Search

The Community Workshop on Developing Requirements for In Situ and Remote Sensing Capabilities in Convective and Turbulent Environments (C-RITE) was carried out to assist the National Science Foundation (NSF) in defining next generation technologies and observing capabilities best suited for studying convection and turbulence in the atmosphere. With the help of over one hundred investigators from the scientific community, the Workshop developed a series of recommendations to NSF.

Tropical cyclone (TC) intensity forecasts continue to be one of the most challenging aspects of hurricane prediction. However, through an innovative new tracking and analysis tool, a high resolution, real-time analysis and greater understanding of the complex dynamics of TCs is possible, both leading to better forecasting skill. The validity of this tool was analyzed in this research. Through the ground-based Velocity Track Display (GBVTD) technique, a single ground-based Doppler radar data can be used to retrieve the three-dimensional (3-D) kinematic structure of a TC. The GBVTD technique and GBVTD-simplex center finding algorithm formed the basis of the Vortex Objective Radar Tracking and Circulation (VORTRAC) software package for this purpose. VORTRAC is capable of automatic center tracking and retrieval of storm-scale circulation, structure, and central pressure tendency of a TC, for use in real-time forecasting and research. VORTRAC was utilized herein to analyze Hurricane Danny (1997). This analysis was then compared to the National Hurricane Center (NHC) best track (considered to be the "ground truth") and the manual GBVTD analysis previously performed on Danny. Additionally, the VORTRAC runs from two separate radars were compared. The VORTRAC software package was shown to be a viable tool for the retrieval of the detailed 3-D structure of a TC. VORTRAC was shown to agree well with observations and with manual GBVTD analyses, while efficiently processing data at a high temporal resolution. This promises to be a useful tool for both forecasting and research.

While it is common for tornadoes to occur in conjunction with land falling tropical cyclones (TCs), characteristics of the hurricane-spawned tornadoes themselves remain poorly documented. This study (i) documents and describes supercells embedded within the outer rainbands of Hurricane Katrina (2005) and (ii) compares Katrina’s supercell storms to past studies containing documented hurricane-spawned tornadoes. Radar reflectivity and velocity data collected on 29 August 2005 by WSR-88D (NEXRAD) radars in Slidell, Louisiana and Mobile, Alabama were used to track the supercells. Atmospheric soundings from 29 August 2005 showed that environmental conditions were comparable to those in previous hurricane-spawned tornado studies. Twenty-three storms from 0300 UTC - 0900 UTC were tracked, and single-Doppler radar analyses examined characteristics such as shear and rotational velocity. Storms were initially classified as mesocyclonic or non-mesocyclonic, and were then classified as tornadic if they reached 18 m/sec (F0 intensity) at any time during their duration. Because tornadoes are not resolvable several kilometers from the radar, the intensity of the mesocyclone was used to infer the occurrence of a tornado. Eighteen of Katrina’s supercells were classified as tornadic, three reached mesocyclone strength but never reached F0 intensity, and two storms never reached mesocyclone criteria. Remarkably, the majority of the supercells formed over the Gulf of Mexico and not over land, which contrasts with previous studies. Furthermore, the ground-relative speeds of these tornadic mesocyclones in the outer rainbands can reach strong Category 4 intensities despite sustained winds in Katrina’s only reaching Category 3 at landfall.

Dual-Doppler analysis of the 17 April 1995 data collected by the National Center for Atmospheric Research airborne Electra Doppler Radar during the Verification of the Origins of Rotation in Tornadoes Experiment was used to resolve the kinematic structure of the storm, which was used to perform a hail trajectory analysis. Hailstone embryos introduced into the right flank of the storm followed the cyclonic branch of the horizontal flow which was found to favor the largest hail growth. A placement of the axis of maximum vertical vorticity upstream of the updraft both directs particles into the updraft, and redirects particles into the updraft after initial descent. This configuration was found to allow for two passes over the updraft or recirculation, as described in past studies. These findings agree with the current understanding of hail growth within thunderstorms. It was found in this study that the placement of a deep, relatively weak updraft around the maximum cloud liquid water content level played a key role in supporting large hail growth in both size and quantity, as well as concentrating the hail fallout into a small area. A stronger maximum updraft that lifted the particles above the level of maximum liquid water content was found to provide a smaller concentration of large hail growth, as well as spreading the particle fallout over a larger area due to stronger advection in the midlevels of the storm. These findings seem at odds with the notion that the strength of the updraft alone determines final hail sizes.

Analysis of the 17 April 1995 data collected by the National Center for Atmospheric Research airborne Electra Doppler Radar (ELDORA) during the Verification of the Origins of Rotation in Tornadoes Experiment (VORTEX) was used to investigate the wind structure hypothesized in conceptual models of hail growth in supercell storms proposed by Browning and Foote in 1976. For the first time Doppler radar data resolved a horizontal vortex tube that resembles Browning and Foote's model. This supercell produced 1.75 inch diameter hail and an F1 tornado. In conjunction with 3-D variational analysis, ELDORA's data resolved fine scale convective structures and internal circulations within the supercell storm. The wind field and vertical vorticity calculation confirmed the existence of a mesocyclone-strength cyclonic rotation that coincided with the reported tornado. The horizontal vorticity and wind field also confirmed the existence of an anticyclonic horizontal vortex tube with a vorticity comparable to the mesocyclone. Particle trajectory analyses confirm that the horizontal vortex tube does support hail formation, but strong updraft regions also support hail formation. A delicate balance between hailstone embryo size, location, trajectory, and storm structure was realized.

A single Doppler radar wind retrieval technique, Ground Based Velocity Track Display (GBVTD), was used to retrieve the primary circulation of Hurricane Bret. Radar data from Brownsville (KBRO) and Corpus Christi (KCRP), TX from 22:22:32 UTC to 23:03:03 UTC were used in the application of GBVTD and the results compared to each other. Only the axisymmetric circulations (wavenumber 0) were analyzed for this project. GBVTD needed the circulation center to be specified within 2km of the true center for an accurate representation of the circulations, so a simplex center-finding technique was employed. The circulation centers returned were within 1km of each other, so were assumed to fit the 2km criterion. The tangential and radial velocities were retrieved using the simplex centers. The mean tangential winds derived by GBVTD showed the de-intensification of the storm after landfall equally well for both radars. However, KCRP consistently returned higher tangential velocities than KBRO. An explanation lies in the effect of a mean wind on the tangential velocity, where the cross-beam component of the mean wind added to the KCRP wind estimates and subtracted from the KBRO wind estimates. The radial winds retrieved had a similar problem, but with KBRO having stronger inflow where KCRP had strong outflow. An along-beam mean wind effect was hypothesized as the primary reason for this difference. The total axisymmetric wind retrieved by GBVTD for both KBRO and KCRP are consistent at all radii and altitudes.

The purpose of this article is to document the DORADE format version 1, which will be used to exchange data collected by the ELDORA/ASTRAIA airborne Doppler radar and the NCAR ground-based radars. The physical description and definitions of the DORADE format are in Section 2. Section 3 describes the data structure. Appendix A provides the schematics of the terminologies used in Section 3. For the purposes of this document, only detailed formats for radar data recording are discussed. Exchanged data should always be corrected as best as the facility making the tape can do, i.e., aircraft motion removed, range delay corrected, etc.