GOES-14 SRSOR: Fires in northern California

August 21st, 2012
GOES-14 3.9 µm shortwave IR (left) and 0.63 µm visible (right) images (click image to play animation)

GOES-14 3.9 µm shortwave IR (left) and 0.63 µm visible (right) images (click image to play animation)

McIDAS images of GOES-14 1-minute interval Super Rapid Scan Operations for GOES-R (SRSOR) 4-km resolution 3.9 µm shortwave IR channel data and 1-km resolution 0.63 µm visible channel data (above; click image to play animation) showed a number of significant wildfires burning across parts of northern California on 21 August 2012. The largest and most intense fires exhibited pronounced “hot spots” (black to yellow to red color enhancement) on the shortwave IR imagery, with optically-thick smoke plumes on the corresponding visible imagery.

The GOES-14 satellite has been brought out of on-orbit storage to be tested in SRSOR mode through the end of October 2012, allowing it to provide images at 1-minute intervals for an extended period of time over special regions of interest (similar to the future GOES-R satellite, which will be capable of producing imagery at 30-second intervals over special sectors of interest).

During the previous night-time hours, a comparison of AWIPS images of 1-km resolution MODIS 3.7 µm data with the corresponding 4-km resolution GOES-15 3.9 µm shortwave IR data (below) demonstrated the value of improved spatial resolution for identifying the location of smaller fires, as well as more accurately assessing the location and shape of the more intense portions of larger actively burning fires.

MODIS 3.7 µm vs GOES-15 3.9 µm shortwave IR images

MODIS 3.7 µm vs GOES-15 3.9 µm shortwave IR images

Fog and Low Stratus Detection

August 21st, 2012
Toggle between GOES-East Brightness Temperature difference (10.7µm and 3.9 µm) and Aqua MODIS Brightness Temperature difference (11 µm and 3.7 µm)

Toggle between GOES-East Brightness Temperature difference (10.7µm and 3.9 µm) and Aqua MODIS Brightness Temperature difference (11 µm and 3.7 µm) (click image to play animation)

Toggle between GOES-East Brightness Temperature difference (10.7µm and 3.9 µm) and Suomi/NPP VIIRS Brightness Temperature difference (10.8 µm and 3.74 µm)

Toggle between GOES-East Brightness Temperature difference (10.7 µm and 3.9 µm) and Suomi/NPP VIIRS Brightness Temperature difference (10.8 µm and 3.74 µm) (click image to play animation)

Water clouds, such as those that develop when fog and low stratus form, have different emissivity properties at short IR wavelengths (around 3.9 µm) compared to longer IR wavelengths (around 11 µm). At shorter IR wavelengths, water cloud do not emit as blackbodies, meaning less radiation is emitted than is theoretically possible. Emissions at longer IR wavelengths are far closer to the theoretical maximum. The perceived temperature, then (computed based on the assumption that emissions are as blackbodies), based on the detected radiation at shorter IR wavelengths are cooler than the perceived temperature based on radiation detected at longer IR wavelenghts. This brightness temperature difference (BTD) field can be used to highlight regions of low clouds, as shown in the imagery above. Calm conditions over the central Appalachians have allowed fog formation especially in river valleys. The 1-km resolution of MODIS imagery (from Aqua) and VIIRS imagery (from Suomi/NPP) shows far higher detail than is available from GOES imagery.

A fog and low stratus detection scheme based solely on BTD fields, however, has shortcomings. For example, in the imagery above, no signal is available underneath the cirrus canopy that stretches from Georgia and South Carolina northeastward into southern New England. In addition, VIIRS, MODIS and GOES show fog/low stratus over western Ohio, but this plot of visibilities shows no serious obstructions to visibility in western Ohio where the BTD signal is strong (and significant visibility obstructions underneath the cirrus canopy in eastern Pennsylvania and Maryland where the BTD is weak). How can this satellite-based signal be improved?

MODIS-based IFR Probabilities, 0651 UTC on 21 August 2012

MODIS 3.7 µm vs GOES-15 3.9 µm shortwave IR images

Model data — in this case, from the Rapid Refresh — can be used in a fused product to add information in regions where satellite data cannot be used, and to refine the satellite data elsewhere. IFR Probabilities — an algorithm developed for use with GOES-R ABI data, but applicable to both MODIS and GOES-East data — using MODIS data, above, show very high probabilities over the river valleys of Appalachia where both satellite and model predictors agree that fog or low stratus is likely. Lower, but still significant, probabilities that are based solely on model data, are present underneath the cirrus canopy in southeastern Pennsylvania southwestward to Georgia (because the satellite data cannot be used here, a somewhat lower probability occurs). And, significantly, the stratus deck over western Ohio, which is not associated with IFR conditions, is de-emphasized in the IFR probability field, because the model data shows IFR conditions are unlikely even though the satellite signal — caused by elevated stratus — is strong. Model data adds to and refines the satellite-only BTD field. A plot similar to the MODIS IFR probability, but using GOES Imager data, is here.

Finally, note in the toggle between the VIIRS BTD product and the GOES BTD product, above, that a signal in the GOES Imagery along the eastern shore of Lake Huron is not replicated in the data from Suomi/NPP. The one-pixel co-registration offset between the 3.9 and 10.7 channels, first noted here, persists.

(This blog post is an expanded version of one first posted here.)

GOES-14 SRSOR: Severe thunderstorms in the Texas panhandle region

August 20th, 2012
GOES-14 0.63 µm visible channel images (click image to play animation)

GOES-14 0.63 µm visible channel images (click image to play animation)

McIDAS images of 1-km resolution GOES-14 1-minute interval Super Rapid Scan Operations for GOES-R (SRSOR) 0.63 µm visible channel data (above; click image to play animation) showed the development of severe thunderstorms in the Texas panhandle region during the afternoon and early evening hours on 20 August 2012. These storms produced numerous reports of damaging winds (as high as 76 knots or 88 mph) and hail up to 1.75 inch in diameter (SPC storm reports).

The GOES-14 satellite has been brought out of on-orbit storage to be tested in SRSOR mode through the end of October 2012, allowing it to provide images at 1-minute intervals for an extended period of time over special regions of interest (similar to the future GOES-R satellite, which will be capable of producing imagery at 30-second intervals over special sectors of interest).

GOES-14 SRSOR 1-minute interval images of Tropical Depression Helene

August 18th, 2012
GOES-14 0.63 µm visible channel images (click image to play animation)

GOES-14 0.63 µm visible channel images (click image to play animation)

On 18 August 2012, GOES-14 provided Super Rapid Scan Operations for GOES-R (SRSOR) imagery of Tropical Depression Helene making landfall along the eastern coast of Mexico — McIDAS images of 1-km resolution 0.63 µm visible channel data (above; click image to play large 128 MegaByte animation) showed convective bursts near the center of Helene as it moved inland, along with a larger convective band that persisted just offshore over the far western Gulf of Mexico (which eventually produced well-defined low-level convective outflow boundaries along the northern end).

The GOES-14 satellite has been brought out of on-orbit storage to be tested in SRSOR mode through the end of October 2012, allowing it to provide images at 1-minute intervals for an extended period of time over special regions of interest (similar to the future GOES-R satellite, which will be capable of producing imagery at 30-second intervals over special sectors of interest).

A comparison of AWIPS images of 1-km resolution POES AVHRR 0.63 µm visible channel and 10.8 µm IR channel data at 14:18 UTC (below) showed some of the earlier low-level convective outflow boundaries approaching far southern Texas, as well as embedded areas of convection exhibiting IR brightness temperature as cold as -77º C both near the inland center of Helene and also within the offshore convective band.

POES AVHRR 0.63 µm visible channel and 10.8 µm IR channel images

POES AVHRR 0.63 µm visible channel and 10.8 µm IR channel images