Category Archives: Midwest

Dense Fog over Missouri

GOES-R IFR Probability Fields, and surface reports of Ceilings and Visibilities, hourly from 0315 to 1315 UTC on 15 August 2017 (Click to enlarge)

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

IFR Conditions and Dense Fog developed over southeastern Missouri during the early morning on 15 August 2017, leading to the issuance of Dense Fog Advisories. GOES-R IFR Probabilities, above, showed increasing values as the ceilings lowered and visibilities dropped.  The IFR Probability fields over southern Missouri and northern Arkansas (and Kansas and Oklahoma) have the characteristic uniformity that arises when Rapid Refresh data alone are used to drive the IFR Probability values.  In these regions, high clouds (associated with convection over Arkansas) are blocking the satellite view of lower clouds.

Such high clouds will make it difficult for a satellite-only product to identify the regions of clouds.  For example, the Brightness Temperature Difference field below (10.3 µm – 3.9 µm) from GOES-16, color-enhanced so that low clouds are green and that cirrus (at night) are purple) shows widespread low cloudiness at the start of the animation, including some obvious river fog over Missouri (river valleys that are not well-resolved with GOES-13), but developing convection over Arkansas eventually prevents the view of low clouds.

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

GOES-16 Brightness Temperature Difference field (10.3 µm – 3.9 µm), hourly from 0312 to 1312 UTC on 15 August 2017 (Click to enlarge)

Because the Brightness Temperature Difference field (helpfully called the ‘Fog’ Channel Difference in AWIPS) is challenged by high clouds in viewing the low clouds, RGB Products that use the brightness temperature difference field are also similarly impeded by high clouds.  Consider, for example, the stations in southern Missouri shrouded by the cirrus shield.  IFR Conditions are occurring there and a strong signal of that appears in the IFR Probability fields.

GOES-R IFR Probability computed with GOES-13 and Rapid Refresh Data, GOES-16 Fog (10.3 µm – 3.9 µm) Brightness Temperature Difference and GOES-16 Advanced Nighttime Microphysics RGB, all near 1115 UTC on 15 August 2017 (Click to enlarge)

IFR Conditions under multiple cloud decks in the Upper Midwest

GOES-R IFR Probability Field, along with observations of surface visibility and ceiling heights, 1100 UTC on 17 May 2017 (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

The GOES-16 data posted on this page are preliminary, non-operational data and are undergoing testing

A morning screenshot from the Aviation Weather Center website shows a wide region of IFR and Low IFR Conditions reported over the Upper Midwest, from eastern North Dakota eastward to Lake Superior. The GOES-R IFR Probability field, above, from 1100 UTC on 17 May 2017, shows high probabilities over that region.

Over much of Minnesota and Michigan, the character of the IFR Probability field is flat.  This is typical of IFR Probability when high clouds prevent satellite data from being used as a statistical predictor for IFR Conditions.  If high clouds are present, the satellite cannot detect the presence of low clouds, and the chief predictor of IFR conditions will therefore be model data that typically does not vary strongly from gridpoint to gridpoint when IFR conditions are present.  The pronounced boundary apparent in the IFR Probability field that extends nortwestward from Green Bay in Wisconsin is the boundary between night-time predictors (to the west) and daytime predictors (to the east).

GOES-R IFR Probability values are largest in the region of IFR conditions over North Dakota.  In this region, high clouds are not present and the satellite is able to detect low clouds, and that information is part of the computation of IFR Probabilities.  Note also the region in east-central Minnesota where satellite data are also being used in the computation of IFR Probabilities;  the resultant field there is pixelated.

The toggle below shows the brightness temperature difference field between the shortwave and longwave infrared window channels from GOES-13 (3.9 µm and 10.7 µm) and GOES-16 (3.9 µm and 10.33 µm).  This brightness temperature difference field is used to detect stratus, and by inference fog, because stratus cloud tops composed of water droplets emit radiation around 10.3-10.7 µm as a blackbody, but do not emit 3.9 µm radiation as a blackbody.  Satellite detection of radiation, and computation of the inferred temperature of the emitting surface, assumes blackbody emissions.  Consequently, the brightness temperature computed using detected 3.9 µm radiation is colder than that computed using 10.7 µm (or 10.33 µm) radiation.

Both satellites capture the region of low stratus/fog over North Dakota, and the superior spatial resolution of GOES-16 is apparent. Note, however, that neither satellite can detect low clouds associated with dense fog in regions of higher clouds — over Michigan’s Keewenaw Peninsula, for example, or along the shore of Lake Superior in Minnesota. This is an unavoidable shortcoming of satellite-only-based detection of low clouds/fog.

GOES-13 (3.9 µm – 10.7 µm) and GOES-16 (10.33 µm – 3.9 µm) Brightness Temperature Difference fields, 1100 UTC on 17 May 2017 (Click to enlarge)

Dense Fog from the Ohio Valley to North Carolina

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.

The weather.gov website on Wednesday morning 10 May 2017 showed two dense fog advisories, one near Cincinnati, OH and one near Greensboro, NC. The aviation weather website showed an IFR Sigmet in between the two regions of dense fog. The fog formed along a stationary front that sat over the region.

How well did GOES-R IFR Probabilities and GOES-13 Brightness Temperature Difference fields capture this event? The animation of GOES-R IFR Probability, below, computed using data from GOES-13 and the Rapid Refresh Model, shows enhanced probabilities early in the evening that increased with time. The orientation of the field — from west-northwest to east-southeast — aligns well with the regions of developing fog.

GOES-R IFR Probability fields, 0100, 0400 and 0700 – 1100 UTC on 10 May 2017 (Click to enlarge)

The brightness temperature difference field, below, did not perform as well in outlining the region of low ceilings/reduced visibilities because of the presence of high clouds that interfered with the ability to detect low clouds. Consequently, the highest brightness temperature differences (3.9 µm – 10.7 µm) do not align so well with the regions of developing fog. Note also that at the end of the animation — 1100 UTC — increasing amounts of reflected solar 3.9 µm radiation is changing the character of the field from negative to positive. In contrast, the IFR Probability fields (above) maintain a consistent signal through sunrise.

GOES-13 Brightness Temperature Difference fields (3.9 µm – 10.7 µm), 0700-1100 UTC on 10 May 2017 (Click to enlarge)

Dense fog in Ohio, Indiana and Illinois

Toggle between 1100 UTC GOES-R IFR Probability fields and GOES-13 Brightness Temperature Difference (3.9 µm – 10.7 µm) fields, with surface observations of ceilings and observations. (Click to enlarge)

Dense Fog developed over central/southern Ohio, Indiana and Illinois on the morning of 20 February 2017 (Screen shot from here; IFR Depiction from here). The toggle above includes the GOES-R IFR Probability fields; large values of IFR Probability overspread the region of low ceilings and visibilities. In contrast, high clouds (dark grey in the enhancement used) are preventing the Brightness Temperature Difference field from articulating where the fog/low ceilings might be occurring. Because satellite data cannot be used as a predictor at 1100 UTC, the character of the IFR Probability field is mostly uniform, lacking the pixelation that occurs when low clouds can be viewed from the satellite (as over southwestern Pennsylvania, West Virginia, central Kentucky, parts of central Illinois, and elsewhere).

High clouds have not moved into Ohio and Indiana at 0500 UTC on 20 February as the fog was developing (they are present over Illinois, however). The toggle below, from that time, shows low probabilities of IFR conditions over Ohio, and only a few reports of IFR conditions, mostly over central and southern Ohio.

Toggle between 0500 UTC GOES-R IFR Probability fields and GOES-13 Brightness Temperature Difference (3.9 µm – 10.7 µm) fields, with surface observations of ceilings and observations. (Click to enlarge)

Toggle between 0700 UTC GOES-R IFR Probability fields and GOES-13 Brightness Temperature Difference (3.9 µm – 10.7 µm) fields, with surface observations of ceilings and observations. (Click to enlarge)

By 0700 UTC (above), IFR conditions are becoming more widespread as IFR Probabilities increase.  The dark regions in the Brightness Temperature Difference fields over Wisconsin and Lower Michigan and surrounding regions show the advance of high clouds.  By 1000 UTC, those clouds have overspread Ohio and Indiana, and the brightness temperature difference field loses utility as far as low-cloud detection goes.  Because the IFR Probability fields incorporate low-level saturation information from the Rapid Refresh Model, however, IFR Probability fields can continue to provide a useful signal when high clouds are present or move in.

Toggle between 1000 UTC GOES-R IFR Probability fields and GOES-13 Brightness Temperature Difference (3.9 µm – 10.7 µm) fields, with surface observations of ceilings and observations. (Click to enlarge)

Fog and Ice Fog over the southern Plains

ifrp_0400_1215_05dec2016anim

GOES-R IFR Probabilities, hourly from 0400 through 1215 UTC on 5 December 2016 (Click to enlarge)

Dense Fog developed over the southern Plains early on Monday 5 December, and the GOES-R IFR Probability fields, above, were a tool that could be used to monitor the evolution of this event. A challenge presented on this date was the widespread cirrus (Here’s the 0700 UTC GOES-13 Water Vapor (6.5 µm) Image, for example) that prevented satellite detection of low clouds. The Brightness Temperature Difference fields, below, at 3-hourly intervals, also show a signature (dark grey/black in the enhancement used) of high clouds, although they are shifting east with time — by 1300 UTC there is a signature (orange/yellow in the enhancement used) of stratus clouds over central and eastern Oklahoma.

The IFR Probability fields, above, have a characteristic flat nature over Arkansas and Missouri, that is, a uniformity to the field, that is typical when model data are driving the probabilities. The more pixelated nature to the fields over Kansas and Oklahoma, especially near the end of the animation, typifies what the fields look like when both satellite data and model data are driving the computation of probabilities. Careful inspection of the fields over Arkansas shows regions — around Fayetteville, for example, around 1000 UTC where IFR Probabilities are too low given the observation at the airport of IFR conditions. This inconsistency gives information on either the small-scale nature of the fog (unlikely in this case) or on the accuracy of the Rapid Refresh model simulation that is contributing to the probabilities. In general, the Rapid Refresh model has accurately captured this event, and therefore the IFR Probabilities are mostly overlapping regions of IFR or near-IFR conditions. The region over southern Illinois that has stratus, and low probabilities of IFR conditions, for example. Adjacent regions have higher IFR Probabilities and lower ceilings and/or reduced visibilities. A screen shot from the National Weather Service, and from the Aviation Weather Center, at about 1300 UTC document the advisories that were issued for this event.

btd_0400_1300_05dec2016step

Brightness Temperature Difference fields (3.9 µm – 10.7 µm), 0400, 0700, 1000 and 1300 UTC on 5 December 2016 (Click to enlarge)

IFR Probability Motion as a forecast tool

ifrprobability_13oct2016_0200_1407anim

GOES-R IFR Probability Fields, 0200-1400 UTC on 13 October 2016 along with surface observations of ceilings/visibility (Click to enlarge)

Because GOES-R IFR Probability fields are computed with the same time latency as GOES imagery, motion of the IFR Probability fields can have predictive value.  In the animation above, higher GOES-R IFR Probability  is moving eastward;  IFR Conditions are reported as the higher IFR conditions move overhead (consider, for example, Bowling Green, KY, or Clarksville, TN), and ceilings / visibilities improve as the band of higher IFR conditions moves eastward from a station (over southern Illinois, for example).

Fog over the lower Ohio River Valley

goesr_ifrp_3october2016anim_0100_0700

GOES-R IFR Probability Fields, 0200-0700 UTC on 3 October 2016 (Click to enlarge)

Fog developed over portions of the Ohio River Valley from Indiana westward to the Mississippi River at Cairo IL on the morning of 3 October 2016.   Dense Fog Advisories were issued by the National Weather Service Offices in St. Louis, Lincoln IL and Paducah between 3:30 and 4:15 CDT (0830 to 0915 UTC).  A SIGMET was also issued.  How effective was Satellite detection of this developing fog? 

The brightness temperature difference product, below, shows hourly measures of water-based clouds, a detection that keys off the emissivity differences of water based clouds for 3.9 radiation (at which wavelength near-blackbody emission is not occurring) and 10.7 radaiation (at which wavelength near-blackbody emission is occurring).  Significant changes to the brightness temperature difference field did not occur until after 0500 UTC.   In addition, Brightness Temperature Difference fields overestimated the region of developing fog.  In contrast, the GOES-R IFR Probability field, above, showed a more gradual increase from 0200 UTC onward, and the region of the strong signal was better confined to where dense fog developed. On this day, GOES-R IFR Probability fields were better for situational awareness, generating an earlier alert for forecasters to the potential for fog. In addition, the GOES-R IFR Probability fields better defined the region of hazardous ceilings and visibilities.

btd_anim_03octob2016_0100_0700

Brightness Temperature Difference Field (3.9 – 10.7) from 0100 to 0700 UTC on 3 October 2016 (Click the enlarge)

Cloud Thickness and Dissipation Time

goes-rcloudthickness_1130utc_20sept2016

GOES-R Cloud Thickness Fields, 1130 UTC on 20 September 2016 (Click to enlarge)

GOES-R Cloud Thickness is created from a look-up table created from observations of 3.9 µm emissivity and sodar observations of cloud thickness off the west coast of the United States.  The product is not computed during twilight conditions when rapid changes in reflected solar radiation (either increases around sunrise or decreases around sunset).  The image above shows the GOES-R Cloud Thickness field over the midwest just before sunrise on 20 September 2016 (Radiation fog formed subsequent to late-afternoon and evening thunderstorms over Wisconsin and Illinois).  This scatterplot relates the last pre-sunrise value to dissipation time.  GOES-R Cloud thickness shows values over the Wisconsin River Valley in southwest Wisconsin, and over regions south of Military Ridge. Largest values — 1100 feet over Illinois and Iowa — suggest (from the scatterplot) a dissipation time of around 4 hours, which would be 1130 UTC (the time of the image) + 4 hours, or 1530 UTC.  There is also a region of thick clouds on northwest Indiana on the shore of Lake Michigan.  It’s these regions where you should expect large-scale fog/low clouds to dissipate last.   The animation below shows that to be true.  Fog over the river valleys is taking a bit longer to dissipate than expected, however. Note: navigation in the animation shows the effect of the loss of one star-tracker on GOES-13.

goes13_flswi_20sept2016_1245_1515anim

GOES-13 Visible (0.63 µm) animation, 1245-1515 UTC on 20 September 2016 (Click to enlarge)

The Day Night band on the VIIRS instrument on board Suomi NPP produces visible imagery at night that showed the regions of fog distinctly shortly after 0800 UTC on 20 September as shown below.

daynightband_wi_0827_20sep2016

VIIRS Day/Night Band Visible (0.70 µm) Imagery from Suomi NPP at 0827 UTC on 20 September (Click to enlarge)

Fog under High Clouds

FourPanel_1045UTC_18August2016

GOES-R IFR Probability Fields (Upper left), GOES-East Brightness Temperature Difference (3.9 – 10.7) (Upper Right), GOES-R Cloud Thickness (Lower Left) and GOES-East Water Vapor imagery (Lower Right), all at 1045 UTC on 18 August 2016. Surface observations of ceilings and visibilities at 1100 UTC are included in the upper right (Click to enlarge)

Dense Fog developed over southern Indiana on the morning of August 18 (and advisories were hoisted).  The single image above demonstrates an advantage of GOES-R IFR Probability fields in determining the areal extent of fog:  the traditional method of night-time fog detection from satellite fails in regions where cirrus clouds obscure the view of low clouds.  That was the case over the Ohio River Valley where IFR conditions were occurring.  GOES-R IFR Probability fields have a signal where high clouds exist in regions where Rapid Refresh model output shows low-level saturation, as over southwestern Indiana.  Because satellite data cannot be used there to compute IFR Probabilities, the magnitude of the probability is smaller.  Tailor your interpretation of the IFR Values based on the presence of high clouds.  The presence of high clouds changes the character of the IFR Probability field, from a pixelated field where satellite data are present to a flatter field where only model data can be used.

GOES-R Cloud Thickness can be used to estimate fog dissipation time (using this scatterplot, where the thickness values are from the last pre-sunrise scene).  That field, however, is only produced where the satellite has an unimpeded view of the low clouds (therefore, where cirrus clouds are present, as over the Ohio River Valley, Cloud Thickness is not produced).   Note the line parallel to the terminator over eastern Ohio:  GOES-R Cloud Thickness is not produced during twilight times around sunrise or sunset.  This 1045 UTC image is the final one over Indiana before sunrise.  Maximum thickness values are just over 1000 feet over southwest Indiana, suggesting a dissipation time of about three hours, that is, around 1345 UTC.

Changes in Model Fields show up in IFR Probability

GOES-R_IFR_1045_1100UTC_15August2016toggle

When GOES-R IFR Probability fields are governed solely by Rapid Refresh model output because of thick cloudiness (as was the case over Illinois on 15 August 2016), there can be changes in the field at the top of the hour that are related to changes in the Rapid Refresh model output — that is, changes in which hour Rapid Refresh Model is used.  The toggle above shows the IFR Probability fields at 1045 UTC and 1100 UTC on 15 August.  Both fields are characterized by smooth values that come with IFR Probability that is driven by Rapid Refresh model output, output that is smooth and not pixelated like satellite data.  It’s pretty noticeable, however, that values increase (from ~39% to ~52%) in those 15 minutes.  Why?

The image below shows Rapid Refresh Model Predictions of 1000-700 mb Relative Humidity at 1100 UTC from the 0800 UTC model run (that is, a 3-hour forecast, left) and from the 0900 UTC model run (that is, a 2-hour forecast, right).  Relative Humidity Values from the 0800 UTC Run (interpolated to 1045) are used in the computation of IFR Probabilities at 1045 UTC;  values from the 0900 UTC Run are used in the computation of IFR Probabilities at 1100 UTC.  It’s not this relative humidity field (value from 1000-700 hPa) precisely that is used, but rather maximum values in the vertical.  Certainly there are changes in the predicted low-level relative humidity field at 1100 UTC between the sequential model runs;  it’s more likely that saturation is occurring in the later model run, and that greater likelihood of saturation is reflected in the change of IFR Probability from 1045 UTC (when 0800 UTC Rapid Refresh Model fields are used) to 1100 UTC (when 0900 UTC Rapid Refresh Model fields are used).

RAP_RH_11UTC_15AUG2016

Rapid Refresh model predictions of 1000-700 mb Relative Humidity; 3-hour forecast from the 0800 UTC Rapid Refresh Model (left) and 2-hour forecast from the 0900 UTC Rapid Refresh Model (Right), both from 15 August 2016 (Click to enlarge)