Premium
THE MAGNETOTELLURIC IMPEDANCE TENSOR—ITS RANDOM AND BIAS ERRORS *
Author(s) -
PEDERSEN L. B.
Publication year - 1982
Publication title -
geophysical prospecting
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.735
H-Index - 79
eISSN - 1365-2478
pISSN - 0016-8025
DOI - 10.1111/j.1365-2478.1982.tb01298.x
Subject(s) - magnetotellurics , computation , noise (video) , random error , magnetic field , signal (programming language) , regional geology , random noise , tensor (intrinsic definition) , random field , electrical impedance , algorithm , computational physics , statistical physics , statistics , mathematics , physics , computer science , geology , hydrogeology , electrical resistivity and conductivity , artificial intelligence , geometry , geotechnical engineering , metamorphic petrology , quantum mechanics , image (mathematics) , programming language
A bstract Approaches to the reduction of bias in the computation of the elements of the magnetotelluric impedance tensor have been proposed in the past by several authors. In this paper a clear distinction is made between random errors and bias errors. No effort is made to reduce either, but the emphasis is on their estimation. Both types of errors depend critically upon the polarization of the magnetic field. The random error increases with increasing noise‐to‐signal ratio in the electrical field, and it is rather insensitive to noise in the magnetic field. The bias error increases with increasing noise‐to‐signal ratio in the magnetic field. Expressions for random errors and maximum bias errors are developed and discussed using a single station set‐up. Random errors with a reference station set‐up are also calculated.