Reduction of a two‐gravimeter base network by linear regression
In the exploration environment, a primary application of gravity surveying is regional reconnaissance. The first step in such a survey is to establish a base‐station network. Since an error in the network will propagate to many stations in the subsequent survey, careful field work and accurate reduction of these data are particularly critical. Optimally, successive base stations are tied by minimum‐time loops using at least two meters read simultaneously. Using two meters has the obvious advantage of doubling the number of ties with minimal increase in time and cost. Erroneous readings are also much easier to detect and correct with two meters. Furthermore, the simultaneous operation of the meters allows calibrations of the two to be compared by computing a linear regression of the readings of one meter against the corresponding readings of the other. If the meter calibrations are identical, the regression line should have a slope of 1. A significant deviation from 1 indicates a systematic variation in calibration.