It appears that there is no way of placing an event in time across a cultural discontinuity (aliens, bombed back into the stone age). Referencing an event in the past by an event in the sky is no good if the new culture has no record of that event.
For pulsar phases and radioactive decay, things get fuzzy quickly as you try to extrapolate back in time from present behavior. Also, the old culture had to have had the technology to make the phase or decay measurements, but we're talking about our own culture as it is today. Dating with Carbon 14 is not certain, even if the laws of atomic decay don't change (and we're talking about a time before proton decay becomes prevalent). C14 is assumed to be created at a constant rate by gamma rays in the upper atmosphere. When a living thing dies, it always has the same ratio of C14 to C12 because C14 is assumed to be distributed uniformly through the atmosphere. After death, the amount of C14 decays with a half life of about 6000 years. This is adequate for human history, but not on a cosmological scale. See http://en.wikipedia.org/wiki/Radiocarbon_dating for the full story. The rate of C14 formation depends on the strength of the Earth's magnetic field as it affects gamma rays. The amount of atmospheric C14 doubled during atomic testing. The distribution of C14 in the atmosphere is not uniform. And so on. Too bad. I was thinking that isotope dating could get you in the neighborhood, where you could fine tune the estimate with pulsar phase relationships -- if you could locate the pulsars and predict their behavior at that time. Bill Hawkins -----Original Message----- From: Magnus Danielson Sent: Saturday, October 30, 2010 10:42 AM To: [email protected] Subject: Re: [time-nuts] Time of death-Again On 10/30/2010 04:48 PM, Jim Lux wrote: > > > On Oct 30, 2010, at 3:31 AM, Javier Herrero<[email protected]> wrote: > >> El 30/10/2010 10:31, Hal Murray escribió: >>> >>> >>> Suppose you have to go back a zillion years. Now the fuzz on the period adds >>> to the fuzz on measuring an individual pulse. >>> >>> >> Not to forget that pulsar frequencies spins down as the energy that they emits is ultimately drawn from its rotational energy (PSR B1937+21 spins down at 1.05 x 10^-19 seconds per second). In a zillion years this could amount a bit of time (several hundred microseconds over the 2.29 x 10e8 years life of this pulsar if the spin down rate would have been constant - too much drift for a real time-nut ;) ) >> > > But if the decay rate is known, you could factor that into your calculation. In fact wouldn't specifying the time of an event as the observed phase of the pulsars be unique. That is if something is at a time of 75%A,22%B,54%C, does that specify a unique time and place? If you would have constant rates and three phase observables that would give you a unique time modulus the beating period of this oscillator combination. To solve it you would use the chinese reminder. However, the chinese reminder theorem does not handle case with changing rates. Also, the ability to predict deep into the futures is severely limited by the precision of our modelling of these sources and possible other effects playing in on them. Also, the precision by which we makes the measures is a limiting factor. Sorting all of this out would > Percent, here, is where the event occurred in the period between the pulses Degrees would have helped more. I think pulsars alone would be difficult for long-term, omni-positional timing. We haven't even touched on stellar movements over long time periods. Also, our observation will be in movement from the various sources, so doppler compensation would be required. It's not a completely unsolveable thing, but a lot of things shifting over time throwing in massive of unknown parameters which would need to be estimated with sufficient precision if we where to build from scratch the time and time-scale. Just agreeing on the definitions of the star-book to traverse cultural, time and position boundaries would be an interesting problem. Cheers, Magnus _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
