Project 2: Hurricane Data Analysis

Overview

This project explores the dynamics of hurricane impacts over time, leveraging the hurricNamed.csv dataset to analyze patterns in hurricane frequency, severity, and consequences. The analysis focuses on:

• Economic Impact: Assessing damage in 2014 USD and identifying influencing factors.
• Mortality Patterns: Correlating hurricane intensity (wind speed, pressure) with death tolls.
• Geographical Trends: Mapping hurricane activity across U.S. states.
• Gender of Hurricane Names: Investigating if naming conventions affect perceived or actual severity.

Key Components

Data

• Dataset: hurricNamed.csv
• Variables:
  • LF.WindsMPH (Maximum Wind Speed at Landfall in MPH)
  • LF.PressureMB (Lowest Pressure at Landfall in MB)
  • deaths (Number of deaths caused by the hurricane)
  • BaseDam2014 (Economic damage adjusted to 2014 USD)
  • Year (Year of hurricane occurrence)
  • AffectedStates (States affected by the hurricane)
  • mf (Gender of the hurricane name)

Analysis Tools

• R with libraries:
  • shiny for interactive web applications
  • ggplot2 for static and dynamic plotting
  • dplyr, tidyr for data manipulation
  • plotly for interactive plots
  • maps for geospatial visualizations

Project Structure

Shiny Application

• UI:

  • Overview filters for variable selection.
  • Tabs for different analysis types:
    • Scatter Plot for examining relationships between hurricane metrics.
    • Trends Over Time to visualize changes in hurricane characteristics.
    • Insights for deep dives into economic impacts.
    • Correlation Heatmap to identify variable interrelations.
    • Geospatial Map showcasing hurricane distribution by state.

• Server:

  • Reactive computations for dynamic data visualization.
  • Custom functions for data processing and visualization.

Static Visualizations

• Wind Speed vs. Deaths: Scatter plot to visualize the impact of wind speed on mortality.
• Pressure vs. Deaths: Similar analysis but with pressure as the variable.
• Hurricane Frequency by State: Bar chart showing hurricane counts over years for top states.
• Economic Damage Distribution: Frequency polygon to understand damage spread across states.
• Trends Over Time: Faceted plot showing multiple hurricane metrics' evolution.
• Impact of Hurricane Name Gender: Violin and Box plots to assess if hurricane name gender correlates with damage.

How to Use

Running the Shiny App

• Ensure you have R and RStudio installed.
• Install required packages with:

  install.packages(c("shiny", "ggplot2", "dplyr", "tidyr", "plotly", "maps"))
  

Viewing Static Visualizations

• Open the Rmd file in RStudio and knit it to generate an HTML document with all visualizations.