Monday, 28 August 2017: 4:45 PM
Vevey (Swissotel Chicago)
Manuscript
(2.7 MB)
Large hail is a frequently occurring sometimes devastating phenomenon, especially during summertime convective events. Many occasions of hail stones with diameters above 5 cm up to 14 cm have been reported and confirmed in the last years in central Europe.
Concepts and algorithms to detect and distinguish large hail from small hail or rain with polarimetric radars at S-Band have been discussed and published (e.g. Ryzhkov et al. 2013, Ortega et al. 2016), but not yet implemented for operational use e.g. in Germany. We use the German polarimetric radar network covering an area of more than 357000 km2 as testbed for hail size detection and discrimination at C-band. Scattering model simulations and two strongly overlapping X-band radars of Geoverbund ABC/J in Germany are utilized for the development of algorithms. For evaluation, observations of hail occurrence and size are taken from the national disdrometer network and the growing European Severe Weather Database (ESWD). Also, the use of Circular Depolarization Ratio (CDR), a potential, but seldom used polarimetric variable, will be explored together with the classical polarimetric moments differential reflectivity, differential phase and cross-correlation coefficient. While the strong differential attenuation at C-band allows for the identification of hail mixed with precipitation and for the discrimination between different hail sizes, it hampers quantitative precipitation estimation. Total signal extinction due to large, wet hail may occur, which makes the interpretation of signals beyond hail-affected range bins challenging. Therefore, a reliable attenuation correction is mandatory.
Testud et al. (2000) introduced the ZPHI method to correct attenuation in rain using the differential phase ΦDP as a constraint. Ryzhkov et al. (2013) presented a modified ZPHI method, which is capable to correct attenuation in hail at S-band but requires assumptions regarding the distribution of hydrometeor types involved along the radial in the shadow of hail. We will present extensions and improvements of the methodology, which allow for arbitrary numbers of segments containing either single or multiple hydrometeor types (e.g. rain, melting hail, dry hail or mixtures of these) in arbitrary order and even multiple hail-bearing cells (e.g. with different hail sizes) behind each other. This enables the algorithm to resolve attenuation correction with higher radial resolution and with any kind of rain and hail mixture. The method introduced additionally estimates backscatter differential phase δ using polarimetric variables (primarily total differential phase) and idealized attenuation estimates. This extension allows to improve the attenuation correction in and beyond the hailcore of convective cells. First evaluation results of the hail attenuation correction will be discussed along with the algorithm using disdrometer observations, ground truth provided by the European Severe Weather Database and backscattering simulations based on a two-layer T-matrix code for wet, melting hail.
Concepts and algorithms to detect and distinguish large hail from small hail or rain with polarimetric radars at S-Band have been discussed and published (e.g. Ryzhkov et al. 2013, Ortega et al. 2016), but not yet implemented for operational use e.g. in Germany. We use the German polarimetric radar network covering an area of more than 357000 km2 as testbed for hail size detection and discrimination at C-band. Scattering model simulations and two strongly overlapping X-band radars of Geoverbund ABC/J in Germany are utilized for the development of algorithms. For evaluation, observations of hail occurrence and size are taken from the national disdrometer network and the growing European Severe Weather Database (ESWD). Also, the use of Circular Depolarization Ratio (CDR), a potential, but seldom used polarimetric variable, will be explored together with the classical polarimetric moments differential reflectivity, differential phase and cross-correlation coefficient. While the strong differential attenuation at C-band allows for the identification of hail mixed with precipitation and for the discrimination between different hail sizes, it hampers quantitative precipitation estimation. Total signal extinction due to large, wet hail may occur, which makes the interpretation of signals beyond hail-affected range bins challenging. Therefore, a reliable attenuation correction is mandatory.
Testud et al. (2000) introduced the ZPHI method to correct attenuation in rain using the differential phase ΦDP as a constraint. Ryzhkov et al. (2013) presented a modified ZPHI method, which is capable to correct attenuation in hail at S-band but requires assumptions regarding the distribution of hydrometeor types involved along the radial in the shadow of hail. We will present extensions and improvements of the methodology, which allow for arbitrary numbers of segments containing either single or multiple hydrometeor types (e.g. rain, melting hail, dry hail or mixtures of these) in arbitrary order and even multiple hail-bearing cells (e.g. with different hail sizes) behind each other. This enables the algorithm to resolve attenuation correction with higher radial resolution and with any kind of rain and hail mixture. The method introduced additionally estimates backscatter differential phase δ using polarimetric variables (primarily total differential phase) and idealized attenuation estimates. This extension allows to improve the attenuation correction in and beyond the hailcore of convective cells. First evaluation results of the hail attenuation correction will be discussed along with the algorithm using disdrometer observations, ground truth provided by the European Severe Weather Database and backscattering simulations based on a two-layer T-matrix code for wet, melting hail.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner