home » about nssl » research divisions » radar r&d » doppler radar & remote sensing
Polarimetric Doppler Radar
IDENTIFYING PROBLEMS AND FINDING SOLUTIONS:
Classifying Precipitation
PROBLEM: Conventional radars have trouble discriminating between different types of precipitation. Forecasters need a way to clearly identify different types of precipitation - like rain, hail or snow - within regions of a cloud.
SOLUTION: Polarized radar helps classify precipitation: Rain and different types of frozen precipitation have unique polarimetric signatures.
- Rain/snow transition lines (indicated by a bright band in the radar reflectivity data) provide reliable estimates of locations of rain vs snow for hydrologic runoff models.
- Melting snow has a unique polarimetric signature.
- Low bright band (freezing level) regions have a unique polarimetric signature.
- Freezing rain can be identified when polarimetric data is used in combination with surface temperatures.
SUCCESSES:
NSSL Hydrometeor
Classification Algorithm: The HCA
can tell the difference between ten types of radar echoes using different
radar variables:
- ground clutter / anomalous propagation (AP occurs when the radar beam is bent downward towards the earth due to inversions or a rapid change in dewpoint) and produces false echoes
- biological scatterers (insects and birds)
- dry snow
- wet snow
- crystals (horizontally or vertically oriented)
- graupel (soft hail)
- big rain drops
- light and moderate rain
- heavy rain
- rain / hail mixture
The key to the HCA is detecting the melting layer using polarimetric measurements. Once the melting layer is determined, the algorithm utilizes five radar variables in a fuzzy logic classification scheme to differentiate between the echoes.
Winter precipitation type classification (Schuur, Ryzhkov, Giangrande): Polarimetric radar data on a dozen winter weather events were collected, providing statistical information to help quantify the polarimetric characteristics of winter precipitation in Oklahoma. Regions of dry and wet snow, sleet, ice pellets, and freezing rain are all identified with polarimetric signatures.
WHAT'S NEXT: An automatic, polarimetric bright band detection algorithm to aid in the identification of the freezing level height in precipitation systems.