After running this project for roughly sixteen days, the CSV file containing all of the temperature measurements (from the MCP9808 sensor) is roughly thirty nine megabytes in size. Which is almost two and a half megabytes of data gathered every day; a little less than originally estimated.

What I have also noticed is every twenty four hour period has a distinct fingerprint or pattern, in terms of the temperature readings collected. This is validated with the visualizations provided for each calendar day using the HTML and JavaScript files available on GitHub. Further still, by referencing the same graphs, I am able to determine the exact minute when environmental services in my home are automatically turned on. All being rather interesting and revealing to me. 

Please see the screenshot below for an example of what is stated in the above paragraph:

This self-quantification project has been a lot of fun to build, document and maintain. I expect to be actively gathering temperature readings for a little less than two more weeks. Resulting in a total of four weeks worth of entries. When completed, I will be moving on to another similar project, to be documented via a Hackaday project page.