Аннотации:
© SGEM 2018. The first results of using adaptive regression modeling (ARM-approach) in description of the latitude periodic change are analyzed. A multicomponent (polyharmonic) model of the latitude part observed dynamics containing between 14 and 21 harmonics requires interpretation and autoregressive forecasting of the 6th order providing high accuracy for the period of two years. This article provides the analysis, modeling, and forecasting of the latitude periodic change using the “Automated System for Robust Modeling” (ASRM-2017) package designed for geoprocessing and it is representing a system in which algorithms of time series analyzing and processing are implemented on the basis of adaptive regression modeling (ARM-approach). It is considered that in almost constant orientation of the Earth's rotation axis relative to “fixed stars” the Earth body moves around the axis of rotation. The points of intersection of the rotation axis with the surface of the Earth are geographic Earth's poles and they move on the Earth’s surface since the Earth oscillates relative to its axis of rotation. According to the accepted terminology, this phenomenon is called periodic motion of the Earth's poles. The amplitude of the pole “swinging” on the circumpolar surface can reach a few dozens of meters. This is a significant value for today's practical applications depending on the quality of coordinate and time support. Therefore, the motion of the poles will certainly be taken into account in modern navigation, astrometry, geodesy, etc. The work also deals with the correlation between the Earth's pole motion and strong earthquakes.
