Plumes of blowing dust (and smoke) in Washington State

July 12th, 2010 |
MODIS 0.65 µm visible image + MODIS Normalized Difference Vegetation Index product

MODIS 0.65 µm visible image + MODIS Normalized Difference Vegetation Index product

Strong northwesterly winds behind a cold front gusted as high as 55 mph at Magee Peak in Washington State on 12 July 2010, causing some traffic accidents and road closures due to low visibility from blowing dust — and the blowing dust ended up restricting surface visibility to 2 miles as far to the east as Spokane (station identifier KGEG). A write-up of the event by the NWS forecast office at Spokane showed some web camera views of the dust. An AWIPS image of the 1-km resolution 0.65 µm visible channel data (above) did reveal two distinct aerosol plumes: one originating to the northwest of Wenatchee (station identifier KEAT), and another originating to the north and northeast of Moses Lake (station identifier KMWH). These aerosol plumes appeared to be originating from areas with a low Normalized Difference Vegetation Index (NDVI) value, suggesting dry land void of crops, trees, or other vegetation.

A comparison of MODIS 0.65 µm visible and MODIS 6.7 µm water vapor channel images (below) indicated that there was a mountain wave (to the lee of the Cascade Range) present over the region with the plumes, which may have acted as a mechanism to help transfer some of the strong momentum aloft downward toward the surface.

MODIS 0.65 µm visible image + MODIS 6.7 µm water vapor channel image

MODIS 0.65 µm visible image + MODIS 6.7 µm water vapor channel image

Note how this mountain wave signature as seen with MODIS was not evident at all on the corresponding 8-km resolution GOES-11 6.7 µm water vapor channel image (below).

1-km resolution MODIS + 8-km resolution GOES-11 6.7 µm water vapor images

1-km resolution MODIS + 8-km resolution GOES-11 6.7 µm water vapor images

A closer look using 250-meter resolution MODIS true color and false color Red/Green/Blue (RGB) images from the SSEC MODIS Today site (below) was helpful in determining that the westernmost plume was actually smoke from a wildfire (the fire hot spots and burn scar showed up as red to pink on the false color image, and the smoke plume itself was brighter on the true color image). In contrast, the blowing dust plume farther to the east had more of a light brown to tan appearance on the true color image.

250-meter resolution MODIS true color and false color RGB images

250-meter resolution MODIS true color and false color RGB images

===================================

MODIS 3.7 µm shortwave IR image

MODIS 3.7 µm shortwave IR image

The presence of an actively-burning fire was confirmed by a large cluster of hot pixels (dark black color enhancement) to the northwest of Wenatchee (station identifier KEAT) on MODIS 3.7 µm shortwave IR imagery (above). Farther to the south and southeast, the larger dark black area seen on the shortwave IR image corresponded to the sparsely-vegetated region around Hanford (station identifier KHMS), which was exhibiting MODIS Land Surface Temperature values in the 120-130º F range (below).

MODIS Land Surface Temperature product

MODIS Land Surface Temperature product

Water Vapor channel “weighting functions”: helping to explain the complexity of water vapor image interpretation

July 12th, 2010 |
GOES-13 Sounder and Imager water vapor channels (00 UTC, 12 July)

GOES-13 Sounder and Imager water vapor channels (00 UTC, 12 July)

An AWIPS 4-panel display of the three GOES-13 Sounder water vapor channel images along with the single GOES-13 Imager water vapor channel image (above) revealed that a pocket of warm brightness temperatures (denoting dryer air aloft) was in place over much of eastern Virginia and North Carolina around 00 UTC on 12 July 2010.

The corresponding plots of the GOES-13 Sounder (red, green, and blue plots) and GOES-13 Imager (black plot) water vapor channel weighting functions (below, calculated using 00 UTC rawinsonde data from Greensboro, North Carolina) showed a relatively uncommon “bi-modal” structure — this indicated that there were significant contributions from two distinct layers, such that the actual altitude of the features being displayed on the water vapor imagery over that region at that time would be difficult to ascertain.

Greensboro NC GOES-13 Sounder and Imager water vapor weighting function plots

Greensboro NC GOES-13 Sounder and Imager water vapor weighting function plots

However, 12 hours later, the same AWIPS 4-panel display of water vapor imagery around 12 UTC (below) showed that significant amounts of moisture and cloudiness had moved into the region that was relatively dry at 00 UTC.

GOES-13 Sounder and Imager water vapor channels (12 UTC, 12 July)

GOES-13 Sounder and Imager water vapor channels (12 UTC, 12 July)

With the increased middle and upper tropospheric moisture, note that the 12 UTC  water vapor weighting function plots (below, calculated using 12 UTC rawinsonde data from Greensboro NC) all peaked at significantly higher altitudes — and by the shape of the weighting function plots, it was easier to tell the layer of the troposphere that was being sampled by each of the individual water vapor channels.

Greensboro NC GOES-13 Sounder and Imager water vapor weighting function plots

Greensboro NC GOES-13 Sounder and Imager water vapor weighting function plots

AWIPS images of the Blended Total Precipitable Water product (below) confirmed that there was a significant increase in total column moisture content over the Greensboro NC region during that 12-hour period. TPW values rose from about 21 mm to around 40 mm over Greensboro during that time.

Blended Total Precipitable Water product

Blended Total Precipitable Water product

Fog and stratus along the entire West Coast of the US

July 9th, 2010 |
GOES-11 0.65 µm visible channel images

GOES-11 0.65 µm visible channel images

GOES-11 0.65 µm visible channel imagery (above) indicated that you would have been hard pressed to find a location along the entire West Coast of the US (from California to Washington State) that did not have fog or stratus clouds overhead during the morning hours on 09 July 2010!

During the preceding night-time hours (before visible imagery was available), a comparison of AWIPS images of the 1-km resolution MODIS fog/stratus product and the 4-km resolution GOES-11 fog/stratus product centered over central California (below) demonstrated the value of higher spatial resolution in determining just how far inland some of the fog and stratus features were located at around 06:30 UTC (11:30 pm local time).

MODIS and GOES-11 fog/stratus product images

MODIS and GOES-11 fog/stratus product images

“Seeing the surface” on water vapor imagery

July 8th, 2010 |
MODIS 6.7 µm water vapor image (with and without map overlay)

MODIS 6.7 µm water vapor image (with and without map overlay)

Under normal atmospheric conditions, the weighting function of most water vapor channels tends to peak at altitudes within the 500-300 hPa pressure range, allowing features within the middle to upper troposphere to be viewed on the water vapor imagery. However, under special conditions — for example, either a very dry or a very cold air mass — the altitude of the water vapor weighting function is shifted downward such that we are able to “see the surface” on water vapor imagery. Such was the case with the MODIS 6.7 µm water vapor image over the Baja California region on 08 July 2010 (above), where the outline of the coast was very obvious on the image.

Even though the water vapor channel was not “seeing the surface” per se, a signal of the strong surface thermal contrast (between the very warm land and the much cooler water) was able to override the weak signal from what little middle-tropospheric water vapor was present. Other cases of strong land/water temperature contrasts have been seen on water vapor imagery, such as with very cold and very dry arctic air masses back in February 2007, December 2006, and January 2004.

However, in this case, the signal of the land/water thermal contrast was not evident on the corresponding GOES-11 6.7 µm / GOES-13 6.5 µm water vapor composite image. Because of the large viewing angle of the geostationary satellites (around 40 degrees for GOES-11 and around 55 degrees for GOES-13 for the Baja California region), the water vapor weighting function was apparently shifted upward to a high enough altitude to preclude detection of the surface land/water thermal signal.

GOES-11 6.7 µm + GOES-13 6.5 µm water vapor composite (with and without map overlay)

GOES-11 6.7 µm + GOES-13 6.5 µm water vapor composite (with and without map overlay)

Surprisingly, not even the GOES-11 sounder 7.4 µm water vapor channel image (below) was able to detect the strong surface thermal signal — the weighting function of this channel often peaks much lower in the troposphere (usually around 850-700 hPa). Again, perhaps the large geostationary satellite viewing angle was a factor. With the MODIS instrument flying directly overhead, there was no corresponding upward shift in the water vapor channel weighting function.

GOES-11 sounder 7.4 µm water vapor image (with and without map overlay)

GOES-11 sounder 7.4 µm water vapor image (with and without map overlay)