Category Archives: MODIS

GOES-R Probabilities evolve as the night progresses

Leave a reply
GOES-R IFR Probabilities computed from GOES-14 (upper left), GOES-14 heritage brightness temperature difference product (upper right), VIIRS heritage brightness temperature difference product (lower left), MODIS heritage brightness temperature difference product (lower right), all from near 0700 UTC.

MODIS and VIIRS yield satellite information that can be used to detect fog/low stratus at high spatial resolution.  In the exa,ple above, the VIIRS brightness temperature difference product highlights many river valleys as possibly cloud-filled over eastern Kentucky, southern Ohio and southwestern West Virginia.  The coarser resolution of the GOES satellite pixel precludes such fine-scale detection.  Note, however, that both satellite platforms detect the presence of stratiform water clouds over north-central Ohio where surface observations show only mid-level cloudiness.  IFR probabilities are confined to the spine of the Appalachians from the Laurel Mountains near Johnstown (PA) southward towards southern West Virginia.  How do things evolve with time?

As above, but from near 0800 UTC

As above, but from near 0915 UTC

As above, but from near 1000 UTC

As above, but from near 1100 UTC

The power of GOES imagery in this case is to show the evolution of the fog/low stratus field.  Even at this only-hourly timestep, the development of regions of IFR conditions is evident, and those developing conditions occur in tandem with increasing probabilities in the GOES-R IFR Probability field.  Throughout the night, the GOES brightness temperature difference field flags the unimportant (from an aviation standpoint) stratus deck over northcentral Ohio, and the IFR probability field, which field also uses Rapid Refresh Model data, discounts the satellite signal.  By 1100 UTC, the river valley signal has strengthened enough in the GOES imagery to appear, and a corresponding increase in IFR probability occurs.

IFR Probability can also be computed using MODIS data (below). The 0739 UTC MODIS data, shown at the top of this blog post, highlights — as does GOES — the stratus deck over northern Ohio.  The MODIS-based IFR probabilities, however, do not highlight that cloud-deck, by design.  Note also that the higher-resolution MODIS imagery, because it detects river valley fogs at 0739 UTC, also has a strong IFR probability signal there.  Pixel resolution on GOES-R will be intermediate between MODIS and present GOES.

As at top, for near 0800 UTC, but with MODIS-based IFR probabilities in the lower right

Difficulties detecting FLS over arid regions using satellite alone.

Leave a reply
GOES-R IFR probabilities (top left) and the traditional 3.9-11 micron brightness temperature difference (top right) from GOES-West on October 2, 2012 at 13:30 UTC. The blue circles are surface observations with the surface visibility (in miles) below. 
The maximum relative humidity in the lowest 1000ft layer above ground level from the Rapid Refresh forecast model valid at 13:15 UTC on October 2, 2012.
The traditional 3.9-11 micron brightness temperature difference (BTD) used to detect liquid stratus clouds exploits the emissivity differences at the 3.9 and 11 micron wavelengths. Liquid water clouds typically emit less at 3.9 microns than at 11 microns and this difference can be used to differentiate them from other types of clouds. This method works well when the liquid stratus clouds are the highest cloud layer and are distinguishable from the background surface characteristics. However, fog/low stratus (FLS) detection over arid regions can be difficult using satellite data alone because the very dry land surface typically has a low surface emissivity at the 3.9 micron wavelength as well, resulting in a satellite signal very similar to that of liquid stratus clouds. This is shown in the top right image above where the arid regions of the SW U.S. have a similar BTD signal as the status clouds offshore California even though all the surface observations in the area indicate no clouds are present. 
The GOES-R IFR product (top left image above) can greatly reduce the false satellite signal caused by the land surface by incorporating model relative humidity (RH) data. For the GOES-R IFR product the satellite data is combined with the maximum RH found in the lowest 1000 ft layer above ground level from the Rapid Refresh RH profiles using a naive bayesian probabilistic model. FLS generally occurs when the low-level RH is relatively high (> 80%). The low level RH in arid regions is usually very low (< 30%), so even when the satellite signal is strong, indicating FLS may be present,  the weak model signal works to lower the probability that FLS is present. 
However, sometimes the low-level RH can be elevated over dry land surfaces when clouds are not present. When this occurs, the relatively high model signal along with the relatively high satellite signal (from the land surface) indicate there is a greater chance that FLS may be present so the GOES-R IFR probabilities are slightly elevated. This is shown in the top left image above where an area of elevated GOES-R IFR probabilities in central California indicate that there is a roughly 40% chance that IFR conditions are present even though the surface stations report clear skies. The low-level modeled RH from the Rapid Refresh model (bottom image above) indicates a sliver of elevated RH (> 80%) in central California. This area of elevated low-level RH combined with the relatively higher satellite signal (from the land surface) resulted in the GOES-R IFR probabilities being slightly elevated. GOES-R IFR probabilities around 40% are considered to be relatively low and, combined with the surrounding surface observations, forecasters should feel pretty confident that this is indeed a false signal and that FLS is not present.

GOES-R IFR probabilities (left) and the traditional 11-3.9 micron brightness temperature difference (right) from MODIS on October 2, 2012 at 9:05 UTC. The blue circles are surface observations with the surface visibility (in miles) below.

The higher spatial resolution (1 km) MODIS instrument shows in more detail the elevated traditional BTD satellite signal caused by the arid land surface (right image) over the SW U.S. The GOES-R IFR probabilities applied to MODIS (left image) again show how the GOES-R IFR product significantly reduces the false FLS signal that is seen in the traditional BTD product.

GOES-13 outage, GOES-R IFR products still produced with GOES-15

Leave a reply
GOES-R IFR Probabilities computed with GOES-West during the morning of Sep 24 2012

GOES-13 was placed in standby after experiencing an anomaly on Sunday, so all GOES-R IFR Probabilities are being computed with GOES-West data;  GOES-15 is in FD mode, acquiring a full disk each half hour to provide coverage for CONUS.  GOES-15 data did detect the development of fog/low stratus over southern/coastal Texas and Louisiana overnight under the presence of light winds and high dewpoints.  The regions of highest probabilities correspond well with IFR conditions.  The 1330 UTC image, at the end, shows a noticeable drop in coverage as the daytime predictors replace the nighttime predictors (this includes using the visible imagery as a cloud mask).

Resolution Differences between GOES and MODIS GOES-R IFR products

Leave a reply
Toggle between GOES-R IFR Probabilities computed with GOES-West (at 1200 UTC) and with MODIS (1154 UTC) over the northern Gulf of Alaska and surrounding landmass

In regions where very small-scale terrain effects low cloud distribution — for example, in river valleys — the superior resolution of MODIS (on Terra or Aqua) or VIIRS (on Suomi NPP) allows a much better horizontal definition of cloud boundaries.  In regions of larger scale variability, such as over the ocean, differences between GOES and MODIS IFR estimates aren’t quite so noticeable, as in the imagery above.  Both fields show a large region of low stratus over the northern Gulf of Alaska moving inland.  The regions of lower ceilings and reduced visibilities match the stream of higher IFR probabilities moving northward over the Gulf, from Middlteon Is (PAMD) on the western edge to Sitka (PASI) and Petersburg (PAPG) on the southern/eastern edge.  Note also the break in high IFR probabilities where the Aleutian Peninsula attaches to the mainland, with higher probabilities, and IFR conditions, at PASV (Sparrevohn).

The character of the IFR probability field should alert you to the presence of a stream of high clouds over the western third of this image.  In that region, only model values are being used to compute IFR probabilities, and the total probabilities are therefore lower.

Fog and low Stratus Chances near Philadelphia

Leave a reply

The National Weather Service in Philadelphia/Mt. Holly NJ noted the possibility of fog formation in the moist airmass that supported showers and thundershowers over Pennsylvania on late Thursday.  From the forecast discussion issued at 0059 UTC on 7 September:

000
FXUS61 KPHI 070059
AFDPHI

AREA FORECAST DISCUSSION
NATIONAL WEATHER SERVICE MOUNT HOLLY NJ
859 PM EDT THU SEP 6 2012 
.... 
 
.NEAR TERM /UNTIL 6 AM FRIDAY MORNING/...

-- Changed Discussion --

DENSE GROUND FOG CONTINUES TO BE POTENTIAL PROBLEM OF THE NIGHT.
PUBLIC PRODUCTS UPDATED AT 555PM HAVE INCLUDED PATCHY DENSE FOG
IN MANY OF THE ZONES AND THE SHOWERS HAVE ENDED IN DE.

LIGHT WIND...CLEARING SKIES IN THE WAKE OF THE SHOWERS/SPRINKLES
EARLIER TODAY AND DEWPOINTS IN THE UPPER 60S TO AROUND 70 IN WHAT
SHOULD BE A MOSTLY CLEAR - NO WIND NIGHT WITH WEAK HIGH PRES IN PA
SPELLS TROUBLE...ESPECIALLY SINCE PATCHY LOW CIGS HAVE DEVELOPED
VCNTY KACY SINCE ABOUT 19Z...FIRST ALERT TO A BIGGER FOG/LOW CIG
PROBLEM.

GEOCAT SATELLITE AND SREF PROBS ARE NOT INDICATING MUCH FOG/LOW
CIG CHANCE AND MOST OF THE MET AND MAVMOS GUIDANCE IS NOT INTERESTED
IN FOG/LOW CIGS...YET THE MASS FIELDS OF BL AND SFC RH SEEM TO BE
MORE PLAUSIBLE. WE SHOULD KNOW BY 02Z AS A FEW OF THE TEMP/DEWS
START GETTING TO 100 PCT RH IN S NJ/DE. 
 
 
The possibility of fog remained a concern through the night, as noted in the forecast discussion from 0747 UTC:
000
FXUS61 KPHI 070747
AFDPHI

AREA FORECAST DISCUSSION
NATIONAL WEATHER SERVICE MOUNT HOLLY NJ
347 AM EDT FRI SEP 7 2012

........

-- End Changed Discussion --

&&

.NEAR TERM /UNTIL 6 PM THIS EVENING/...

-- Changed Discussion --

.....
WORTHWHILE FOG HAS NOT BEEN DEVELOPING OVERNIGHT. STEPPING OUTSIDE
WHILE IT IS CONSIDERABLY MURKIER THAN LAST NIGHT, ITS AS IF THE DEW
DEPOSITION (WHICH IS QUITE HEAVIER) IS TAKING AWAY FROM THE FOG.
GEOCAT IFR PROBS REMAIN LOW. WE STILL HAVE A COUPLE OF MORE HOURS
TO GO, WE WILL JUST KEEP THE MENTION OF SOME PATCHY/AREAS OF FOG IN
THE GRIDS, BUT STOP THERE. 
 
The GOES-R IFR Probability product, shown below at hourly intervals starting at 0315 UTC,  shows modest probabilities of fog over the Delaware Valley,  and higher probabilities over the Pine Barrens of central New Jersey, justifying the forecast mention of patchy fog.  
GOES-R IFR Probabilities from GOES-East (Upper Left), Traditional Brightness Temperature Difference Product (Upper Right), enhanced window channel (Lower Left), GOES-R IFR Probabilities from MODIS (Lower left)
MODIS data can be used to compute IFR probabilities.  One comparison scene exists at the beginning of the hourly loop, and the finer detail over the valleys of southeast Pennsylvania is evident in the 0322 UTC image.  The MODIS image at 0739 UTC shows (below) similar fine-scale structures that can help a forecaster fine-tune the forecast.  IFR probabilities in the MODIS images definitely jump over central New Jersey between 0322 UTC and 0733 UTC.  Such a notable change is a tool for a forecaster to be alert to the possibility of fog/low stratus and IFR conditions.  Several stations in central New Jersey, including McGuire Air Force base (KWRI), Belmar/Farmingdale (KBLM) and Mill Valley (KMIV) did report borderline IFR conditions between 0700 and 1000 UTC.
GOES-R IFR Probabilities from GOES-East (Upper Left), Traditional Brightness Temperature Difference Product (Upper Right), enhanced window channel (Lower Left), GOES-R IFR Probabilities from MODIS (Lower left) from ca. 0730 UTC

IFR in Alaska when a Large-Scale weather system is present

Leave a reply
Animation of 1400 UTC Water vapor imagery, the 10.7 micron infrared image, the brightness temperature difference (10.7 – 3.9), the GOES-R IFR Probabilities computed from GOES data, the GOES-R IFR Probabilities computed from MODIS data, and the surface observations/ceilings.

The loop above cycles through the 1400 UTC Water vapor imagery, the 10.7 micron infrared image, the brightness temperature difference (10.7 – 3.9), the GOES-R IFR Probabilities computed from GOES data, the GOES-R IFR Probabilities computed from MODIS data, and the surface observations/ceilings.  The complex large-scale weather system over northwest Alaska is means that southerly winds over eastern Alaska are drawing moisture and cloudiness northward from the Gulf of Alaska.  Multiple cloud layers in this moist flow means that the traditional method of fog/low stratus detection (the brightness temperature difference between 10.7 and 3.9 micrometers) will be challenged.  Furthermore, on this particular day, IFR conditions (the observation map is below;  stations with IFR conditions are circled in red) are most frequent underneath the multiple cloud layers in the eastern part of the state, and at high levels, such as in the Brooks Range.

The GOES-R IFR probability field suggests higher possibilities of IFR conditions in regions where IFR conditions are observed:  near Anchorage, on the Aleutian peninsula and in the Brooks Range.

Observations over Alaska at 1500 UTC 31 August.  IFR conditions highlighted by red circles.

IFR over SW Alaska

Leave a reply
GOES-R IFR Probabilities (upper left) computed from GOES-West, GOES-R IFR Probabilities computed from MODIS (upper right), Visible Imagery (bottom left), Topography (bottom right)

GOES-R probabilities are a fused product between satellite data and the Rapid Refresh model.  Model data are used only where multiple cloud layers are present and or where a single cirrus cloud level exists.  The character of the IFR probability field looks different when model data only is used.  IFR probabilities are lower when only model data are used.

IFR probabilities are well related to observations at Kodiak, for example.  As the higher probabilities increase from the southwest, ceilings lower, and eventually IFR conditions occur.  The better resolution of the MODIS imagery, below, allows far finer-scale structures to be resolved in the imagery.

GOES-R IFR Probabilities (upper left) computed from GOES-West, GOES-R IFR Probabilities computed from MODIS (upper right), Visible Imagery (bottom left), Topography (bottom right)

Note how the smaller probabilities are downwind of the Aleutians.  Visible imagery — at the end of the animation — distinctly shows the clear region.

Excellent example of the importance of model data

Leave a reply
GOES-R IFR Probabilities computed using GOES-East data (Upper Left), GOES-R IFR Probabilities computed using MODIS data (Upper Right), Surface Observations and Cloud Ceilings Above Ground level (Lower Left), Suomi-NPP VIIRS Brightness Temperature Difference field, 10.8  µm – 3.74 µm (Lower Right).  Times as indicated.

Three different satellite sensors — the GOES Imager on GOES-East, MODIS on Aqua, and VIIRS on Suomi/NPP — viewed data from the occurrence of Valley Fog over the Appalachians (and surroundings) early in the morning of 21 August.  A shortcoming of the Brightness Temperature Difference field in the lower left is immediately apparent:  no fog/low stratus is indicated where high clouds exist, even though observations do show IFR conditions.  In contrast, the fused product does show heightened probabilities underneath that high cloud deck.  Probabilities are not as high as they are where both satellite and model predictors can be used to evaluate the presence of fog and low stratus, and the resolution of the field is different, obviously limited to the horizontal resolution of the Rapid Refresh, meaning that small river valleys, that are very obvious in the regions where satellite data are used (and even much more obvious when high-resolution MODIS or VIIRS data are used).  Note also how GOES-R IFR Probabilities de-emphasize the signal over western Ohio, where IFR conditions are not reported.  Brightness temperature difference fields from MODIS and from GOES both see a signal there, as also shown in the VIIRS field, but these stratus clouds are not obstructing visibility.

Bottom line:  MODIS data’s higher resolution observes the big differences between river valleys and adjacent cloud-free ridge tops.  GOES-East has difficulty in resolving those differences.  So MODIS IFR fields better highlight river fog.  Model data can help discern between fog on the ground, and stratus that is off the ground.

The Challenge of Identifying Fog in River Valleys

Leave a reply
Animation of GOES-R IFR probabilities computed from GOES-East (upper left), Brightness Temperature difference (11 – 3.9) from GOES-East (upper right), Surface visibilities and ceilings and GOES-East Visible imagery (lower left), GOES-R IFR probabilities computed from MODIS (lower right)

Radiation fog that forms first — or only — in river valleys is a challenge to detect.  In the example above from Pennsylvania for the morning of August 8 2012, the satellite signal starts to appear over the West Branch of the Susquehanna around 0500 UTC.  It is very unlikely that a numerical model with a horizontal resolution of (only) 13 km — such as the Rapid Refresh — will be able to forecast the development of such a small-scale feature, so the satellite observations are key.  Nominal 4-km resolution of the infrared channels is the principle limiting factor in detection.

The presence of high clouds over southeast Pennsylvania precludes the detection by satellite of low clouds/fog, so the GOES-R IFR product there is driven primarily by model output, and the scale of the river fogs are simply too small to be simulated in the model.  Fog probabilities do increase near Selinsgrove (KSEG) around daybreak because the model relative humidity does reach high enough values.  This also happens over southeast Pennsylvania.  Typically, model relative humidities in the lowest kilometer of the model must be greater than 80-85% for a strong GOES-R IFR signal to be present.

GOES-R IFR probabilities computed from GOES-East (upper left), Brightness Temperature difference (11 – 3.9) from GOES-East (upper right), Surface visibilities and ceilings and GOES-East Visible imagery (lower left), GOES-R IFR probabilities computed from MODIS (lower right), all valid at 0700-0715 UTC 8 August 2012

Note that at 0715 UTC there is a comparison between the better resolution of the MODIS imagery.  When GOES-R is operational, resolution will be in between that of current GOES and MODIS.  The MODIS GOES-R IFR probabilities are much higher, and show the different river valleys far more cleanly than present GOES.

GOES-R Probabilities are too low: Why?

Leave a reply
GOES-R IFR Probabilities (upper left), GOES-East Visible imagery (upper right), Brightness temperature difference between 10.7 and 3.9 micrometers (lower left), GOES-R Cloud Thickness (lower right)

The imagery above shows high IFR probabilities over western Massachusetts — where IFR conditions are not observed — and very low probabilities in central Massachusetts in and near the Connecticut River Valley where IFR conditions are observed.  The brightness temperature difference in central Massachusetts is not suggestive of low clouds and fog.  For the IFR probability to be high there, then, would require that the Rapid Refresh Model showed saturation in the lower part of the model atmosphere.  Thus, the fused product could show higher probabilities in this region where fog is observed.  However, as shown below, relative humidity in the lowest part of the model was actually a relative minimum over central Massachusetts.

Two-hour forecasts of Relative Humidity from the Rapid Refresh, all valid at 0600 UTC from the 0400 UTC model run;  Lowest 30 mb of the model (upper left), lowest 60 mb of the model (upper right), lower 90 mb of the model (lower left), surface (lower right)
Two-hour forecasts of Relative Humidity from the Rapid Refresh, all valid at 0800 UTC from the 0600 UTC model run;  Lowest 30 mb of the model (upper left), lowest 60 mb of the model (upper right), lower 90 mb of the model (lower left), surface (lower right)

By 0730 UTC, the satellite brightness temperature difference product (below) is starting to suggest that fog/low clouds are more widespread (A MODIS image at the same time tells the same tale).  As that happens, the IFR probabilities start to increase.

GOES-R IFR Probabilities computed from GOES-East imager data (upper left), GOES-R IFR probabilities computed from MODIS data (upper right), Brightness temperature difference between 10.7 and 3.9 micrometers (lower left), GOES-R Cloud Thickness (lower right)
GOES-R IFR Probabilities (upper left), GOES-East Visible imagery (upper right), Brightness temperature difference between 10.7 and 3.9 micrometers (lower left) at 1032 UTC, GOES-R Cloud Thickness (lower right) not shown because of twilight conditions

The 1032 UTC imagery (above) shows the very small scale of this fog feature that is in central Massachusetts.