That answer makes a ton of assumptions, including if it is even an ongoing, continuous process with a useful enough replacement rate that can be used to compare

That answer makes a ton of assumptions, including if it is even an ongoing, continuous process with a useful enough replacement rate that can be used to compare on an annual basis. That's the more important reason to bring up timescales: if it takes millions of years to happen, estimates of "annual" production are a fun game, but just as likely to be statistical lies as actual useful on the ground facts. (Also, as a reply points out, there's thoughts that time periods after the Jurassic epoch, if not some point earlier in the mid-to-late Cretaceous, may be much less "fertile" in petroleum and coal-producing raw elements and that for the most part new "production" by "nature" has been "stopped" and we've long just been mining further and further back in time. Which is to suggest there is no real, meaningful replacement rate at all with which to discuss.) Given the timescales involved, I know I'm far more inclined towards pessimism that any discussion on replacement rate with respect to human lifetimes, especially on an "annual" basis rather than cumulative across the human epoch, is ignoring the forest for some statistically insignificant trees.