Intracomparison tests of the FG5 absolute gravity meters

The FG5 series absolute gravimeters have an estimated instrumental accuracy of 1–2 µGal (1‐2 × 10 −8 ms −2 ). A number of instrument comparisons were conducted with six FG5 instruments over a period of 9 months; the predecessor series JILA4 instrument was also used. The standard deviation of mean g values (averaged over 24–48 h), as observed by different instruments, is 1.8 µGal. Observations taken within 48 h typically agree within 2 µGal, and the maximum observed disagreement is 6 µGal for two observations taken 37 days apart. Individual data sets of 100–250 measurements (16–42 min duration) show minimum standard deviations of 7–8 µGal, placing an upper bound on instrument noise. The data are not corrected for local water‐table effects or regional atmospheric loading, and thus the observed repeatability is probably an underestimate of the instrument accuracy. The observed instrument agreement is consistent with the instrumental accuracy estimate. Ignoring deficiencies in modeling environmental gravity signals, this accuracy level should allow resolution of vertical motions at the sub‐centimeter level.