Open AccessComment on ‘Explicit, approximate expressions for the resolution and a posteriori covariance of massive tomographic systems’ by G. Nolet, R. Montelli and J. Virieux
Author(s) -
Yao Z. S.,
Roberts R. G.,
Tryggvason A.
Publication year - 2001
Publication title -
geophysical journal international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.302
H-Index - 168
eISSN - 1365-246X
pISSN - 0956-540X
DOI - 10.1111/j.1365-246x.2001.00346.x
Subject(s) - inversion (geology) , resolution (logic) , inverse theory , projection (relational algebra) , tomographic reconstruction , algorithm , matrix (chemical analysis) , mathematics , computer science , calculus (dental) , artificial intelligence , iterative reconstruction , geology , telecommunications , medicine , paleontology , dentistry , structural basin , surface wave , materials science , composite material
Summary When solving tomographic inversion problems, the resolution matrix for the equation system can provide very useful information on how well the model parameters can be resolved. Recently, Nolet et al. (1999)—referred to as NMV hereafter—proposed a one‐step projection method to estimate the resolution matrix for massive tomographic problems. It was claimed by NMV that the approximations inherent in the one‐step approach are of limited significance. We point out some of the problems associated with these approximations, and question whether the issue of the level of approximation has in practice been sufficiently addressed. We also suggest a modified approach that avoids some of the possible problems.