Conformal reduction of a spacetime metric (Q4699733)

The reviewer is somewhat puzzled by this paper. It is not clear what is claimed to be new. It concerns the situation of a space-time metric which admits a conformal Killing vector field and when the metric is conformal to another metric for which this vector field is a Killing vector field. It does not discuss the local or global nature of the problem and considers only a single vector field. However much more is known (and the author actually refers to a paper which contains extended results -- hence the reviewer's confusion!). The extended theory seems to have originated in [\textit{R. F. Biljalov}, Sov. Phys., Dokl. 8, 878-880 (1964; Zbl 0142.24002)] and was further developed in [\textit{L. Defrise-Carter}, Commun. Math. Phys. 40, 273-282 (1975; Zbl 0322.53008)]. The result quoted here is the strongest one known to the reviewer and is given in [\textit{G. S. Hall} and \textit{J. D. Steele}, J. Math. Phys. 32, 1847-1853 (1991; Zbl 0737.53022)] (see also [\textit{G. S. Hall, M. S. Capocci} and \textit{R. Beig}, Classical Quantum Gravity 14, L49--L52 (1997; Zbl 0877.53016)]). It is known that if a space-time with metric \(g\) admits a Lie algebra \(A\) of conformal Killing vector fields then the metric may be replaced locally by a new metric \(h\) which is conformally related to the original one and with respect to which each member of \(A\) is special conformal. If the original metric \(g\) is not conformally flat then \(h\) may be chosen (locally) such that each member of \(A\) is homothetic with respect to \(h\). If \(g\) is not locally a plane wave then \(h\) may be chosen locally such that each member of \(A\) is Killing with respect to \(h\). (The only clauses are a technical one excluding certain local symmetries and one related to the ``constancy'' of the orbit structure and Petrov type.) The results of the present paper seem to the reviewer to be special cases of the above. The author also introduces certain restrictions which do not seem to be important. For example his result 1 (page 2903) does not need the ``null'' restriction (though it is needed in his result 2) His results are local in nature as are those in the general theorem above but, actually, some global features of the result are known [\textit{M. S. Capocci} and \textit{G. S. Hall}, Gravitation Cosmology 3, 1-7 (1997)]. The author also ignores two rather important points. First, he disregards points where the vector field concerned vanishes. This is, in fact, perhaps the most difficult thing to deal with! Also, he ignores the fact that a conformal vector field is not necessarily null everywhere or non-null everywhere. His main result centres on a metric admitting a non-null conformal Killing vector, when a conformally related metric exists for which this vector field is Killing and when special coordinate systems exist in which the vector field takes a very simple form. The paper does, however, deal with some further details of the conformal reduction and some applications to cosmology which may be new.