I'm using code from a Fortran program called solid.f  that I converted to C using F2C and modified to use more accurate sun / moon positions.  The solid.f  program is based on an edited version of the dehanttideinelMJD.f source code provided by Professor V. Dehant. This code is an implementation of the solid earth tide computation found in section 7.1.2 of the IERS Conventions (2003) , IERS Technical Note No. 32.

http://geodesyworld.github.io/SOFTS/solid.htm

The gravity offset calculation is based upon TVB's  tides.c:

http://www.leapsecond.com/tools/tides.c

"calculated" the offsets (using known values from somewhere), or "measured"
it by very long term averaging of the GPS position information?

Hi

There are some interesting interactions between liquid (sea) tides and solid tides. Big blobs of water sloshing around in the ocean
really close to shore do turn out to have an effect. Not a big deal for those of us wise enough to live well inland :) Probably a lot of fun
to plot and validate if you are in the “right” location.

One gotcha pulling this stuff from precision survey data: The post processing software very much knows about these things and
“helpfully” scrubs them out of your data for you. That’s a not ideal if your goal was to validate the calculations. It sort of is if their
result comes out as zero, but that’s not very exciting.

Bob

Mark,

There is misunderstanding on this topic so let me clarify.

1. The quasi-periodic relative motions of earth / moon / sun mean that the acceleration-of-gravity, aka little 'g' (~9.8 m/s^2), varies a tiny bit over hours, days, and months, roughly down at the 0.1 ppm level.

2. Since the period of a pendulum is approximately T = 2pi*sqrt(L/g), a tiny change in g results in a tiny change in T, at least in theory.

3. But as we all know from working with quartz and atomic oscillators there are many sources of frequency instability in a clock; they are generally additive; they tend to follow power-law noise types, and they are highly dependent on the averaging interval, or tau. The ADEV statistic nicely displays all of this in a single log-log plot. At any given tau a few noise sources are dominant; the rest are "lost in the noise".

For example, if your ADEV is 1e-6 then a 1e-9 2-g turnover effect is meaningless. Or if you have a 1e-9 tempco, then 1e-11 power supply variations are insignificant. And so on.

4. For most [*] pendulum clocks, the internal and external instability (noise) far exceeds the tiny effect of tides. Thus they do not and cannot "detect" tides. This is not a "big issue".

[*] To put this in perspective, so far we have digital records of only two pendulum clocks ever made that were stable enough so that the effect of tides is visible in their timekeeping. More info here: http://leapsecond.com/hsn2006/

/tvb

----- Original Message -----
From: "Mark Sims" holrum@hotmail.com
To: time-nuts@febo.com
Sent: Tuesday, June 20, 2017 4:43 PM
Subject: [time-nuts] Plate Tectonics was: GPS Antenna on Tower.