How Many Decimals of Pi Do We Really Need? (2016) (jpl.nasa.gov)

  >>> len("3.141592653589793") - 1
  16
  >> math.log(2**51) / math.log(10)
  15.35252977886304

The precision of pi used by NASA is suspiciously close the precision limit of IEEE 64-bit floating point integers, a reason which is not given the article :). Interesting nonetheless.

edit: formatting