exploring the feasibility of tropical cyclone intensity estimation using cloud radar and radiometer measurements

Open Access
- Author:
- Sieron, Scott Buku
- Area of Honors:
- Meteorology
- Degree:
- Bachelor of Science
- Document Type:
- Thesis
- Thesis Supervisors:
- Dr. Fuqing Zhang, Thesis Supervisor
Dr. Fuqing Zhang, Thesis Honors Advisor
Eugene Edmund Clothiaux, Faculty Reader - Keywords:
- hurricanes
tropical cyclones
intensity estimation
radar
satellite
CloudSat
WRF
OSSE
Hurricane Katrina - Abstract:
- This study evaluates using satellite-based cloud radar and Moderate Resolution Imaging Spectroradiometer Measurements toward a proposed technique of estimating the maximum sustained surface winds of a tropical cyclone. The use of this intensity estimation technique requires measuring the change in cloud-top heights and temperatures of deep moist convection in the eyewall and in the outer region. The cloud-profiling radar data give cloud-top heights and the specific locations of deep moist convection; however, these data are collected in only a single narrow swath. In order to test the extent to which the exact cross-sectional path across a tropical cyclone can impact the resulting diagnosed wind speed, this study uses Weather Research and Forecasting (WRF) model output of a simulation of Hurricane Katrina. Four cross sections through the storm center, simulating CloudSat radar observations, are constructed at several forecast times. The analyzed wind speeds from the four cross sections at a given forecast time are compared against each other to assess variability and are compared against the maximum surface wind in the model output to assess accuracy. Several combinations of different assumptions and techniques for the diagnosis procedure are tested. The results from all methodological combinations have too great variability or too great inaccuracy (or both) for the technique to be considered supplementary to existing intensity estimation techniques. As an alternative to an analysis constructed on cross sections, all cells of deep moist convection within a range of radii of the storm center are located and the cloud-top height and temperature of each cell are used in a wind diagnosis. The resulting dozens of wind speed diagnoses exhibit a similar magnitude of spread as the cross-sections. These results also indicate the impact of some of the physical limitations of the diagnostic technique. In addition to model output analyses, several real cases of CloudSat eyewall overpasses are analyzed; a similar degree of inaccuracy is seen in these results.