304
Multiscale Hydrologic Remote Sensing: Perspectives and Applications
Accurate QPE requires accurate informative radar measurements. Traditional
weather radars measure the single-polarization radar reflectivity factor (Z), which is
used for QPE. Because this observational information is limited by several aspects,
the accuracy of QPE is constrained by such limitations as well (Atlas and Ulbrich
1990). Since the 1970s, radar polarimetry has attracted intensive research interest in
the radar meteorology community (Seliga and Bringi 1976; Doviak et al. 2000; Bringi
and Chandrasekar 2001; Zhang et al. 2001; Brandes et al. 2002; Matrosov et al. 2002;
Ryzhkov et al. 2005a,b). The additional polarimetric radar measurements of differential reflectivity (Z dr or Z DR ), specific differential phase (K DP ), and copolarization correlation coefficient (ρ hv ) provide new insight into precipitation microphysics and allow
for more accurate rainfall estimation. Through its 30 years of research and development, radar polarimetry has matured as a valuable technique in QPE.
Many studies have shown that QPE can be improved with polarization diversity (e.g., Bringi and Chandrasekar 2001; Zhang et al. 2001; Brandes et al. 2002;
Matrosov et al. 2002; Ryzhkov et al. 2005a,b). The dual-polarization radar is gradually taking the place of the single-polarization radar in current operational networks (Doviak et al. 2000). For example, the dual-polarization upgrade of the U.S.
NEXRAD network began in 2009 and will be completed in 2012. The number of
polarimetric radars in the European network, OPERA, has also grown (Holleman
et al. 2008). QPE based on polarimetric radar data could become popular for major
operational radar networks in the near future.
The common methods for polarimetric radar–rain estimation are based on empirical relations or raindrop size distribution (DSD) retrievals (Bringi and Chandrasekar
2001). Polarimetric relations are normally in power-law form and can be regarded
as the revision of traditional R–Z relations with the polarimetric parameters Z dr and/
or K DP . Different combinations of the relations are usually recommended for different situations (Ryzhkov et al. 2005a). DSD retrieval was not attractive, because
radar reflectivity used alone allows for only a simple DSD model (e.g., Marshall
and Palmer 1948). With the introduction of dual- polarization observations without
sacrificing the variability of DSDs, more complicated models can then be applied to
retrieve DSDs, resulting in improved rain estimation (Ulbrich 1983). Recently, DSD
retrieval has become a hot topic for dual-polarization radar applications (Zhang et
al. 2001; Bringi et al. 2002; Gorgucci et al. 2002, 2008; Brandes et al. 2004a,b;
Anagnostou et al. 2008). DSD retrieval is also applied frequently in rain estimation
using dual-frequency radars (Meneghini and Liao 2007). Current DSD retrievals
generally apply a two-parameter model such as the exponential model, the constrained-gamma (C-G) model, or a one-parameter-fixed gamma model. This chapter
addresses the advancement of rain estimation using polarimetric radar data.
13.2 POLARIMETRIC RADAR MEASUREMENTS
When radar is used to measure precipitation, what it measures are compositive backscattering signals from hydrometeors within a radar resolution volume. Each particle
contributes to the total signal received by the radar, depending on its size, shape,
orientation, composition, location, and other factors such as temperature, radar frequency, antenna pattern, and scanning. As a result, the distribution of hydrometeors
Multiscale Hydrologic Remote Sensing: Perspectives and Applications
Accurate QPE requires accurate informative radar measurements. Traditional
weather radars measure the single-polarization radar reflectivity factor (Z), which is
used for QPE. Because this observational information is limited by several aspects,
the accuracy of QPE is constrained by such limitations as well (Atlas and Ulbrich
1990). Since the 1970s, radar polarimetry has attracted intensive research interest in
the radar meteorology community (Seliga and Bringi 1976; Doviak et al. 2000; Bringi
and Chandrasekar 2001; Zhang et al. 2001; Brandes et al. 2002; Matrosov et al. 2002;
Ryzhkov et al. 2005a,b). The additional polarimetric radar measurements of differential reflectivity (Z dr or Z DR ), specific differential phase (K DP ), and copolarization correlation coefficient (ρ hv ) provide new insight into precipitation microphysics and allow
for more accurate rainfall estimation. Through its 30 years of research and development, radar polarimetry has matured as a valuable technique in QPE.
Many studies have shown that QPE can be improved with polarization diversity (e.g., Bringi and Chandrasekar 2001; Zhang et al. 2001; Brandes et al. 2002;
Matrosov et al. 2002; Ryzhkov et al. 2005a,b). The dual-polarization radar is gradually taking the place of the single-polarization radar in current operational networks (Doviak et al. 2000). For example, the dual-polarization upgrade of the U.S.
NEXRAD network began in 2009 and will be completed in 2012. The number of
polarimetric radars in the European network, OPERA, has also grown (Holleman
et al. 2008). QPE based on polarimetric radar data could become popular for major
operational radar networks in the near future.
The common methods for polarimetric radar–rain estimation are based on empirical relations or raindrop size distribution (DSD) retrievals (Bringi and Chandrasekar
2001). Polarimetric relations are normally in power-law form and can be regarded
as the revision of traditional R–Z relations with the polarimetric parameters Z dr and/
or K DP . Different combinations of the relations are usually recommended for different situations (Ryzhkov et al. 2005a). DSD retrieval was not attractive, because
radar reflectivity used alone allows for only a simple DSD model (e.g., Marshall
and Palmer 1948). With the introduction of dual- polarization observations without
sacrificing the variability of DSDs, more complicated models can then be applied to
retrieve DSDs, resulting in improved rain estimation (Ulbrich 1983). Recently, DSD
retrieval has become a hot topic for dual-polarization radar applications (Zhang et
al. 2001; Bringi et al. 2002; Gorgucci et al. 2002, 2008; Brandes et al. 2004a,b;
Anagnostou et al. 2008). DSD retrieval is also applied frequently in rain estimation
using dual-frequency radars (Meneghini and Liao 2007). Current DSD retrievals
generally apply a two-parameter model such as the exponential model, the constrained-gamma (C-G) model, or a one-parameter-fixed gamma model. This chapter
addresses the advancement of rain estimation using polarimetric radar data.
13.2 POLARIMETRIC RADAR MEASUREMENTS
When radar is used to measure precipitation, what it measures are compositive backscattering signals from hydrometeors within a radar resolution volume. Each particle
contributes to the total signal received by the radar, depending on its size, shape,
orientation, composition, location, and other factors such as temperature, radar frequency, antenna pattern, and scanning. As a result, the distribution of hydrometeors