Turn the National Hurricane Center archives into individual GeoJSON files.
First, grab the historical data archives from the NHC data website and save them locally. Scroll down to the 'Best Track Data (HURDAT2)' section and find the two database files.
Note: The filenames below are the current ones as of this date (7/29/15), check the link above for updated versions each season.
$ wget -O atlantic.txt http://www.nhc.noaa.gov/data/hurdat/hurdat2-1851-2014-060415.txt
$ wget -O pacific.txt http://www.nhc.noaa.gov/data/hurdat/hurdat2-nencpac-1949-2013-070714.txt
Create an output directory, then create atlantic and pacific subdirectories in it:
$ mkdir -p output/atlantic
$ mkdir -p output/pacific
With the atlantic.txt and pacific.txt files in your current directory, run the process.py command:
$ python process.py
The script will run and put the resulting GeoJSON files in the output/atlantic and output/pacific directories. It will also create a manifest file that lists all the storms in both basins.