Wednesday, 30 August 2023: 1:45 PM
Great Lakes BC (Hyatt Regency Minneapolis)
Supercells tend to produce the strongest and longest-lived tornadoes, causing insurmountable destruction of property, loss of life, and impacts to society. To mitigate these impacts, understanding the formation of supercell tornadoes is crucial. While there have been many studies that have focused on how supercell tornadoes form, it is largely unknown what differentiates storms that produce weaker and shorter-lived tornadoes versus stronger and longer-lived tornadoes. Some studies have found a correlation between updraft and differential reflectivity (ZDR) column width with tornadic strength while modeling studies have looked at environmental factors that influence tornadic strength and formation. This project aims to look at relationships between dual-polarization radar signatures and storm dynamics within a simulation ensemble framework. The use of simulations allows for high-spatial and temporal resolution that exceeds the capabilities of current, operational and research weather radars. With this in mind, the dataset consists of an ensemble of 27 tornado-producing supercells modeled in Cloud Model 1 (CM1). These tornadoes vary in strength and longevity to provide a variety of storms for analysis. To compare polarimetric signatures across cases, dual-polarization variables are obtained using the CM1 microphysical output with a parametrized forward operator. The forward operator allows for the computation of radar reflectivity factor (ZH), ZDR, specific differential phase (KDP), and co-polar cross-correlation coefficient (𝜌hv).
With the polarimetric variables and the CM1 model output, ZDR and KDP column height and width, the orientation of the KDP foot, and the location of and magnitude of values in the ZDR arc will be compared across ensemble members. Results from the first ensemble member show KDP foot and ZDR column area increase prior to vertical vorticity intensification. Following tornadogenesis, ZDR column area oscillates periodically in association with updraft pulses at 1 km above freezing level. Additionally, since few studies have applied statistical methods to look at leading patterns in radar data, this project will apply a new statistical method (Entropy Field Decomposition, EFD) to the simulated radar dataset. Applying methodology such as EFD to radar data and determining patterns across both dynamic and polarimetric radar variables is needed to objectively examine high-resolution radar and model datasets. In particular, objective methods that can analyze the entirety of very large, 4D radar or model datasets are especially needed. Being able to identify features and patterns unique to strongly tornadic supercells prior to tornadogenesis is crucial as it allows for the opportunity for forecasters to differentiate between the formation of weakly and strongly tornadic supercells. Ideally, this will lead to longer lead times for warnings in addition to providing extra forecast information for operational meteorologists.
With the polarimetric variables and the CM1 model output, ZDR and KDP column height and width, the orientation of the KDP foot, and the location of and magnitude of values in the ZDR arc will be compared across ensemble members. Results from the first ensemble member show KDP foot and ZDR column area increase prior to vertical vorticity intensification. Following tornadogenesis, ZDR column area oscillates periodically in association with updraft pulses at 1 km above freezing level. Additionally, since few studies have applied statistical methods to look at leading patterns in radar data, this project will apply a new statistical method (Entropy Field Decomposition, EFD) to the simulated radar dataset. Applying methodology such as EFD to radar data and determining patterns across both dynamic and polarimetric radar variables is needed to objectively examine high-resolution radar and model datasets. In particular, objective methods that can analyze the entirety of very large, 4D radar or model datasets are especially needed. Being able to identify features and patterns unique to strongly tornadic supercells prior to tornadogenesis is crucial as it allows for the opportunity for forecasters to differentiate between the formation of weakly and strongly tornadic supercells. Ideally, this will lead to longer lead times for warnings in addition to providing extra forecast information for operational meteorologists.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner