Feeds:
Posts
Comments

Archive for the ‘Uncategorized’ Category

MODIS Air Mass RGB Imagery with limb correction applied to the water vapor and ozone channels.  1859 UTC, 13 May 2014

MODIS Air Mass RGB Imagery with limb correction applied to the water vapor and ozone channels. 1859 UTC, 13 May 2014

The Air Mass RGB imagery product via MODIS has often appeared to lack “green” coloring near the edge of the swath and there have been noticeable differences between the channels from Aqua and Terra used within the RGB.  Forecasters from the Great Falls, MT and Albuquerque, NM WFOs applying this experimental data noted these issues.  The above image is a limb and bias corrected version of the Air Mass RGB.  The water vapor and ozone channels tend to “cool” near the swath edge as they pass through more atmosphere and the differences in satellite instrument quality result in physical characteristics between the images having different coloring.  SPoRT has worked to develop a non-linear function to correct much of the limb cooling as well as a bias correction, both through comparison of the MODIS instruments to the EUMETSAT SEVIRI instrument.  Annotations to the image attempt to classify the various features indicated by the resulting composite color during a MODIS pass from 1859 UTC on 13 May 2014 when a cold air mass was moving into the upper Midwest.  Simple interpretation guides can be found via SPoRT’s Training page or EUMETSAT. For comparison, additional plots of GOES Water Vapor,  and NAM 500mb Temperature, Humidity, and Height 0-hour analysis and 6-hour forecasts are provided below for reference. There is also a single image of the Hybrid GEO/LEO Water Vapor / Air Mass RGB product that loops GOES Water Vapor imagery and inserts the MODIS Air Mass RGB swath as it is available because the RGB is largely made up of water vapor channels.  Both the Hybrid and single-swath MODIS files are available in netCDF format for use in AWIPS I or II as well as KML format.

This new limb/bias corrected Air Mass RGB product is credited in large part to graduate student work being done at the University of Alabama Huntsville in conjunction with NASA/SPoRT. Primary contributors are:
Nicolas Elmer (UAH graduate student)
Dr. Emily Berndt (NASA/SPoRT Post-Doctoral Scientist)
Dr. Gary Jedlovec (NASA/SPoRT PI)

Additional contributors include:
Frank LaFontaine (Raytheon, Data processing and analysis)
Kevin McGrath (Jacobs, Product code development and real-time processing)
Matthew Smith (UAH, Data processing and product code development)
Dr. Andrew Molthan (NASA/SPoRT, RGB code development and research science)

g13.2014133.1845_US_wv

GOES Water Vapor Imagery at 1845 UTC, for 13 May 2014

 

 

 

 

NAM 500mb, 0-hour forecast valid 1200 UTC, 12 May 2014 of Temperature, Humidity, and Height via

NAM 500mb, 0-hour forecast valid 1200 UTC, 13 May 2014 of Temperature, Humidity, and Height via NCAR RAL website

NAM 500mb, 0-hour forecast valid 1200 UTC, 13 May 2014 of Temperature, Humidity, and Height via NCAR RAL website

N

NAM 500mb, 6-hour forecast valid 1800 UTC, 13 May 2014 of Temperature, Humidity, and Height via NCAR RAL website

NAM 500mb, 6-hour forecast valid 1800 UTC, 13 May 2014 of Temperature, Humidity, and Height via NCAR RAL website

CAR RAL website

Example: SPoRT Hybrid GEO/LEO Water Vapor and Air Mass RGBimagery

Example: SPoRT Hybrid GEO/LEO Water Vapor and Air Mass RGBimagery

Read Full Post »

Despite the hours of darkness becoming rare over Alaska as the northern hemisphere approaches its summer solstice, the RGB Night-Time Microphysics product still has some utility in Alaska south of the Arctic Circle right around midnight. Just before midnight Alaska Time on May 6, 2014 (0743 UTC, May 7) an RGB NT Micro image derived from the SNPP VIIRS instrument depicted a deck of moderately low marine stratus clouds over the northeastern Bering Sea, as outlined in the black box in Figure 1.

Image

Figure 1: RGB NT Micro product derived from VIIRS data, 1143pm Alaska Daylight Time May 6, 2014. Area of interest noted in the black box.

A closer view of this area is shown in Figure 2, along with the ceiling and visibility data from surface observing sites. In this scenario, ceilings, rather than visibility, are the problematic weather element, with the exception of Nome where the imagery shows a localized area of higher conditions. It can be challenging to discern ceilings and visibilities from satellite imagery, and in this respect the RGB NT Micro product has an advantage over conventional satellite imagery. Per the Quick Guide available at http://weather.msfc.nasa.gov/sport/training/rgb_ntmicro/RGB%20Night-time%20Microphysics%20Reference%20Guide%20AK%20by%20SPoRT.pdf and as demonstrated in the Alaskan training module http://weather.msfc.nasa.gov/sport/training/aviationForecasting_Alaska/launcher.html a tan to light green appearance indicates low clouds, but not necessarily fog, in colder climate regions such as Alaska. Surface observations on Saint Lawrence Island and in the Yukon Delta area indicate MVFR ceilings of between one and three thousand feet, but no reduction to visibility due to fog.

Image

Figure2 : the same RGB NT Micro product as in Figure 1, zoomed into the northeastern Bering Sea. Ceiling and visibility data from surface observation sites are also shown in green.

Read Full Post »

On April 28, 2014 a tornado outbreak occurred across the Southeast, with the states of Mississippi, Alabama and Tennessee being hit the hardest. Several EF-3 and EF-4 tornadoes occurred across these states. As skies cleared over the area from west to eas, the International Space Station (ISS) orbit tracked across the city of Louisville, MS on Sunday May 4. This town was hard hit by an EF-4 tornado.  True color imagery from the ISS SERVIR Environmental Research and Visualization System (ISERV) captured the track of this tornado in its entirely.

 

The above ISERV image shows the tornado track on the southeastern portion of the town with the EF Scale Damage Indicators overlaid on top.

The above ISERV image shows the tornado track on the southeastern portion of the town with the EF Scale Damage Indicators overlaid on top.

 

SPoRT uploaded all the ISERV images to an online viewer where one can explore for more detail. That viewer can be found here.

For a comparison between pre-storm Landsat imagery and post-storm ISERV imagery, see the SERVIR team link and story here.

 

Read Full Post »

Here at the Huntsville National Weather Service Forecast Office (NWSFO), we’ve been using various SPoRT LIS parameters, namely absolute and relative soil moisture and skin and soil temperatures, for situational awareness purposes for some time now.  Specifically, I’ve been using soil moisture data to help with drought monitoring application and subsequent feedback to the U.S. Drought Monitor for a few years.  But, we’ll leave that particular use for another post.  Today, I wanted to post about the relevance of these data on the far other end of the hydrologic spectrum, that is, flooding.  Let’s take the most recent example here in Alabama and adjacent areas of Mississippi and Florida.  I’m going to start with a graphic of precipitation over the region during the 24 hour period ending at 12UTC 7 April 2014.

Stage-IV Precipitation (NWS RFCs) for the 24 Hour period ending 12 UTC 7 April 2014

Stage-IV Precipitation (NWS RFCs) for the 24 Hour period ending 12 UTC 7 April 2014

 

As shown in the graphic above, some locations on the south side of the Birmingham metro received around five to six inches of rain in about a 12 hour period.  This lead to significant flash flooding across parts of the metro where some water rescues were even necessary.  Rainfall amounts here in northern Alabama ranged from around one inch in the far northwestern part of the state to about 3.5-4 inches in parts of north central and northeastern Alabama.

Now, let’s investigate the antecedent soil moisture conditions before heavy rainfall moved across the region.  Let’s begin by taking a look at the shallow layer (0-10 cm) relative soil moisture produced by the SPoRT LIS.

SPoRT LIS 0-10 cm Relative Soil Moisture (%) 12 UTC 6 April 2014

SPoRT LIS 0-10 cm Relative Soil Moisture (%) 12 UTC 6 April 2014

Relative soil moisture values generally south of Birmingham, and in particular in the “black belt” region of Alabama, a region characterized by rich, black topsoil, but underlying less permeable chalks and clay soils, were around 75% to 90%.  So, even before widespread heavy rains of three or more inches moved across the region, soil moisture values were relatively high.  Note here in northern Alabama however, that values were lower, at around 50%.  The combination of lower antecedent soil moisture conditions and similar to lower rainfall led to less incidence of flash flooding in northern Alabama.  In fact, only two flash flood warnings were issued for the event and were for the same location: encompassing parts of DeKalb, Jackson and Marshall Counties in northeastern Alabama.  Incidentally, soil moisture values were higher in the Big Wills Valley in DeKalb County where most of the flash flooding occurred.  Although, other factors should be considered there such as the narrow and steep-walled drainage basin characteristics.

Now, let’s take a look at the deep layer (0-200 cm) antecedent relative soil moisture conditions across the region (below).

SPoRT LIS 0-200 cm Relative Soil Moisture (%) 12 UTC 6 April 2014

SPoRT LIS 0-200 cm Relative Soil Moisture (%) 12 UTC 6 April 2014

 

In this image, the solid darker green indicates relative soil moisture values around 60% or higher, while blue shades indicate values in the 70s-90s%.  Notice that values were this high or higher across much of Mississippi and central and southern portions of Mississippi.  Next is a map of flooding locations as of this morning (April 11th…unfortunately, I don’t have one from earlier).

River Flood Summary courtesy of the SERFC valid as of ~14 UTC 11 April 2014

River Flood Summary courtesy of the SERFC valid as of ~14 UTC 11 April 2014

 

Streams and rivers with orange and red boxes indicate locations of minor and moderate flooding, respectively.  Notice that many of these are located where rains were heaviest, but also where antecedent soil moisture values were highest.  The main take away item here is that while rainfall values of this magnitude can occur in this part of the country, importantly, soil moisture values must be included to make a more complete assessment of the threat for flooding.  We certainly have had higher rainfall amounts in the region with much less flooding.  Such was the case with the passage of Tropical Storm Lee in September 2011, especially here in northern Alabama.

Here at the Huntsville NWSFO it has taken some time, but we have noticed that when deep layer (0-200 cm) relative soil moisture values exceed about 60%, we are at a greater risk for longer term flooding on local stream and river basins when we receive a “typical” synoptic rainfall event totaling around 2-3 or more inches.  Values in northern Alabama before this event were generally under 60%, with the exception of the Big Wills Valley in DeKalb County and in some small portions of the Paint Rock Valley in Jackson County.  Incidentally, in addition to the flooding along the Big Wills Creek in Fort Payne, minor flooding occurred along the Paint Rock River in Jackson County here in northern Alabama.  This was one of the locations with some indication of relatively wet antecedent soils (relative soil moisture values ~60%).  Thus, once again, we received further confirmation in these rough thresholds.  These two locations contain streams that are particularly prone to flooding, but it has been difficult to gauge what rainfall amounts are necessary.  This is because an important component of that assessment was lacking until the advent of LIS soil moisture data into our operations beginning over a year ago.

Thanks to the SPoRT team, we now have these data back in AWIPS (II) and can overlay other important data, such as QPF from the Weather Prediction Center.  Although these flooding threat analyses are mostly qualitative and subjective at this point, the plan is to undertake a more objective study of soil moisture and rainfall thresholds in our more problematic drainage basins to have a better understanding of threats for flooding in the future.

 

 

Read Full Post »

wfohunkris:

A fantastic post by Satellite Liaison Michael Folmer about the use of SPoRT and CIRA RGB products for examining the very rapid development of Tropical Cyclone Hellen in the Indian Ocean

Originally posted on Satellite Liaison Blog:

First of all, welcome to the brand new “Satellite Liaison Blog”!  After some consideration, the other satellite liaisons and I decided to revamp this blog to fit the Proving Ground activities at all of the National Weather Service National Centers and Field Offices.  We hope to keep you informed and spur excitement about new satellite products as we move closer to the launch of GOES-R and JPSS-1.  Please bookmark the site (satelliteliaisonblog.wordpress.com) for future reference, although you will be forwarded here from the old site name for a year.

Now on to this week’s storm of interest. . .

Tropical Cyclone Hellen formed in the northwest Indian Ocean between Mozambique and Madagascar on 03/28/14.  Hellen rapidly intensified from a 60 kt (70 mph) tropical storm to a 130 kt (150 mph) Category-4 tropical cyclone in 18 hours (satellite-based)!  Since the Atlantic hasn’t shown us a strong hurricane (Category 3+, but…

View original 391 more words

Read Full Post »

Originally posted on TFX-shoptalk:

A fast moving shortwave trough is forecast to bring snow through central Montana this morning. The radar has been active (Fig. 1), but snow is not yet being reported at area stations except at Havre, where its only just began. It’s night, and only a few web cams are illuminated. The nightime microphysics RGB (Nt-micro) adds some value to the situation (Fig. 2). At 0448 UTC the Nt-micro image shows widespread red-orange-purple colors across the state. These are best interpreted as optically thick mid to high level clouds. Ceilings above 6000 feet AGL are consistent with this interpretation. The radar has been active in these areas for much of the night in clear-air mode, so likely picking up on ice crystals within the cloud.  Four hours later, the scene has changed (Fig. 3).  Notice the introduction of bright yellow clouds in southern Alberta, just sneaking into northern Glacier County. These…

View original 178 more words

Read Full Post »

The severe to extreme drought conditions in the Colorado, Kansas, Oklahoma, Texas, and New Mexico regions surrounding the panhandle of Oklahoma (see U.S. Drought Monitor: http://droughtmonitor.unl.edu/) served as a source region for a large-scale dust event impacting these five states.  The VIIRS and MODIS RGB Dust imagery shows this dust event well with MODIS providing an early indication of the event prior to obvious dust signatures in current GOES imagery. See the “Real-Time Data” tab on the NASA/SPoRT webpage: http://weather.msfc.nasa.gov/sport/.

At 1807 UTC MODIS imagery (see below) indicated two areas of blowing dust in the afore mentioned region.  At this time the GOES visible imagery was hinting at the dust region in southeast Colorado but the GOES imagery was not clearly showing the dust in Texas.  By the 1947 UTC VIIRS Dust RGB imagery (second image below), both the GOES visible and IR imagery showed the two large dust regions, but VIIRS Dust RGB imagery provided a more clear definition of the dust plume impacting north TX, originating in eastern NM.  By 2202 UTC the dust plume from Colorado had fanned out and could be well seen in the GOES visible imagery (see below) impacting NM, OK, and TX while the dust from the eastern NM source region had spread across north TX and into OK.  I’m sure we’ll see many photos of this extreme event over a wide area; feel free to post them to this blog.

RGB Dust Imagery from MODIS at 1807 UTC 18 March 2014 (taken from NASA/SPoRT webpage)

RGB Dust Imagery from MODIS at 1807 UTC 18 March 2014 (taken from NASA/SPoRT webpage)

Dust RGB Imagery from VIIRS at 1947 UTC 18 March 2014 (taken from NASA/SPoRT webpage)

Dust RGB Imagery from VIIRS at 1947 UTC 18 March 2014 (taken from NASA/SPoRT webpage)

GOES Visible Imagery from 2202 UTC 18 March 2014 (taken via UCAR/RAL webpage)

GOES Visible Imagery from 2202 UTC 18 March 2014 (taken via UCAR/RAL webpage)

 

 

 

 

Read Full Post »

Jonathan Case of NASA/SPoRT and Dr. Ashutosh Limaye of NASA/SERVIR traveled to Nairobi, Kenya this week to collaborate with the Kenya Meteorological Service (KMS; see figure below).  The goals of this visit were to learn more about KMS operations, transition the SPoRT-Model Evaluation Tools (MET) verification scripting package for operational use at KMS, and to demonstrate initializing the Weather Research and Forecasting (WRF) model with high-resolution Land Information System (LIS) land surface fields.  Mr. Case and Dr. Limaye worked closely with Mr. John Mungai and Mr. Vincent Sakwa of KMS to install and test the SPoRT-MET scripting package during the course of the visit.  Installation and testing of the SPoRT-MET scripts were successful for generating surface point and gridded precipitation verification statistics, including the generation of graphical “quick plots” using embedded Perl utilities.  Mr. Mungai and Mr. Sakwa indicated the high utility that this verification system will provide KMS to improve operational efficiency through routine, automated verification of WRF model forecast accuracy, as well as other models the KMS uses in-house for which they have access to GRIB files.  SPoRT and SERVIR will continue collaborating with KMS representatives to ensure automation of SPoRT-MET capabilities, and to offer additional enhancements to the KMS-WRF, such as future real-time vegetation data to improve LIS and WRF simulations.

Jonathan Case of SPoRT and Dr. Ashutosh Limaye of SERVIR work with John Mungai and Vincent Sakwa of the Kenya Meteorological Service to install and test the SPoRT-MET scripts for conducting automated model verification.

Jonathan Case of SPoRT and Dr. Ashutosh Limaye of SERVIR work with John Mungai and Vincent Sakwa of the Kenya Meteorological Service to install and test the SPoRT-MET scripts for conducting automated model verification.

Read Full Post »

radar_Feb262014_1024UTCsfr_Feb262014_1022UTC

A weather system moved across the area during the overnight hours into the morning hours on Feb 26 2013. As a result…a band of snow developed and moved southeast. As of 8 AM…most locations were reporting 1.5 to 2.5 inches…with some higher anmounts.

I have attached two images from approximately the same time. The first image shows a radar mosaic from 1024 UTC for West Virginia and surrounding areas while the second image is an SFR image from 1022 UTC. The radar mosaic shows the heaviest snow across eastern Kentucky into the southern counties of West Virginia. The SFR image shows the best signal for snow across northern portions of West Virginia. Surface observations generally showed the heavier snowfall…based on visibility…was falling from eastern Kentucky across southern West Virginia into the northeastern West Virginia counties.

Each dataset had it’s strengths and weaknesses. However, using a combination of all three datasets was the best method to show the overall coverage of the heavier snow.

Read Full Post »

GOES IR (left), NESDIS Snowfall Rate Product (center) and Composite NEXRAD Base Reflectivity (right)           for around 1045 UTC 28 January 2014.

GOES IR (left), NESDIS Snowfall Rate Product (center) and Composite NEXRAD Base Reflectivity (right) for                 around 1045 UTC 28 January 2014.

In the early morning hours of 28 January (approximately 1047 UTC or 4:47 AM CST), the Operational NESDIS Snowfall Rate Product (SFR; above center), that uses the Advanced Microwave Sounder Unit (AMSU-A) and Microwave Humidity Sounder (MHS) on-board four Low Earth Orbiting (LEO) NOAA POES and EUMETSAT Metop satellites to produce a liquid equivalent snowfall rate over land, was detecting snow in the clouds over Central Alabama (red circle) and an area from northern Mississippi into Central Tennessee (yellow oval).   In both areas there was some enhanced cooler cloud tops on the geostationary infrared imagery (above, left) and light precipitation  detected by the composited NEXRAD base reflectivity (above, right) in approximately the same area that the SFR was detecting snow in the clouds.  However, the only snow being observed (light snow at that) at the surface at the time was in Northwest Mississippi at Tunica Municipal Airport.  The SFR has a known high bias for sensing low liquid equivalent rates/light snow and so forecasters need to understand that just because the product is sensing snow in the clouds, the relatively low moisture that may exist close to the surface at the start of an event can initially prevent the precipitation from reaching the ground.  Now let’s focus on the Central Alabama snow being detected by the SFR at 1047 UTC in the enhanced satellite geostationary IR cloud tops and aloft by the radar.  At this time no snow was being observed around the Birmingham, Alabama area.  Still the SFR would have provided valuable information to forecasters that the snow detected in the clouds could be seeding existing clouds (a precursor) and with increased moisture more easily support snow eventually reaching the ground later in the morning.   This is what actually occurred as Birmingham-Shuttlesworth International Airport started to observe snow reaching the ground shortly before 16 UTC (10 AM CST) and was confirmed by the approximate 1615 UTC SFR (below, center), GOES IR (below, left) and composite NEXRAD (below, right) imagery/data; about 5 hours after the 1047 UTC SFR was detecting snow in the clouds.   NESDIS researchers and analysts are carefully looking at the SFR product’s sensing of the snow in the cooler enhanced IR cloud tops as a precursor to snow eventually reaching the ground; thus helping alert forecasters to the impending snow reaching the ground.  And in this case once the snow started reaching the ground in the Birmingham area just before 16 UTC, its intensity increased through the rest of the morning and into the afternoon on the 28th (snow-water equivalent reached 0.03″ through 18 UTC and an additional 0.05″ fell between 18 and 21 UTC .

Approximate 1615 UTC GOES IR (left), Snowfall Rate Product (center), NEXRAD Composite Base Reflectivity (right)

GOES IR (left), NESDIS Snowfall Rate Product (center), Composite NEXRAD Base Reflectivity (right) for                        around 1615 UTC 28 January 2014. Black Circle depicting snow in and around Birmingham, Alabama.

 

Read Full Post »

Older Posts »

Follow

Get every new post delivered to your Inbox.

Join 93 other followers