Polar orbiting and geostationary lake ice monitoring

Monitoring lake ice coverage over the Great Lakes via satellite is vital and affects shipping industries, tourism and recreation, especially over the winter months when ice develops, grows and expands over the lakes. According to the Great Lakes Surface Environmental Analysis (GLSEA) and NOAA CoastWatch, the total ice coverage between all 5 lakes is 80% as of 8 March 2019. GLSEA and NOAA CoastWatch’s diagram below highlights lake ice coverage (i.e. ice depicted as grey, dark grey and black colors) and the areas of open water seen via different shades of blue (i.e. represented via water temperatures, ~0°C-5°C).

Satellite observations, combined with other data sets, are vital in producing ice coverage percentages over the Great Lakes. On 8 March 2019, under moderate clear-sky conditions, polar-orbiting and geostationary satellites observe the Great Lakes at high spatial resolution. Note, geostationary observations express high temporal resolution as well, however polar-orbiting observations exhibit coarser temporal resolution. Satellite imagery and products are shown below and are provided from RAMSDIS Online, CIRA SLIDER and RealEarth.

S-NPP Day/Night Band (DNB) – Solar Reflectance (0.7um) at 1850Z, 8 March 2019

DNB solar reflectance acts like ‘daytime visible imagery’ (i.e. 0.64um) where DNB’s satellite sensor observes the solar reflectance from atmospheric or surface features that exhibit high albedos. DNB provides imagery (below) at 750-m resolution and shows open water, land, ice and clouds, above and around the Great Lakes. However, how can users differentiate between the aforementioned scene types?

S-NPP VIIRS Snow/Cloud Layers at 1850Z, 8 March 2019

Look no further than to the polar-orbiting VIIRS Snow/Cloud Layers product which is at 750-m resolution. Observing the same domain as DNB, the VIIRS Snow/Cloud Layers differentiates between land (green), snow and ice (white), low (yellow) and high (pink) clouds and bodies of water (dark blue/black).

GOES-16 Snow/Cloud Layers from 1832-1927Z, 8 March 2019

Now incorporate a similar product, except derived from the geostationary satellite, GOES-16, users can observe the Great Lakes at high temporal resolution. Temporal resolution is from the CONUS sector, that is, 5-minute geostationary data. Notice the lake ice movement (i.e. moving white features over the Great Lakes) along with the low and high clouds moving to the east. Lake ice motion can be seen more conspicuously over Lake Huron.

 

S-NPP VIIRS Flood Detection Product at 1900Z, 8 March 2019

Additionally, another polar-orbiting product that users can observe the Great Lakes and differentiate between surface and atmospheric features, is the VIIRS Flood Detection product. Product is at 375-m resolution, discriminates between different scene types: ice = aqua, supra-snow ice (water on top of ice, or melting ice) = purple, open water = blue, clouds = grey, snow = white, and land = brown. The product also calculates the floodwater fraction percentage from 0-100% (green-red colors) as well. VIIRS Flood product can be accessed in AWIPS-II via LDM.

Posted in POES | Leave a comment

3 March 2019 – Severe thunderstorm and heavy rainfall event

Posted in Miscellaneous | Leave a comment

Radar and satellite ‘snowfall rate’ observations

Forecasting snowfall and snowfall rates can be quite challenging, especially in radar-limited and or radar-deprived regions. A polar-orbiting satellite ‘Snowfall Rate’ product can be used together with radar observations to help anticipate snowfall rates, identify snowfall areal extent and snowfall maximas. To highlight the product’s capabilities, refer to the following snowfall case event over Northern Colorado, between 3-15Z, 2 March 2019 comprising of surface, radar and satellite observations.

Surface Observations over Northern Colorado from 3-15Z, 2 March 2019.

The 13-hr loop shows a decrease in air temperatures across Northern Colorado, exhibiting below-freezing temperatures. Over time, notice a surface low develop, just north of Denver, CO as southeasterly, upslope flow moved into the area. Additionally, the surface low in complement with an upper-level jet maxima (not pictured) and an increase in low-to-mid level moisture, produced enhanced snowfall totals over Northern Colorado.

Radar observations over Northern Colorado from 3-15Z, 2 March 2019.

Base Reflectivity radar observations (via Denver radar from the COD website) during the same time period, shows the evolution of higher reflectivity values (between 15-35 dBZ) observed over Northern Colorado. Notably from 11-13Z, a bright ‘snow band’ (an elongated reflectivity maxima) was observed, indicating heavy precipitation, or in this case, heavy snowfall over Larimer and Weld counties. But what snowfall rates are being observed? This is where the Snowfall Rate (SFR) product can be utilized.

Collocated Snowfall Rate (SFR) product and Radar Observations at ~11Z, 2 March 2019.

The SFR product is derived from passive microwave observations via polar-orbiting satellites, where SFR observations are displayed in units of liquid equivalent ‘inches per hour’. The image below is a direct comparison of SFR in relation to the radar (albeit, offset by two minutes) at ~11Z, 2 March 2019. Notice the line of higher liquid equivalent snowfall rates (0.04-0.1 inches per hour) correspond well with the bright ‘snow band’ seen in the radar. Additionally, the SFR product has the ability to observe the snowfall rate areal extent, seen throughout Wyoming, Nebraska and South Dakota via 1 satellite swath (i.e. DMSP). In contrast, the Denver radar exhibits a limited range, where an adjacent radar needs to be utilized to see snowfall occurring in nearby domains (i.e. users would need to refer to the Cheyenne, WY or North Platte, NE radars).

Snowfall Analysis and snow totals (ending ~9AM MST, 2 March 2019).

Via NWS/NOAA snowfall analyses, note the snow totals from this event, ranging from 0-6 inches at low elevations, and 6-plus inches at higher elevations.

For interested readers, NASA-SPoRT has an additional product, denoted as the Merged SFR product (i.e. incorporates radar and SFR together) that can be accessed online via the following link.

Posted in POES, Satellites, Winter Weather | Leave a comment

Low cloud / fog over snow covered ground on 25 February 2019

During the overnight hours of 25 February 2019, low clouds and fog developed over portions of northwest Kansas, eastern Colorado and southwest Nebraska.  The low cloud and fog developed over a field of snow on the ground from a recent blizzard.  Low cloud and fog on top of snow on the ground can be difficult to detect in some satellite imagery, while in other satellite imagery it is easy to detect, for example see this 4 panel GOES-16 imagery:

http://rammb.cira.colostate.edu/templates/loop_directory.asp?data_folder=training/visit/loops/25feb19/4panel&loop_speed_ms=100

The loop spans the nighttime to daytime hours.

Upper left is the GeoColor product.

Upper right is the Day Cloud Phase Distinction product.

Lower left is the nighttime microphysics product.

Lower right is the 10.3 micron (IR) band.

During daytime hours, note how difficult it is to discriminate between low cloud / fog versus snow on the ground in the GeoColor product, both features appear white.  However, during the daytime we can make the discrimination between low cloud / fog versus snow on the ground in the day cloud phase distinction RGB.  Snow on the ground appears green, why?  There is little contribution from Red (10.3), large contribution from Green (highly reflective at 0.64 microns) and small contribution from Blue (absorptive at 1.6 microns).  The low cloud and fog appears cyan since the contribution from Blue is larger (liquid water clouds reflect much more than snow on the ground at 1.6 microns).

The low cloud and fog during the nighttime hours is observed as bright green in the nighttime microphysics product and light blue in the GeoColor product.  It may be seen in the 10.3 micron band as well, but is much more difficult to detect due to the lack of contrast relative to the other 2 RGB products.  High clouds are also observed moving over the low cloud / fog region in northwest Kansas, these are observed as black or dark red colors in the nighttime microphysics product.  The high clouds acted to seed ice crystals into the low level clouds underneath, leading to snow flurries across the area.

Another product that shows all of this quite well is the CIRA Snow/Cloud-Layers product:

This loop spans daytime hours only since the discrimination between clouds versus snow on the ground can only be made during the daytime in this product.  Snow appears white in color, which may be more intuitive compared to other RGB products.  Bare ground is dark green, low clouds or fog are yellowish-green and high level clouds are pink.

Posted in Ceilings, Fog, Visibility | Leave a comment