Annotation

In the course of the geodetic training ground of the Lviv National Agrarian University in 2009, a linear base of the 2nd category was constructed, intended for metrological certification of various geodetic instruments. It has some preliminary observations.

In order to identify possible altitudinal displacements, it is suggested to perform current cyclical observations. In the autumn of 2019, the following observations were made with geometric leveling.

The Observation Program provides five options for measuring excesses between baselines. The first option is a mid-level alignment between two adjacent points. The second variant is leveling along the 15-1 direction, the third one is in the direction of 1-15, the fourth option is leveling from one station from the middle on the left side of the base, and the fifth variant is leveling from one station from the middle on the right side of the basis.

The program uses a proven H-3 leveler and a special device, a steel ruler with a spherical level, attached to a power screw with forced centering.

The influence of the proposed methods (variants) of observations is considered, namely: "Readings according to the direction"; "Readings from the middle" on the results of observations. The homogeneity of the results of the observations by the proposed methods with respect to the grouping center (some average value of the required value) was investigated by means of the analysis of variance.

To test the hypothesis Н₀: when using the proposed methods, the systematic influence on the measurement results is absent or in one order, and the methods lead to the results of the same accuracy (alternative hypothesis Н₁ – systematic effects are significant, the methods are not equivalent) we used the Fisher test  to estimate the difference between the disparities  factor and residual  for significance level α = 0.01 and α = 0.005. It is determined that for all significance levels F<F_α.

It can be argued that the null hypothesis H₀ is accepted, that is, the influence of the above observation methods does not have significant systematic errors, and the variance estimates differ from one another due to random factors and the used methods lead to the results of the same accuracy.