| > which requires the satellites themselves to have precise clocks that are synchronized with each other. Yes, the satellites themselves need to keep track of their GR-induced drift relative to the other satellites. But we don’t typically describe this as needing a new timekeeping system (or new timezone, or whatever) for each satellite. We just say the GPS system makes corrections for GR, which it does. And in any case, this takes place on the satellites, not the lunar surface, which is what is mentioned in the announcement. > I thought an accurate clock would be necessary to derive a reference meter and kilogram measurement. So first, the fundamental SI definitions are almost never used in the field to make measurements. For the vast majority of field experiments, the measuring equipment is calibrated back at a lab using a reference system, which itself was calibrated using a chain of multiple intermediaries that eventually trace back to the fundamental SI definition. But the equipment necessary to connect to the fundamental definitions are extremely expensive and delicate, which is why it’s only done occasionally and in a few labs specialized for the purpose. Second, even if you were measuring absolute lengths in the field with the distance light traveled in 10^X ticks of a cesium atomic clock or whatever (as opposed to the much more mundane task of measuring relative length changes with interferometry), it would all work perfectly fine if you did it within a local region of uniform gravitational field. The only reason you need to worry about the difference in clock speeds is if you are comparing event timing being done in different places where the clocks are running differently (as happens with GPS, or something like the LISA gravitational wave experiment, but which is otherwise quite unusual). Like, don’t get me wrong, I can sort of imagine that NASA is actually planning such experiments. But then they should be mentioned in an announcement like this. |