NUCAPS, Part Two: Field Campaign and Observations

Currently at the Cooperative Institute for Research in the Atmosphere (CIRA), there has been a field campaign underway comparing satellite temperature and moisture soundings (also known as the NOAA Unique Combined Atmospheric Processing System, NUCAPS) to RAwinsonde OBservation (RAOB) soundings along the Colorado Front Range (Figure 1). The focus of the field campaign is to analyze NUCAPS soundings in a pre-convective environment, examine how they compare to surface-based observations (i.e., RAOB) and how NUCAPS soundings are beneficial for National Weather Service (NWS) forecasters.


Figure 1: CIRA personnel setting up a RAOB launch site located along the Colorado Front Range. Notice the balloon in the middle of the figure is accompanied by a radiosonde (not shown) that collects an atmospheric profile of environmental data (i.e., temperature, dew-point, wind and pressure) starting from the surface to the upper levels of the atmosphere.

Only a certain percentage of NWS Weather Forecast Offices (WFO) produce RAOB observations. From RAOB observations, this subset of WFO’s are able to analyze the current state of the atmosphere wherein they’re able to formulate the proper forecast for their WFO. In contrast, WFO’s that do not produce RAOB observations have to rely on either forecast models (e.g., HRRR, NAM, GFS soundings) or RAOB observations produced from nearby WFO’s to help them assess the current state of the atmosphere.

This is where NUCAPS soundings come into play and can help assist WFO’s that do not produce RAOB observations. NUCAPS soundings come in a series of swaths from the Suomi-National Polar-orbiting Partnership (Suomi-NPP) satellite overpasses that occur once during the early morning (local time) and once in the afternoon (local time). The NUCAPS soundings are approximately 50 kilometers apart (~31 miles) from one another. Figure 2 below shows the NUCAPS soundings (green) compared to where CIRA produced a RAOB sounding (red) in Colorado on 19 May 2016 at 20Z.


Figure 2: Highlights several NUCAPS sounding (green) that are found across the state of Colorado. It is also important to see where the nearest NUCAPS sounding is located relative to the produced RAOB sounding by CIRA (red) on 19 May 2016 at 20Z.  

The two soundings, the RAOB sounding produced by CIRA and the nearest NUCAPS sounding are then compared to show the similarities and differences. Note the distance between the two soundings of interest is less than 50 kilometers apart and can be visually seen in Figure 2. Additionally, NUCAPS soundings are ‘volumetric’ as compared to a ‘point’ sounding provided by RAOB observations.

Figure 3 below shows the comparison between the RAOB observation (left) and NUCAPS sounding (right). As previously mentioned, the date of comparison is 19 May 2016 at approximately 20Z. Both soundings show the pressure in hectopascals (hPa) on the vertical axis and the temperature (red line) and dew-point temperature (green line) in degree Celsius along the horizontal axis.


Figure 3: The comparison of the RAOB observation (left) and NUCAPS sounding (right) along the Colorado Front Range on 19 May 2016 at 20Z. 

One can see that the two soundings look similar to one another from the surface (~850 hPa) to the upper levels of the atmosphere (~200 hPa) inferring the satellite retrieval and the surface-based observation are in relative agreement on the current state of the atmosphere. Although it is only a small sample size, such agreement could increase confidence in the utility of NUCAPS for NWS forecasters in their daily operations.

However, an apparent difference to note is NUCAPS expresses a coarser vertical resolution than the RAOB observation. This can be seen in Figure 3 by the ‘smoother’ temperature/dew-point lines seen by NUCAPS in contrast to the rigid lines seen by the RAOB observation (not so smooth temperature/dew-point lines).The RAOB observation is also able to see an upper-level inversion near 500 hPa which NUCAPS does not clearly see due to the coarser vertical resolution. Furthermore, although it is not apparent in Figure 3, NUCAPS is limited in accurately seeing the lower-levels of the atmosphere due to the influence from clouds.

Now with the the limitations stated above, how can NUCAPS be modified to best represent atmospheric profiles similar to RAOB observations? First of all, NUCAPS data can be seen visually (e.g., Figure 3) in AWIPS-II, a forecasting and software display package that NWS forecasters use in their day-to-day operations. Secondly, within the AWIPS-II interface, NUCAPS data has a feature where it can be modified manually by the NWS forecaster to better represent the current state of the atmosphere. For the interested reader, this modification will be highlighted in the next blog update.

Posted in Miscellaneous | Leave a comment

Fort McMurray Wildfires and Near-Constant Contrast (NCC) Imagery

The Fort McMurray Wildfires started in the city of Fort McMurray, located in the northeastern part of Alberta, a province of Canada. The wildfires started 01 May 2016 and are still currently burning. The wildfires have burned over 1,200,000 plus acres of land and has reached into parts of western Saskatchewan. Over 2,400 plus homes and businesses were lost within the Fort McMurray area (The Globe and Mail and Estimated insured losses from the fires are between 3-7 billion U.S. dollars (Insurance Journal). According to the Washington Post, the wildfires have produced an estimated 85 million tons of carbon dioxide equivalent emissions as of 20 May 2016.

The sequence of the estimated fire perimeters can be shown through the animation below.


A subset of operations for National Weather Service (NWS) is focused on forecasting and monitoring wildfire potential and growth (i.e., Fort McMurray wildfires) during the daytime and nighttime. NWS forecasters can monitor wildfires utilizing visible imagery from satellite during the day however, monitoring wildfires during the night can become cumbersome. To assist NWS forecasters during the nighttime, polar-orbiting satellite data from the Suomi National Polar-orbiting Partnership (Suomi-NPP) satellite is considered. On-board the Suomi-NPP is the Visible Infrared Imaging Radiometer Suite (VIIRS) instrument that consists of 22 spectral channels, where one of those channels is the Day-Night Band (DNB).

The DNB provides the capability of observing night-time light emissions (e.g. wildfires) and atmospheric features across the globe and monitors the global distribution of clouds (Miller et al 2014). DNB has the ability to detect broad ranges of light intensities ranging from full sunlight during the daytime to faint atmospheric glow on moonless nights (8 orders of magnitude in radiance space). The broad range leads to difficulties in displaying an images without losing detail at either end of the radiance scale. Near-Constant Contrast (NCC) was developed to mitigate the enhancement issues utilizing a sun/moon model to convert DNB radiance values into a reflectance-like value. In short, the NCC is a derived product of the DNB.

Currently, NCC is available for NWS forecasters in the Advanced Weather Interactive Processing System-II (AWIPS-II); a weather forecasting display and analysis package. To access NCC data in AWIPS-II refer to Figure 1 below.


Figure 1: Screenshot of the AWIPS-II interface, where users can access NCC data under the ‘Satellite’ tab.

For the interested reader, a link to the Quick Guide for Imagery Enhancement involving NCC in AWIPS-II can be found below.

Furthermore, to highlight the capabilities of the NCC we will take a closer look at the Fort McMurray wildfires. Displayed screenshots of NCC imagery were taken before the fire and during the fire shown in Figures 2-4. Emitted light sources from the active fires, city lights, gas flares, and also atmospheric features such as clouds and smoke are seen in the satellite imagery.

NCC Imagery Before the Fire


Figure 2: NCC imagery was taken a week in a half before the initiation of the Fort McMurray fires at 1013Z, 18 April 2016. Note the city lights of Fort McMurray, the gas flares from Tar Island, and the clouds in the vicinity.


NCC Imagery of Fort McMurray Wildfire – 17 May at 0930 UTC


Figure 3: NCC imagery taken during the Fort McMurray fires at 0930Z, 17 May 2016. One can see the emitted light from the fires and fire perimeter line that is forming.


NCC Imagery of Fort McMurray Wildfire – 18 May at 0915 UTC


Figure 4: NCC imagery taken during the Fort McMurray fires at 0915Z, 18 May 2016. Note the active fires along the fire perimeter line and the large amount of smoke produced.


Lastly, a comparison was conducted between the estimated fire perimeter and NCC imagery for 17 May 2016 (Figure 5). Both images show relatively similar shape and size of the fire perimeter line. It is important to note the time stamps for each image in the comparison is offset by a few hours, however, the polygonal shape of the fire perimeter is still apparent.

Picture9Figure 5: The comparison between the estimated fire perimeter and NCC imagery for 17 May 2016. A polygonal shape of the active fire line perimeter is evident in both images.

Posted in Miscellaneous | Leave a comment