Author Archives: Scott Lindstrom

Radiation fog over coastal North Carolina

GOES-R IFR Probabilities with observations of ceilings/visibilities (Upper Left), GOES-East Brightness Temperature Difference (10.7µm – 3.9µm) (Upper Right), GOES-R Cloud Thickness (Lower Left), Suomi/NPP VIIRS Day/Night Band (0.70 µm, at night) or GOES-East Visible imagery (0.63 µm, during day), Times as indicated (Click to enlarge)

Clear skies allowed for radiation fog with IFR conditions to develop over eastern North Carolina overnight. GOES-R Cloud Thickness as observed at the last time before twilight conditions begin — in this case at 1000 UTC (6 AM EDT) — can be used as a predictor for fog dissipation time using this chart. At 1000 UTC, maximum cloud thickness was 1000 feet, which suggests a dissipation time around 1300-1330 UTC.

The visible animation, below, is in agreement with the prediction from the cloud thickness field.

GOES-East Visible Imagery (0.62 µm) Times as indicated (Click to enlarge)

A toggle of two images with Day/Night band imagery, below, shows the difficulty in using the Day/Night band to identify regions of fog/low clouds consistently. In the 0615 UTC image, the developing low clouds show up well (Of course, it’s hard to tell if the clouds are low stratus or mid-level stratus), but the picture at 0745 UTC is a lot less distinct.

Radiation Fog in the Florida Panhandle

GOES-R IFR Probabilities computed from GOES-East (Upper Left), GOES-East Brightness Temperature Difference (10.7 µm – 3.9 µm) (Upper right), GOES-R Cloud Thickness (Lower Left), GOES-13 Heritage Cloud Thickness Product (Lower Right), times as indicated (Click to enlarge)

Clear skies and light winds over the Florida Panhandle allowed radiation fog to develop, and the GOES-R IFR Probability field ably captured the region of IFR conditions, as shown above. Note how the GOES-R Fields provide information that looks less ‘noisy’: the heritage products — Brightness Temperature Difference and Cloud Thickness — both have signals over interior Alabama and Mississippi where IFR conditions were not reported. In addition, the well-known co-registration error in the 10.7 and 3.9 sensors produces a faulty signal along the Florida coast east of Appalachee Bay. Eventually this signal because so strong that it bleeds into the GOES-R IFR probability.

Cloud thickness can be used to forecast fog dissipation time, using this chart and tempered by experience. The last pre-sunrise cloud thickness is shown at 1045 UTC in the loop above, and is around 1000 feet, suggesting a dissipation time around 1345 UTC. The visible animation below corroborates this estimate.

GOES-R Visible (0.63 µm), times as indicated (Click to enlarge)

Advection fog over Lake Michigan

GOES-R IFR Probabilities computed from GOES-East (Upper Left), GOES-East Brightness Temperature Differences (10.7 µm – 3.9 µm) (Upper Right), GOES-R Cloud Thickness (Lower Left), GOES-R IFR Probabilities computed from MODIS, or GOES-East Visible Imagery, times as indicated on 29 April 2014 (click to enlarge)

The GOES-R IFR Probability fields computed from GOES-East captured the onset of Lake fog that moved onshore over eastern Wisconsin on April 29th. Multiple cloud layers associated with a strong extratropical cyclone precluded the use of the brightness temperature difference product (the heritage method of detecting fog/low stratus). However, the IFR Probability field aligns well with the reductions in visibility associated with the Lake fog. The character of the IFR Probability field can be used to infer whether of not satellite data predictors are being used. For example, the relatively flat field over southeast Wisconsin at the start of the animation is a region where satellite predictors are not used. The use of satellite predictors generally leads to a pixelated field. A flatter field as over southeast Wisconsin reflects the smoother model fields that are driving the probability field computation.

Cloud thickness is computed in regions where the highest cloud, as seen by the satellite, is a water-based cloud. And that is also usually the region where satellite predictors are used in the computation of IFR Probabilities. Note in the animation above how cloud thickness generally overlays regions of IFR Probability that are pixelated. Cloud thickness is not computed where only model data are used to compute IFR Probabilities. (Cloud thickness is also not computed in the hour or so around sunrise and sunset, during twilight conditions).

The slow northward movement of the fog bank is apparent in the first part of the animation above, from 0615 through 0745 UTC. Note also how the MODIS IFR Probability fields give a very similar solution to the GOES-13-based fields at 0745 UTC. Differences in resolution are apparent over southwest Wisconsin, however, where river valleys are more accurately captured by the MODIS fields.

In the visible imagery at the end of the animation (1355 UTC), the rapid saturation of moisture-laden air moving northward from Indiana over the cold waters of southern Lake Michigan is very apparent.

Fog Dissipation in southern Alabama

GOES-13 Visible Imagery (0.63 µm), Times as Indicated (click to enlarge)

GOES-R Fog/Low Stratus products can be used to estimate the time of fog dissipation, an example of which dissipation from 23 April 2014 is shown above (complete with a temporary hole in the subsequent cumulus development). Complete dissipation of the radiation fog occurred by 1632 UTC. This chart is a scatterplot of GOES-R Cloud Thickness in the last pre-sunrise image vs. dissipation time (measured as hours after that pre-sunrise image). The 1100 UTC GOES-R Cloud Thickness is shown below; it is the last pre-sunrise image over Alabama (note that no values are present over Georgia because twilight conditions are present there). GOES-R Cloud Thickness values in the foggy region (where IFR Probabilities are high, and where ceilings are low and visibilities obstructed) are around 1000 feet. Values from the best-fit line at this link suggest, then, a dissipation time of a little over 3 hours, but the spread of values shown in the scatterplot is from less than two hours to more than four. In this present example, fog dissipated after about five hours. IFR conditions ended after about three hours.

GOES-R IFR Probabilities computed from GOES-East (Upper Left), GOES-East Brightness Temperature Differences (10.7 µm – 3.9 µm) (Upper Right), GOES-R Cloud Thickness (Lower Left), GOES-13 Heritage Low Cloud Base Product (Lower Right), 1102 UTC on 23 April 2014 (click to enlarge)

Fog in Southwestern Louisiana

Dense fog developed along the I-10 corridor over southern Louisiana during the early morning hours of 21 April 2014. From the AFDs issued by Lake Charles:

000
FXUS64 KLCH 210153
AFDLCH

AREA FORECAST DISCUSSION
NATIONAL WEATHER SERVICE LAKE CHARLES LA
853 PM CDT SUN APR 20 2014

.UPDATE…A WEAK TROF WILL APPROACH FROM THE WEST TONIGHT,
BRINGING A LIGHT SOUTHERLY FLOW AND SOME MOISTURE NEAR THE SURFACE.
THIS MOISTURE SHOULD BE SUFFICIENT FOR THE FORMATION OF PATCHY
GROUND FOG LATER TNITE…WHICH IS ALREADY IN THE FORECAST.

The 319 AM CDT Forecast Discussion noted the increase in IFR Probabilities:

000
FXUS64 KLCH 210819
AFDLCH

AREA FORECAST DISCUSSION
NATIONAL WEATHER SERVICE LAKE CHARLES LA
319 AM CDT MON APR 21 2014

.DISCUSSION…
MID AND HIGH LEVEL CLOUDS STREAMING ACROSS INTERIOR SOUTHEAST
TEXAS…OTHERWISE…TEMPERATURES MAINLY IN THE UPPER 50S TO LOWER 60S
AREAWIDE. CLEAR SKIES AND LIGHT WINDS SETTING UP AGAIN FOR FOG
DEVELOPMENT MAINLY ALONG THE I-10 CORRIDOR FROM BEAUMONT EAST TO
SHORT OF LAFAYETTE LOUISIANA. GOES-E/MODIS MVFR PRODUCT SHOWING
INCREASING PROBABILITY OF LOW VISIBILITY FOR DEVELOPING IN THE
BPT AREA WHICH LATEST OBSERVATION CONFIRM THAT TREND…AND ANOTHER
AREA FROM LAKE CHARLES TO NEAR LAFAYETTE. THEREFORE…ASKING FOLKS
TO DRIVE CAREFULLY THIS MORNING IF YOUR DRIVING IN THESE AREAS.

Shortly after sunrise, a Dense Fog Advisory was issued:

000
FXUS64 KLCH 211150
AFDLCH

AREA FORECAST DISCUSSION
NATIONAL WEATHER SERVICE LAKE CHARLES LA
650 AM CDT MON APR 21 2014

.UPDATE…
ISSUED DENSE FOG ADVISORY FOR THE I-10 CORRIDOR AREA FROM
SOUTHEAST TEXAS TO LAKE CHARLES TO LAFAYETTE. MAIN CONCERN WAS
LIGHT WINDS AND STRONG RADIATIONAL COOLING WHICH HAS RESULTED IN
A SHARP DROP IN VISIBILITIES OVER A SHORT TIME PERIOD. DENSE FOG
MAY BOUNCE UP AND DOWN A BIT DURING THE ADVISORY TIMES. DENSE FOG
SHOULD DISIPATE BY 9 AM.

How did the IFR probability forecasts do during this event? At 0400 UTC (below), neither the IFR probabilities nor the traditional method of fog/low cloud detection suggest fog/low clouds are present.

GOES-R IFR Probabilities computed from GOES-East (Upper Left), GOES-East Brightness Temperature Differences (10.7 µm – 3.9 µm) (Upper Right), GOES-R Cloud Thickness (Lower Left), Suomi/NPP VIIRS Brightness Temperature Difference (11 µm – 3.74 µm) (Lower Right), 0400 UTC on 21 April 2014 (click to enlarge) (click to enlarge)

By 0600 UTC, IFR Probabilities have increased to around 10% in/around Lake Charles as visibilities have dropped. GOES-R Cloud Thickness values are around 600 feet.

As above, but at 0615 UTC on 21 April 2014 (click to enlarge)

Just before 0800 UTC, a Suomi/NPP overpass provided high-resolution data. Neither the GOES-East nor the VIIRS data brightness temperature difference products show a distinct fog/stratus signal over southwestern Louisiana, where GOES-R Cloud Thickness values persist at around 600 feet, and where IFR probabilities have increased past 50%.

As above, but at 0745 UTC on 21 April 2014 (click to enlarge)

By 1100 UTC (below), when the Dense Fog Advisory is issued (hours after the IFR Proabilities first increased), the Brightness Temperature Difference product from GOES-East is finally showing a faint (albeit noisy) signal of fog/low stratus over southwestern Louisiana. A stronger signal extends northeast and southwest from Baton Rouge.

As above, but at 1100 UTC on 21 April 2014 (click to enlarge)

Fog/Low Stratus and Stratus behind a late-season Snow Storm

A late-season snowstorm moved through the upper midwest on the 16th/17th of April, dropping up to 18″ of snow over northern Minnesota and northern Wisconsin. The storm and its aftermath produced IFR conditions; how did the GOES-R IFR Probability field do?

GOES-R IFR Probabilities computed from GOES-East (Upper Left), GOES-East Brightness Temperature Differences (10.7 µm – 3.9 µm) (Upper Right), GOES-R Cloud Thickness (Lower Left), Toggle between GOES-R IFR Probabilities computed from MODIS and MODIS Brightness Temperature Difference Fields (Lower Right), times as indicated (click to enlarge)

The toggle above shows data from ~0445 UTC on 17 April. The MODIS Brightness Temperature Difference field suggests stratus over a large region where IFR conditions are not observed; the IFR Probability fields better approximate the regions of low ceilings/reduced visibilities. The inclusion of Rapid Refresh model data better defines regions of low-level saturation so that the IFR Probability Fields (compared to the brightness temperature difference fields) are better aligned with low ceilings/reduced visibilities. Imagery at ~0900 UTC, below, shows an expansion of the IFR conditions over southern Minnesota as the storm moves away. Note also how IFR Probabiilties are enhanced over northern Wisconsin, where multiple cloud layers make the traditional fog/low cloud detection mechanism, the brightness temperature difference field, difficult. Probabilities are smaller there because the satellite predictors cannot be used in the algorithm that computes IFR Probabilities, and the variability of the values is less, reflecting the smoother fields present in the model (compared to the pixels of the satellite data).

GOES-R IFR Probabilities computed from GOES-East (Upper Left), GOES-East Brightness Temperature Differences (10.7 µm – 3.9 µm) (Upper Right), GOES-R Cloud Thickness (Lower Left), Toggle between GOES-R IFR Probabilities computed from MODIS and MODIS Brightness Temperature Difference Fields (Lower Right), times as indicated (click to enlarge)

Fog/Low Stratus near the Gulf of Maine

GOES-R IFR Probabilities computed from GOES-13 and surface plots of ceilings/visibilities, times as indicated (click to enlarge)

When relatively high dewpoints move over the cold waters of the Gulf of Maine in Spring, advection fog can form. Sometimes this happens underneath clear skies, sometimes it happens underneath high clouds. In both cases, the IFR Probability field should give a reasonable answer — but how can that prediction be validated? In the example above, the apparent fog/low cloud bank propagates off to the east, and when it is fully ashore near Yarmouth, NS, visibilities drop from near-IFR to IFR conditions. The IFR probability field has an appearance that suggests plenty of high clouds are overlaying the visibility-restricting lower clouds, yet a consistent signal of higher probabilities of IFR conditions is maintained in/around the Gulf of Maine northward into Maine.

Fog vs. Stratus over southern California

Brightness temperature difference fields over the Pacific Ocean offshore of southern California showed a solid field of clouds overnight. How can this information about the top of the cloud be used to predict where low clouds and fog (IFR conditions) might exist? If you blend the satellite predictors with predictors from the Rapid Refresh model, you have information about the presence of clouds (the satellite predictors) and about the likelihood of saturation in the lowest kilometer of the model atmosphere. Consider the example below from 0400 UTC on 25 March 2014.

GOES-R IFR Probabilities computed from GOES-15 (Upper Left), GOES-West Brightness Temperature Differences (10.7 µm – 3.9 µm) (Upper Right), GOES-R Cloud Thickness (Lower Left), GOES-R IFR Probabilities computed from MODIS (Lower Right), all near 0400 UTC on 25 March 2014 (click to enlarge)

Brightness temperature differences suggest a deck of stratus over most of the Gulf of Santa Catalina (and over the Channel Islands as well). The IFR Probability field, however, suggests that the stratus just offshore of southern California, between the mainland and the Islands, is not reducing surface visibilities, and IFR conditions do not exist right along the coast of the mainland. The station reporting IFR conditions, KAVX, Catalina Airport on Catalina Island, is 1600′ above sea level.

By 1000 UTC, below, visibilities have lowered (and IFR Probabilities have increased) near Vandenberg AFB north of Point Conception. IFR conditions persist over Catalina Island (demonstrating the importance of knowing the elevation of the stations!).

As above, but at 1000 UTC on 25 March 2014 (click to enlarge)

Stratus vs. Fog in the upper Midwest

GOES-R IFR Probabilities computed from GOES-13 (Upper Left), GOES-East Brightness Temperature Differences (10.7 µm – 3.9 µm) (Upper Right), GOES-R Cloud Thickness (Lower Left), Suomi-NPP Brightness Temperature (Lower Right), all near 0200 UTC on 20 March 2014 (click to enlarge)

Low clouds lingered over the upper midwest behind a departing low pressure system late on Wednesday the 19th. A strong signal was evident in the brightness temperature difference field from GOES-East, above, from 0200 UTC, extending northwest to southeast over eastern Minnesota into northern Indiana. Note, however, that ceilings in this region were indicative of mid-level stratus rather than fog. IFR Probabilities are correctly very small underneath this stratus.

GOES-R IFR Probabilities computed from GOES-13 (Upper Left), GOES-East Brightness Temperature Differences (10.7 µm – 3.9 µm) (Upper Right), GOES-R Cloud Thickness (Lower Left), MODIS-based IFR Probabilities and Suomi-NPP Day/Night Band (Lower Right), times as indicated (click to enlarge)

An animation of the fields, above, shows the development of a low IFR conditions over western Minnesota. The brightness temperature difference fields also show their development, and the combination of satellite predictors and model predictors lead to very high IFR Probabilities in that region, both in the GOES-based fields, shown half-hourly, and in the MODIS-based fields, shown when available.

Suomi/NPP Day/Night band and brightness temperature difference field, 0744 UTC on 20 March 2014 (click to enlarge)

The near-full Moon provided ample illumination for the clouds, and the day/night band reveals the extensive cloud cover over the upper midwest, but as it only shows the top of the clouds, it is difficult to determine if visibility restrictions are also present. The Brightness temperature difference produce is also shown, which field is helpful in screening out snow cover and city lights.

Low clouds and Fog along the West Coast

Low clouds and fog developed along the west coast this morning. From the Monterey (CA) AFD:

FXUS66 KMTR 141143
AFDMTR

AREA FORECAST DISCUSSION
NATIONAL WEATHER SERVICE SAN FRANCISCO BAY AREA
443 AM PDT FRI MAR 14 2014

.SYNOPSIS…AFTER A BIT OF COOLING TODAY…A WARM AND DRY UPPER LEVEL
RIDGE OF HIGH PRESSURE WILL BUILD STRONGLY INTO THE WEST COAST
OVER THE WEEKEND. THIS WILL RESULT IN AFTERNOON TEMPERATURES
REACHING WELL ABOVE SEASONAL NORMS…AND POSSIBLY TO NEAR RECORD
LEVELS FOR THESE DATES. THIS WARM-UP WILL BE SHORT-LIVED HOWEVER…WITH
INCREASED ONSHORE FLOW AND A SIGNIFICANTLY COOLER AIR MASS MOVING
IN ALOFT THE FIRST PART OF NEXT WEEK. DRY CONDITIONS ARE EXPECTED
TO CONTINUE THROUGH MID WEEK…BUT THEN WITH UPPER LEVEL TROUGHING
AND A CHANCE OF RAIN FOR THE OUTER PORTION OF THE FORECAST PERIOD.

.DISCUSSION…AS OF 4:10 AM PDT FRIDAY…THE DRY TAIL END OF A
WEATHER SYSTEM MOVING IN TO THE PACIFIC NORTHWEST IS APPROACHING
OUR DISTRICT…AND RESULTING IN ENHANCEMENT OF THE MARINE LAYER
AND A RETURN OF THE MARINE STRATUS. LATEST GOES FOG PRODUCT
IMAGERY…AND IN RATHER SPECTACULAR DETAIL JUST REC’D SUOMI VIIRS
NIGHTTIME HIGH RES VISUAL IMAGE…SHOW COVERAGE ALONG MUCH OF THE
COAST FROM PT REYES SOUTH TO THE VICINITY OF THE MONTEREY
PENINSULA…AND A BROAD SWATH EXTENDING INLAND ACROSS SAN
FRANCISCO AND THROUGH THE GOLDEN GATE TO THE EAST BAY. LATEST
BODEGA BAY AND FT ORD PROFILER DATA INDICATE A MARINE LAYER DEPTH
OF ABOUT 1300 FT. SOME THIN HIGH CLOUDS ARE ALSO PASSING THROUGH ABOVE.

NAM MODEL AND IN-HOUSE LOCAL WRF MODEL BOUNDARY LAYER RH OUTPUT
BOTH INDICATE STRATUS SHOULD GENERALLY CLEAR BY MIDDAY…EXCEPT
ALONG THE SAN MATEO COAST AND IN THE VICINITY OF THE MONTEREY
PENINSULA. EXPECT AFTERNOON HIGHS TO BE AROUND 3 TO 5 DEGS COOLER
THAN ON THURSDAY…BUT STILL WELL ABOVE SEASONAL NORMS ESPECIALLY
INLAND.

GOES-R IFR Probabilities computed from GOES-15 (Upper Left), GOES-West Brightness Temperature Differences (10.7 µm – 3.9 µm) (Upper Right), GOES-R Cloud Thickness (Lower Left), Suomi/NPP Day/Night Band (Lower Right), all hourly times as indicated (click to enlarge)

The animation of satellite and satellite-derived fields, above, shows how the GOES-R and GOES-West fields depicted the development of the low clouds. Note how the brightness temperature difference fields over CA and NV throughout the animation have a speckled appearance. These positive signals are due not to the presence of fog/low clouds but rather to differences in emissivity properties of the dry land. Near the end of the animation, high clouds are widespread over northern California. For such cases, the brightness temperature difference product provides little information about low-level clouds. However, the GOES-R IFR Probability field, because it blends together information from satellite and from Rapid Refresh does provide a signal under clouds. It is a much smoother signal because it does vary from one satellite pixel to the next, and the Probability values are smaller because satellite predictors cannot be used in the algorithm.

The AFD above notes the Day/Night band, and also the depth of the marine stratus. The toggle of Cloud Thickness, Day/Night Band, and brightness temperature difference, below (useful to distinguish white clouds from white city lights!), shows a nice overlap between the GOES-R product and the clouds detected at high resolution by Suomi/NPP. Cloud thickness is around 1150 feet at Bodega Bay, and closer to 1250 feet at Fort Ord, in good agreement with the profile data cited.

VIIRS_DNB__REF_GOES_CLD_THICK_20140314_09

Toggle of GOES-R Cloud Thickness, Suomi/NPP Brightness Temperature Difference and Day/Night Band (click to enlarge)

GOES-R Fog Product Examples

Fog detection fusing GOES, Terra/Aqua or Suomi/NPP Satellite data with Model output