 |
| Titel |
Quantitative analysis of X-band weather radar attenuation correction accuracy |
| VerfasserIn |
A. Berne, R. Uijlenhoet |
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| ISSN |
1561-8633
|
| Digitales Dokument |
URL |
| Erschienen |
In: Natural Hazards and Earth System Science ; 6, no. 3 ; Nr. 6, no. 3 (2006-06-02), S.419-425 |
| Datensatznummer |
250003505
|
| Publikation (Nr.) |
 copernicus.org/nhess-6-419-2006.pdf |
|
|
|
|
|
| Zusammenfassung |
| At short wavelengths, especially C-, X-, and K-band, weather radar signals
are attenuated by the precipitation along their paths. This constitutes a
major source of error for radar rainfall estimation, in particular for
intense precipitation. A recently developed stochastic simulator of range
profiles of raindrop size distributions (DSD) provides a controlled
experiment framework to investigate the accuracy and robustness of
attenuation correction algorithms. The work presented here focuses on the
quantification of the influence of uncertainties concerning radar
calibration, the parameterization of power law relations between the integral
variables (radar reflectivity Z and specific attenuation k), and total
path integrated attenuation (PIA) estimates at X-band. The analysis concerns
single frequency, incoherent and non-polarimetric radar systems. Two
attenuation correction algorithms, based on a forward and a backward
implementation respectively, are studied. From DSD range profiles, the
corresponding profiles of integral radar variables are derived. Using a Monte
Carlo approach, the accuracy and robustness of the two algorithms are
quantified for the different sources of error previously mentioned. This
framework of realistic DSD variability provides a robust way to confirm that,
under realistic assumptions concerning the PIA estimation uncertainty, the
forward algorithm outperforms the backward algorithm for PIA values below
10 dB. |
| |
|
| Teil von |
|
|
|
|
|
|
|
|