Comment of the Day: Sewage and Statistics EditionIn today's comments, we learned how to calculate the event horizon of a black hole, where to look for the very best in sci-fi inspired music, and one very surprising fact about where those between-year census figures come from.

In this post calling for the trade secrets from commenters fields of expertise, commenter BrigadierGeneral shared this tidbit from the world of civil engineering:

Many population estimates for cites between census years are based on the increase in volume of sewage handled by treatment facilities.

But just how does the statistically-derived sewage-usage figure compare with the painstakingly-collected census data collected every 10 years?

The correlation can vary. On census years, if the two numbers are not close we would look further into the reason with a demographer. If the census number is higher, than it could be a rural area that doesn't have the most robust infrastructure. If the census number is lower it could be an urban city where information on lower income areas of the city may be undercounted. Overall the number is usually close enough for predicting population growth or reduction. I wouldn't be surprised if the numbers we hear in the news about Detroit's population loss weren't based on sewage at least partially.