Date of Award

2011

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Electrical Engineering

First Advisor

Homaifar, Abdollah Dr.

Abstract

This thesis is focus on developing pattern recognition techniques to trace the origins and propagation of the Pre-Tropical Storm Debby (2006) African easterly waves (AEWs) and mesoscale convective systems (MCSs) using satellite imagery. The results are used to verify a numerical weather prediction (NWP) model. The pre-Debby MCSs’ movement and formation needs to be precisely and objectively tracked. These MCSs could be generated over mountains in North Africa, such as Ethiopian Highlands (EH), Darfur Mountains (DF), and Asir Mountains (AS), going through complicated splitting and merging processes. Thus, an objectively analyzed MCS movement is essential not only to help provide data to verify numerical modeling results, but also to help understand the formation and propagation of the African easterly waves and MCSs. The technique used could be applied to other AEWs and MCSs leading to tropical cyclogenesis. This, in turn, will improve the NWP over the data sparse areas, such as over eastern and central North Africa. The accuracy of numerical simulations of pre-tropical cyclone (TC) AEWs and MCSs is improved by fusing data with different data fusion techniques. In order to provide enhanced information to help predict the weather, this thesis investigates various techniques of data fusion. In many cases, fusing data provides more accurate and complete initial data for the NWP models which will then reduce the errors in weather prediction.

Download

Share

COinS