A potent winter storm system impacted portions of New Mexico on March 26, 2016, ending an extended stretch of very dry weather. Snowfall amounts of 3 to 9 inches were reported from the Sangre de Cristo Mountains eastward across the northeast plains. The MODIS and VIIRS satellite products proved useful for illustrating the extent of snow cover in both the daytime and nighttime scenes. The images below are graphical briefings posted to the NWS Albuquerque web page and shared via Twitter after this much needed snowfall event.
Archive for the ‘RGB’ Category
A powerful jet stream approached NM from the Pacific Northwest on February 23, 2016 and carved out a large scale upper level trough over the southern Rockies. Meanwhile, a strong area of surface high pressure raced southward down the front range of the Rockies and provided an influx of moist, cold air into eastern NM. Winter storm warnings were issued for several counties in northeastern NM with winter weather advisories in a few surrounding zones. Forecasters were eagerly anticipating how the event would unfold on the merged SFR product given the recent stretch of very dry and exceptionally warm weather. Figure 1 below is a Graphical Briefing issued prior to the event with expected snowfall totals.
Rain and high terrain snow developed from north to south late on the 22nd before transitioning to all snow early on the 23rd. Forecast models from the 21st indicated that an area of heavy snow would impact northeastern NM on the 23rd. Figure 2 shows a well-defined axis of higher snowfall rates stretching from Trinidad, CO to Capulin, NM and Tucumcari, NM in the stand-alone SFR product from 0912 UTC 23 February 2016. Sampling this area showed peak liquid equivalent values near 0.07″/hr. The 0900 UTC observation at Trinidad showed the visibility had fallen to 1/2 mile within the snow band while the observation at Raton, NM showed no snowfall where the SFR product had near zero values. Farther south near Tucumcari the observation showed unknown precipitation falling at a temperature of 36° south of the main snow band.
The following merged SFR image at 0940 UTC showed the area of higher snowfall rates persisting (Figure 3 (left)). A merged SFR product from 0950 UTC is shown to note the extent of the radar void area (Figure 3 (right)). The following stand alone SFR product at 1245 UTC indicated the higher rates had shifted farther south but were still impacting at least portions of this same area (Figure 4). In this example the observation at Las Vegas, NM was indicating snowfall with visibilities down to 1 1/4 miles while Tucumcari showed very light snow with no values on the SFR product.
Based on the peak values depicted in the SFR product and the persistence of the snow band in the area forecasters were anticipating snowfall reports in the 2 to 6 inch range. The Snow-Cloud RGB product later in the day in Figure 5 verified this area of snowfall very well (red shades). Spotters reports are overlaid on the RGB imagery. Feedback from forecasters during this event supported accurate observations of the SFR product during the transition from rain to snow as well.
Recently, I had the opportunity to travel to the Tucson NWS office and work with forecasters there concerning a number of experimental data sets transitioned by the SPoRT group. Primarily, this involved the SPoRT LIS, GPM Constellation and IMERG, and NESDIS QPE data sets. However, I also had the opportunity to see how other products were being utilized by forecasters. While taking a look at the Nighttime Microphysics RGB image, I was initially perplexed by the apparent presence of fog and low clouds in parts of the desert southwest. The first image below is a 4-panel image from AWIPS, showing the Longwave (LW) and Shortwave (SW) IR, the LW-SW IR channel difference, and the Nighttime Microphysics RGB from the VIIRS instrument on the morning of Sept 23rd.
The difference in brightness temperatures between the LW and SW IR channels in parts of SW Arizona, SE California and areas of NW Mexico around the Gulf of California, results in relatively large positive values. Notice the yellow colors that appear in these areas in the channel difference imagery (image 1, lower right), and the corresponding appearance of white-aqua colors in the Nighttime Microphysics RGB (the 10.8-3.9 channel difference represents the green color component of the RGB recipe). For a forecaster accustomed to looking at these imagery in other parts of the country (and those will less sandy surfaces), these channel difference values and colors in the RGB would suggest the presence of low stratus and/or fog. However, no clouds or fog were present in those locations during the morning. You can, however, see some low clouds in portions of central and eastern New Mexico, as indicated by the brighter white-aqua colors.
So, what is going on here? Well, as eluded to above, it’s the presence of dry sand. The image below (courtesy of COMET) shows the IR emissivity over several different surface features: tree leaves, red clay, dry sand, and water.
Notice that the emissivity over dry sand changes fairly substantially through portions of the SW and LW portion of the spectrum, and is lower at 3.9 µm than at 10.8 µm. The channel difference between 10.8 and 3.9 µm will result in positive values (given clear sky conditions of course) over dry sandy areas, thus mimicking the presence of low clouds and/or fog, as would be the interpretation in other areas. The next image below demonstrates the LW and SW IR brightness temperatures and differences, along with the Nighttime Microphysics RGB, as sampled over a clear, dry sandy area.
Notice the substantial resulting green color contribution in the Nighttime Microphysics RGB (lower right in above image). These colors are very similar to colors that would be indicative of fog and other low cloud features as they traditionally appear under similar temperature conditions in other areas outside of dry, sandy areas (image 4 below).
With an unseasonably strong upper low approaching New Mexico, forecasters at NWS Albuquerque anticipated high elevation snow and widespread rain with relatively high QFP values for the period of 26-27 April 2015. A winter storm watch was issued at 400 am MDT on Saturday, April 25. Snow was forecast for the highest terrain across the northern and central New Mexico, but significant snow accumulations were expected late Sunday, April 26 through early Monday, April 26. Additionally, rain amounts in excess of an inch were expected across the eastern plains.
The GFE storm total snow from mid-day Saturday, April 25 is shown below. The forecast called for the most significant snow accumulations, just over a foot across the highest peaks, to occur over the Sangre de Cristo Mountains (just to the west of Interstate 25) and the higher elevations along the Colorado border north of Raton, NM. The watch was upgraded to a warning at 4am MDT on Sunday, April 26.
Widespread precipitation was reported during the overnight hours, with 3-.6in of rain in the Albuquerque metro area. The position of the closed low early on the morning of Monday, April 27 is shown below. Snow was still being reported at Angel Fire in the Sangre de Cristo Mountains, but the big story by this point was rain across the eastern plains.
Two Snowfall Rate products were received during the overnight hours around 3am MDT (09Z). The date/time stamp was not included on the images – the first shows SFR at 0838Z and the second at 0913Z. Both include metar observations from 09Z. Angel Fire is reporting snow, though in both images the SFR ends just north of the site. Raton is reporting rain and the SFR products both show the eastern edge of snow accumulations ending just to the west of Raton. Based on very high accumulations south of Angel Fire, the SFR product may be underestimating the area of active snow.
As is often the case, radar cover across the north central mountains is limited. The 0.5 reflectivity mosaic below is from 09Z, between the two SFR products above. Angel Fire is marked by the purple circle. Radar returns over the Sangre de Cristo mountains are greater north of Angel Fire. In eastern New Mexico, Tucumcari (blue circle) is reporting rain associated with the strongest radar returns. Rain continued through the daytime hours with numerous rainfall reports of over one inch. In fact, Tucumcari Aiport reported 1.50″ of rain, the 3rd largest 1-day total in April since 1941!
Snow accumulation reports did verify our forecast of over a foot of snow for this late season event. Determining snow records is more difficult since routine snow observations are few. Highest totals were received in the Sangres, with 18 inches observed at Black Lake, just south of Angel Fire. Smaller accumulations were noted over the San Juan and Jemez Mountains, areas which did not have a Winter Storm Warning in effect. The RGS Snow-Cloud product from April 28 shows new snow cover across much of the north central high terrain. Snow over the San Juan and Jemez Mountains (west of the Sangres) likely accumulated prior to the SFR products above.
I’ve written about the operational utility of Day-Night Band (DNB) RGB imagery several times in the SPoRT blog, and here I’m going to take the chance to do that again. First, just some brief background information in case you’re not familiar with this type of imagery. The DNB RGB is composed of a long wave IR channel (~10.8 µm), which is assigned to the blue color component of the RGB recipe, while the DNB channel (0.7 µm) is assigned equally to the red/green colors of the RGB. SPoRT produces two DNB RGB products: Radiance and Reflectance. I typically prefer the Radiance RGB for operational uses since it is composed of the raw data (emitted and reflected light) from the sensor. Sure, cities are quite bright in the imagery, but the cloud features also stand out better compared to the reflectance product, where the data are normalized by the available amount of moonlight. Below are a few observations from the Suomi-NPP VIIRS instrument during this most recent full moon cycle.
First, take a look at the images below from the SE half of the CONUS on the early morning of December 7th. The top image (Image 1) is a Nighttime Microphysics RGB at approx. 0736 UTC, while Image 2 is a DNB Radiance RGB valid at the same time. While this type of imagery is far superior to legacy IR imagery (even enhanced with fanciful color curves), there are proper operational forecasting/analysis applications that one has to consider. The Nighttime Microphysics RGB is generally more useful for distinguishing different cloud types (e.g., low stratus vs fog, thin cirrus vs thick cirrus, etc). After at least a year of viewing the DNB imagery, I think perhaps the best application of these types of imagery (at least with respect to operational forecasting) lies in the ability to view low clouds through cirrus at night. No other imagery available to forecasters offers this capability currently. Take for example these first two images below and pay close attention to the cloudy regions stretching from the central Plains into the lower Mississippi Valley.
Notice that in the Nighttime Microphysics RGB the expansive deck of low stratus across much of Kansas, southwestern Missouri and Oklahoma is almost entirely obscured by the cold cirrus clouds. Of course, this is only realized upon looking at the DNB imagery. Details in the low stratus can also be observed in the DNB imagery, such as the cloud banding stretching SW-NE across much of northern Louisiana and Mississippi. Since the cloud bases in this imagery were mostly at MVFR and IFR levels with respect to aviation forecast concerns, knowledge about the details and characteristics of the low clouds are very important.
The next series of images from the New England region in the early morning hours of December 9th again demonstrates this application of the DNB imagery.
In the images above, notice that the extensive low cloud deck across the region that spans from Maine to at least as far south as northeastern North Carolina cannot readily be observed either in the legacy IR (10.8 µm ) imagery or in the Nighttime Microphysics RGB. However, more details about the low clouds can be discerned from the DNB imagery. Sure, cirrus clouds are optically thick enough to prevent viewing of any low clouds in the NY metro area. Nevertheless, the advantages of the DNB imagery for detecting low clouds beneath thin cirrus can clearly be seen. Again, as expressed earlier, this type of imagery certainly offers application for aviation forecasting, in particular.
Lastly, here are some observations from just this morning (Dec 10th) over the Rio Grande Valley region.
In the VIIRS IR image (Image 6) just as in previous IR imagery the cirrus clouds obscure the presence of any clouds beneath. However, the patchy low clouds in eastern New Mexico can be much more easily seen in the DNB imagery. In the area between Midland, TX (KMAF) and Fort Stockton (KFST), a forecaster might have made the assumption that the low clouds were continuous based on the observations alone and without the aid of the DNB imagery. Yet, what becomes noticeable in the DNB imagery is that a gap exists in the low cloud deck.
Of course, with all of this said, the availability of the imagery severely limits its application for operational forecasting and analysis. Generally, only one or two passes are available over a given location on any night. Also, due to moonlight limitations, the imagery are only available for about half of the month…at best. I can only lament that the DNB imagery will not be available on a geo-stationary platform (at least anytime soon). Nevertheless, understanding the limitations of the imagery while also appreciating its advantages can offer operational utility when applied properly to a forecast challenge.
So, with the moon now passing into the waning crescent phase, the Day-Night Band imagery is less operationally useful, at least for the detection of fog and other lower level cloud types. That is, at least until the moon is back into the waxing gibbous phase. Nevertheless, when cirrus clouds aren’t present, the Nighttime Microphysics RGB has proven to be a very valuable tool for the detection of fog and other low-level clouds. Just this morning a forecaster at the Huntsville, AL WFO was able to use the imagery not only for the detection of fog, but also to aid in the issuance of a special weather statement about the fog. The image below valid at ~724 UTC (0224 am CDT) 17 Oct shows the fog (whitish-aqua colors) lying across the valley areas of NE Alabama and adjacent areas of southern Tennessee and NW Georgia.
Around the time of this image, the visibility in the foggy locations had decreased to ~1/4 – 1/2 SM or less. Notice the fog in the DeKalb Valley is fainter than the fog in areas to the north and west. Not only is the DeKalb Valley more narrow, but the fog was likely more shallow. This feature of the imagery can also help to guide forecasters in assessing the longevity of the fog once sunrise breaks. Over time, forecasters can develop a sense of pattern recognition with the varying degrees of color shading and tailor forecasts to better match the time of dissipation. In this case, the fog in the DeKalb Valley began to dissipate significantly by about 1430 UTC, while the deeper and more expansive fog to the north and west lasted about an hour longer.
I wanted to follow up with this sooner, but I’ve been rather busy preparing for the NWA conference. Yep, I tend to be one of those “last-minute” people. Anyway, just last week I posted an example showcasing the advantage of the VIIRS Day-Night Band Radiance RGB in detecting fog through cirrus clouds. The next day, SPoRT collaborator Doug Schneider (Morristown WFO), followed up with another great example that I wanted to share. I’m just going to take an excerpt from an email he sent to the Morristown staff…
“I know I’ve been sending out quite a few emails lately about NASA SPoRT products, so please bear with me, but I thought this was a great example of how they can add value.
I’ve mentioned the MODIS/VIIRS Fog product before, but sometimes there are better products available for identifying fog, especially when thin cirrus are present. In the MODISVIIRS Fog product image that is attached, you can see that it is difficult to see the extent of fog. There’s clearly some in the Sequatchie Valley, but there is also some in the NE TN/SW VA area that can’t be easily seen.
The attached MODIS/VIIRS Nighttime Microphysics product also shows fog in the southern areas where it is clear (fog is light blue colors), but cirrus obscures NE sections.
The VIIRS Day/Night Band Radiance RGB product does the best job showing the extent of fog. Fog is clearly identified in the valleys of SW VA and NE TN, despite the presence of cirrus. The extent of fog is also more easily seen in the central and southern valley areas.
The attached menu image shows where you can find the Day/Night Band Radiance RGB product under the Satellite -> NASA SPoRT -> Polar Imager -> MODIS/VIIRS -> SR East menus.
Remember that these products are from a polar-orbiting satellite, and may only be available once a night, usually between 07z and 10z.”
I’ve included the images he referenced below, and circled the specific area of fog in NE Tennessee / SW Virginia that he was mentioning.
This is exactly the kind of inter-office sharing SPoRT is looking for from our close partners. We greatly thank Doug and the Morristown office for their collaborative efforts!