| Abstract: | An experiment is described dealing with the detailed mapping of the vertical velocity V of recent crustal movements in the central Danubian riverside region of the Moesian platform in Northern Bulgaria. The error in V being significant here, the problem of statistical control on the reliability of the initial data and results of such studies is investigated; this question is also relevant for other platform regions in Eastern and Central Europe. With the aim of assessing the significance of a movement, the sampling distribution of V is used. The problem is reduced to a checking of a standard statistical hypothesis and to a probabilistic characterization of V. In order to estimate the statistical significance of the relative velocity δV between neighbouring benchmarks, the distribution of the module G of the horizontal gradient of V is studied. An agreement between the sampling distribution obtained and the theoretical Weibull distribution is established. A sufficiently high probability P>86% is determined for the significance of δV in this region and an optimal interval of the isobases is suggested. The experiment itself embraces three variant solutions: independent mapping of V, independent mapping of G, and mapping of V with respect to G. The analysis and comparison of the results obtained demonstrate the high efficiency of the third solution. |
