Tufte Cloud Chart for Seattle
In The Visual Display of Quantitative Information Edward Tufte includes a chart of annual sunshine as an example of how to pack a lot of information into very little space. Here is the original chart, it was made for the year 1967 for a village in England.
I could not find any data on sunniness to recreate it, but I could find data on cloudiness. At every airport, there are ASOS machines that measure a number weather conditions and create a METAR weather report which pilots use to determine if they should fly or not.
One consideration when flying is how much of the sky is taken up by clouds and the height of the clouds (I’ve been told that it is extremely disorienting to fly a plane into a cloud). METAR includes the Oktas, or eighths, of the sky covered in clouds and the height of those clouds. This is what I used to create the chart, the more clouds, the darker I made the square
I found a cleaned up version of METAR from NOAA that was already formatted in csv and also included sunrise and sunset information. I pulled that into pandas before plotting it manually (I couldn’t figure out a way to achieve the look I wanted with Matplotlib).
The source data for any airport can be found here: https://www.ncdc.noaa.gov/data-access/land-based-station-data/land-based-datasets/quality-controlled-local-climatological-data-qclcd.
My code for generating this chart can be found here: https://github.com/chaltonj/TufteCloudProject/blob/master/DrawClouds.ipynb
