exceptions, such as KHEI (Hettinger, ND, near the border with South Dakota is southwest North Dakota).

Compare the areal extent of the IFR Probability field with that from the Brightness Temperature Difference from GOES, below. The presence of multiple cloud layers prevents any satellite from viewing low clouds, and satellite-only products therefore give little information about near-surface events in much of western North Dakota and Montana.  When Rapid Refresh Model output is controlling the GOES-R IFR Probability field, as happened in this case over Montana and western North Dakota, the IFR Probability will have lower values and a less pixelated look that reflects the coarser model resolution and model smoothing.


GOES Brightness Temperature Difference fields, 0300-1300 UTC, 20 January 2016 (Click to enlarge)

Fog and Stratus over the Northern Plains


GOES-R IFR Probability Fields, 0345-1345 UTC on 21 December 2015 (Click to enlarge)

The National Weather Service in Bismarck issued Dense Fog Advisories for a fog and freezing fog early in the morning on December 21 (Happy Solstice!!) 2015. (Link). The half-hourly animation above shows the GOES-R IFR Probability during the overnight hours of 20-21 December 2015. (Here is a faster animation). Several aspects of this animation warrant comment. First, the western edge of the IFR Conditions matches well with the western edge of highest IFR Probabilities over North Dakota. This is true even as high clouds (obvious in the animation of Brightness Temperature Difference, below) move over North Dakota from the west: When this happens, IFR Probabilities decrease (and the field itself becomes more horizontally uniform) because Rapid Refresh Model Data output is the main predictor being used to diagnose the IFR Probability. The edge of the IFR Probability field moves through Dickinson ND (in the southwest part of the state) as the ceilings and visibilities there improve.

In addition, a persistent region of small IFR Probabilities exists over northern Minnesota in a region where IFR conditions are not reported. The Brightness Temperature Difference field there (below) shows a strong signal. (Click here for a faster animation of Brightness Temperature Difference) Thus, GOES-R IFR Probability is better able to differentiate between mid-level stratus and low stratus/fog. Over northern Minnesota, the Satellite data says water-based clouds are present, but the Rapid Refresh data notes little saturation in the lowest 1000 feet of the model. Thus, IFR Probabilities are small there.


GOES-13 Brightness Temperature Difference Fields (10.7 µm – 3.9 µm), 0345-1345 UTC on 21 December 2015 (Click to enlarge)

Freezing Fog over Minnesota


GOES-R IFR Probability fields, hourly from 0515 through 1300 UTC on Sunday 22 November 2015 (Click to enlarge)

On Sunday 22 November, fog developed over west-central Minnesota with sub-freezing temperatures on the ground. The animation of the GOES-R IFR Probability fields, above, shows the slow enlarging of highest probabilities in the region around Glenwood and Alexandria where IFR Conditions were observed.

The Brightness Temperature Difference fields, below, show a signal over much of the region. One of the benefits of GOES-R IFR probability fields is that the use of Rapid Refresh model output screens out regions where mid-level stratus is more likely. Only regions in the model that have low-level saturation show the highest IFR Probabilities. GOES-R IFR Probabilities also give a consistent signal for IFR conditions when high clouds move over a surface fog. At the end of the animation above, probabilities decrease somewhat in regions over western Minnesota where high clouds are encroaching. Those high clouds prevent the brightness temperature difference field, below, from detecting surface fog.


GOES-13 Brightness Temperature Difference Fields (10.7µm – 3.9µm), hourly from 0515 through 1300 UTC on Sunday 22 November 2015 (Click to enlarge)

Dense Fog Detection over western North Dakota under Cirrus


GOES-R IFR Probabilities (Upper Left), GOES-R Cloud Thickness (Lower Left), GOES-13 Brightness Temperature Difference (10.7 µm – 3.9 µm) (Upper Right), GOES-13 Water Vapor Infrared Imagery (6.5 µm) (Lower Right) (Click to enlarge)

Dense fog formed over western North Dakota early in the morning of Monday 26 October (and advisories were issued). At the same time, cirrus clouds overspread the region, making satellite detection of the low clouds problematic. This is an excellent example of the benefits of a fused product that blends surface and near-surface information from numerical models with satellite detection. When one of the products gives no information, the other can be relied upon to fill in gaps.

In the example above, dense fog develops over western North Dakota, complete with freezing drizzle. Satellite detection of the low clouds on this date was difficult because of widespread cirrus that overspread the state; these cirrus are apparent in both the water vapor (lower right) and the brightness temperature difference (upper right). When model data principally are used to compute the IFR Probability fields — in regions where high clouds prevent a satellite view of low clouds — the character of the field is flatter and less pixelated. In addition, where high clouds cannot view low clouds, the Cloud Thickness cannot be computed (as it relates 3.9µm emissivity to cloud thickness). Note that there are regions where the cirrus thins enough that low clouds can be viewed — in these regions the Cloud Thickness is computed an the IFR Probability increases (where both Cloud Predictors and Model Predictors can be used, IFR Probabilities are larger).