HUSCAP logo Hokkaido Univ. logo

Hokkaido University Collection of Scholarly and Academic Papers >
Theses >
博士 (環境科学) >

An observational study of tropical cyclone intensity estimation, intensity change processes, and intensity forecast

Files in This Item:
Udai_Shimada.pdf27.81 MBPDFView/Open
Please use this identifier to cite or link to this item:
Related Items in HUSCAP:

Title: An observational study of tropical cyclone intensity estimation, intensity change processes, and intensity forecast
Other Titles: 熱帯低気圧の強度推定、強度変化プロセス及び強度予報に関する観測的研究
Authors: 嶋田, 宇大 Browse this author
Issue Date: 25-Dec-2017
Abstract: An observational study of tropical cyclones (TCs) using radar and satellite data is made from the perspective of intensity estimation, intensity change processes, and intensity forecast with the aim of better understanding intensity change and further improving the accuracy of intensity forecast. First, a system using Doppler radar data is developed, where the TC wind field is retrieved and the central pressure is estimated at 5-min intervals. A verification of intensity estimates relative to the best track data shows that the accuracy of the system is comparable to or better than the accuracies of Dvorak and satellite microwave-derived estimates. Next, by using satellite-derived rainfall data and radar data, the statistical relationship between TC rainfall structure and intensity change, and intensity change processes in two TCs are investigated. It is found that, during the development stage, the higher the axisymmetricity (the degree of axial symmetry) of the rainfall distribution in the inner-core region, the larger the intensity change in the next 24 h, depending on the current intensity. The relative relationship between axisymmetric and asymmetric terms of rainfall shows that the larger the axisymmetric term and the smaller the asymmetric term, the larger the intensity change is. Strong vertical wind shear causes the structure of the eyewall to become asymmetric and is generally a hostile environmental condition for TCs to intensify. However, Typhoon Noul (2015) reintensified with the formation of a symmetric eyewall despite vertical wind shear greater than 10 m s?1. The reintensification began with convective bursts. The maximum azimuthal mean tangential wind at 2-km altitude increased from 30 to 50 m s?1 during only 5 h, associated with the increase in azimuthal mean reflectivity inside the radius of maximum wind (RMW). It was when the vortex vertically aligned through vortex precession upshear that a symmetric eyewall formed in strong shear. The radar analysis suggests that the vortex tilt, convective bursts, and subsequent intensification occurred triggered by the increase in shear in the presence of environmental conditions favorable for convection. After the completion of an eyewall replacement cyclone (ERC), a TC often continues to intensify and in some cases rapid intensification (RI) occurs. However, RI processes after an ERC have not yet been investigated. An operational ground-based Doppler radar observed the RI of Typhoon Goni (2015) for 24 h immediately after it completed an ERC. This provides an invaluable opportunity to examine Goni’s RI processes in detail. Around the onset of RI, relatively strong outflow outside the RMW above the boundary layer was observed, which contributed to rapid contraction of the low-level RMW, causing the RMW to slope greatly outward with height. The radial profile of tangential wind became more peaked with time. During RI, the low-level outflow changed into inflow just outside the RMW and the secondary circulation became well established. An updraft peak and the radius of maximum reflectivity between 2- and 9-km altitudes were located inside the RMW. Finally, the possibility of further improvement in TC intensity forecast is examined based on the findings that inner-core structural conditions govern subsequent intensity change. Five new predictors associated with rainfall distribution and structural features, including the axisymmetricity of the rainfall distribution, are incorporated into a multiple regression model, which predicts intensity up to 5 days ahead. Results show that the model produces a 2-7% improvement in the forecast skill, compared against a model without these predictors.
Conffering University: 北海道大学
Degree Report Number: 乙第7038号
Degree Level: 博士
Degree Discipline: 環境科学
Examination Committee Members: (主査) 准教授 堀之内 武, 教授 久保川 厚, 教授 谷本 陽一, 助教 川島 正行, 准教授 筆保 弘徳 (横浜国立大学教育学部)
Degree Affiliation: 環境科学院
Type: theses (doctoral)
Appears in Collections:学位論文 (Theses) > 博士 (環境科学)
論文博士 (Doctorate by way of Dissertation) > 環境科学院(Graduate School of Environmental Science)

Export metadata:

OAI-PMH ( junii2 , jpcoar )


Feedback - Hokkaido University