Project 3 will be on visualizing data with a strong geographic component, in particular tornadoes in Illinois.

View Alpha Release

View Final Release

Github Repository (Project Files)

Github Repository (Website)

Youtube Link

Progress

Weekly Summary: April 6, 2018

Guillermo Rojas Hernandez progress:

  • Look through project requirements and through project data
  • Began work on UI sketches for Alpha Release
  • Yang Hao progress:

  • Working on 5 issues of part C
  • Try to get some new idea from the design of other group's projects.
  • Organize data and integrate it with the requirements of project 3
  • Natasha Rice progress:

  • Created a git-repo for the project
  • Created a rdata file to hopefully speed up processings
  • Created an elementry UI design so work can easily be started
  • Began working on 4 C level issues
  • Siddharth Basu progress:

  • Created git-repo for the project website
  • Looking into data files of the project
  • Began to tackle some of C section issues
  • Experimenting with UI and individual module placement
  • Weekly Summary: April 13, 2018

    Guillermo Rojas Hernandez progress:

  • Made sketches of possible layouts for the leaflet map
  • Organized project requirements for section B
  • Uploaded test project to shinyapps server
  • Worked on adapting layout to new screen dimensions
  • Yang Hao progress:

  • Integrate FIPS data with tornado data
  • Find the way to calculate loss data which include discrete and continuous data.
  • Finished plots for injuries, fatalities and most hit counties.
  • Natasha Rice progress:

  • Created leaflet Map
  • Filtering by listed parameters
  • Filtering by magnitude
  • Switching time frames and distance measurements
  • Siddharth Basu progress:

  • Working on Destructive tornados data variables (B section)
  • Looking into data files of the project for interesting findings
  • Began to learn new leaflet and geosphere library uses and implementations
  • Weekly Summary: April 20, 2018

    Guillermo Rojas Hernandez progress:

  • Created initial UI design for the Alpha Release
  • Looked at sample shiny apps for ideas on how to structure the UI design and created new sketches for the final release.
  • Looked at different shiny mapping libraries to try to find the best libraries for Leaflet.
  • Yang Hao progress:

  • Add new features to data
  • Switch categorical data of loss to numerical values
  • Organize the loss data for different years
  • Updated loss value to tables and charts
  • Natasha Rice progress:

  • Converted Map into a new format
  • Allowed filtering by magnitudes
  • Allowed filtering color/width by several parameters
  • Began looking into color options for the map
  • Siddharth Basu progress:

  • Working on destructive parameters for section B
  • Managed project website and weekly progresses
  • Experimenting with new library uses and implementations
  • Sorting through the data for inconsistences and interesting findings

  • Data

    The data utilized was chosen by the professor of the course, Dr. Andy Johnson.

    For the application, we created bar charts and pie charts using the ggplot library once the data was processed into R Studio.

    Data Manipulation

  • The data was cleaned and reduced to csv, xml, and rds files given by Professor Johnson.
  • The attributes that a Tornado is classified with is magnitude, speed, time, length, which leads to some interesting findings
  • The attribute that makes a tornado destructive is classified by fatalities, injuries, cost loss, crop loss, etc.
  • The library we used for our animation is SliderInput Widget library for over years of number of tornados by count.
  • Libraries Used:

  • library(shiny)
  • library(shinydashboard)
  • library(ggplot2)
  • library(lubridate)
  • library(DT)
  • library(jpeg)
  • library(leaflet)
  • library(dplyr)
  • library(plotly)
  • library(tidyverse)
  • library(reshape)
  • library(geosphere)
  • library(measurements)
  • library(RColorBrewer)
  • library(scales)
  • library(lattice)
  • library(sp)
  • library(leaflet.extras)

  • Insights on How to Use

    File Sytem

  • The two R files: app.R and howtomakeaRdatafile.R which are taking the data from the .csv file
  • A javascript file for animation implementation
  • A css file for styling and proper sizing for class projector
  • Tordnados.rds file to store the metadata for a package and to store the database.
  • The Main Menu System

    About

  • Has developers information and link to this website
  • Tornado Facts

  • Tornado Facts tab contains Tornado Sightings, Tornado Paths, and Tornado injuries, fatalities, an
  • The animation tab has color scheme based on the this scale.
  • The last column is Destructive attributes(injuries, fatalities, loss) in Illinois with comparison to another state.
  • Illinois Tornado Facts

  • Illinois counties
  • Most Powerful Places during a tornado: Heatmap of high magnitude tornados
  • Top 10 most powerful tornados
  • Tornados by Distance from Chicago
  • Tornado Tracks

  • Have a series of view types, from Electric Night View to basic terrain or traditional grid map. Also including choosing a particular year.
  • Allows perspective for the user and connect back to how back to their daily life
  • The legend that filters out the parameters of the different attributes from the user's choice. Ability to compare states similutanous.
  • Parameters: Magnitude from 0 to 5, Zero being weakest and 5 being stongest on scale, detructive attributes that are negative outcomes of these tornados: fatalities, injuries, damage cost, and length of tornado movement.
  • Have Geographic control to either by State to State comparison or county breakdown.
  • Time

  • Ability change time from traditional 12 hour clock or Military hours
  • Unit

  • Ability change Unit of measurement from Imperial or Metric
  • Comparison State

  • Ability change current state being compared
  • Top 10 Most Destructive Tornados

  • The chosen parameters that we used to determine the arbitrary term "Destructive was: fatalities, injuries, cost loss.
  • Then we gave weights to each attribute: Fataliies was 50%, Injuries was 25%, and Cost Loss 25%.
  • We gave the highest weight to fatalities since human life should be above everything else.
  • Weighed Injuries and Cost Loss equal because both are recoverable.
  • Interesting Thoughts:

    Thought#1

  • This visualization module shows the tornados by the distance and the count of tornados with their individual magnitude classified
  • This could help users connect back to the counties surrounding Chicago and relate back.
  • Thought#2

  • Another interesting data table we created is the most hit counties in Illinois
  • Using the xml files to get usps codes and county names to return a count of tornados.
  • This gives the user perspective of the counties they will live in. So for our local counties like Will and Cook county are among the top 10 most hit counties.
  • Chicago for example is in Cook county and they are ranked 10th which shows that this metropolitan area are less effected by open areas like Will county which are suburbs with more farm lands and less high scaled structures.
  • Thought#3

  • The heatmap below is a heatmap of the magnitutes of tornados in each given month
  • The heatmaps above of monthly tornado power is broken into three periods 3 months apart starting at September, January, and May.
  • The transition from low level tornado magnitudes to high magnitudes in January adn back down to low level magnitudes in May
  • This level of transition is shows that the winter season has higher level magnitudes and is a critical time to be aware.
  • Thought#4

  • The map above contains three dates: 1974, 1976, and 1990. The anglulation and data spread between the states Illinois and Indiana to show distance and tornado patterns between our local states.
  • The tornado paths for 2003 are shown in the states: Illinois and Indiana. As you can see that Indiana patterns are minimum so no activity is happening from the Eastern states. So the weather patterns must be effected from the western and southern states assuming.
  • Authors