Degree Type

Creative Component

Semester of Graduation

Spring 2020

Department

Computer Science

First Major Professor

Carl Chang

Second Major Professor

William Gallus

Degree(s)

Master of Science (MS)

Major(s)

Computer Science

Abstract

Severe weather events are caused by underlying weather systems (aka morphologies) that often evolve into one another. The purpose of classifying them is to try to understand the correlation between each system and the severe weather events they cause. As such, there is no established 1-1 mapping between a morphology and a severe weather event.

For the purpose of this research, we considered the following convective storms:

● Flash Floods

● Hail

● Thunderstorm Wind (Damaging winds)

● Tornado

The primary goal of the project is to gather more data with community support so as to develop a correlation between the weather systems and the severe weather events over a prolonged period of time. We approached this problem on two fronts: A full stack application that we developed that provides the users a platform to classify the radar images across 9 regions in the United States; A smaller application hosted on Zooniverse (A citizen science platform).

We recognize that the classifications are subjective and are often ambiguous. So as to make the user’s classification data consistent, we also normalize the classifications if we deem their classifications based on how the systems typically behave. An improved understanding of the relationship between storm morphology (classifications) and severe weather reports could help us specify better the severe hazards during convective storms (thunderstorms) and could thereby lead to a better understanding of the behavior of convective storms.

Copyright Owner

Subramanian, Krishnan

Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License

File Format

PDF

Share

COinS