On the gravimetric contribution to watt balance experiments

It has been recommended that the relative standard uncertainty of the numerical value of the Planck constant required for the redefinition of the kilogram should not exceed 2 × 10 −8 . To reach this goal using experiments based on a watt balance, the free-fall acceleration ( g ) traceable to the SI, at a given point and a given time, needs to be known with a sufficiently small uncertainty well below 2 × 10 −8 . Reducing the uncertainty in g allows the other uncertainties related to the watt balance to be increased. Instead of a simultaneous operation of an absolute gravimeter with a watt balance, we propose an alternative approach and demonstrate that a standard uncertainty below 5 µGal (relative uncertainty of 5 × 10 −9 ) is reachable under the conditions at BIPM. Further decreasing the uncertainty could significantly increase commitments in terms of personnel and equipment and would not significantly improve the uncertainty targeted for the BIPM watt balance experiment. A 5 µGal uncertainty might also satisfy the needs of other watt balance experiments underway or planned. In our approach we combine the following information: (1) the Key Comparison Reference Values obtained from the CCM.G-K1, a key comparison carried out in the frame of the International Comparison of Absolute Gravimeters in 2009 (ICAG2009); (2) the accurate gravity network established using the qualified absolute and relative gravimeters; (3) temporal gravity variations based on observed Earth-tide parameters and modelled effects of polar motion and atmospheric mass redistribution; (4) uncertainty estimates that account for non-modelled effects; (5) the option to carry out absolute gravity measurements once every one or two years with two or more gravimeters for monitoring the stability of the gravity field at the BIPM.