Sample Day 1 SPC Forecast: (05/18/2017 1200Z Day 1 Forecast)

Snippet of text file associated with the forecast:

Each 8 digit code is associated with a latitude-longitude coordinate, or an indication of continuation from one point on the border to another point somewhere else on the border, and represents a vertex of the polygon in the figure.

Associated plot from plot_categorical.py script:

Similar plots for the following events and day outlooks can be created as according to the table below:

• Cleaned up much of the codebase, including condensing five separate plotting scripts into a single script
• Created basic Tkinter GUI allowing for simple forecast lookup, only for most recent forecast so far

• Forecast lookup for any past archived outlook, with GUI support
• (maybe) Determining the set of risks for severe weather given a coordinate
• Optimize the code for efficiency and readability
• Provide GUI and command line support
• Add tutorial to README

• Implement Day 2, 3, 4-8 outlooks
  • Added categorical / severe probability plotting for days 2 & 3
  • Added separate plotting for days 4-8 forecast
• Ability to import text files from previous forecasts
• Added forecast text file lookup

• Eliminated need for JSON file handling by using arrays for each line
• Using shapely library to plot probability boundaries
  • https://pypi.org/project/Shapely/
• Goes to SPC website and pulls most recently available Day 1 convective outlook
• Plotting capabilities using matplotlib library and shapely library
  • Four separate plotting scripts ==> Tornado, Hail, Wind and Categorical probabilities
  • Uses information from 2010 census to plot contiguous US boundaries
    • https://www.census.gov/geo/maps-data/data/cbf/cbf_nation.html

• Pulling most recent Day 1 convective outlook from SPC
  • Saving text file in json format:
    • Currently fixing bug where coordinates are in unordered format
• Finds most recent json file in data folder and determines whether a set of coordinates is in any event/probability combination
  • Bug where discontinuation in json coordinates (see above) leads to false outcomes
• Shell script that determines time and whether or not a new forecast has been released
  • Very basic shell script, not much functionality