Physical interpretations of integration constants and large gauge effects in flat and AdS spacetimes

Abstract

As in other partial differential equations, one ends up with some arbitrary constants or arbitrary functions when one integrates Einstein's equations, or more generally field equations of any other gravity. Interpretation of these arbitrary constants and functions as some physical quantities that can in principle be measured is a non-trivial matter. Concentrating on the case of constants, one usually identifies them as conserved mass, momentum, angular momentum, center of mass, or some other hairs of the solution. This can be done via the Arnowitt-Deser-Misner (ADM)-type construction based on pure geometry, and the solution is typically a black hole. Hence, one talks about the black hole mass and angular momentum etc. Here we show that there are several misunderstandings: First of all, the physical interpretation of the constants of a given geometry depends not only on pure geometry, i.e. the metric, but also on the theory under consideration. This becomes quite important, especially when there is a cosmological constant. Secondly, one usually assigns the maximally symmetric spacetime, say the flat or the (anti)-de Sitter spacetime, to have zero mass and angular momentum, and linear momentum. This declares the maximally symmetric spacetime to be the vacuum of the theory, but such an assignment depends on the coordinates in the ADM-type constructions and their extensions: in fact, one can introduce large gauge transformations (new coordinates) which map, say, the flat spacetime to flat spacetime but the resultant flat spacetime can have a nontrivial mass and angular momentum, if the new coordinates are such that the metric components do not decay properly. These issues, which are often overlooked, will be examined in detail, and a resolution, via the use of a divergence-free rank $(0,4)$-tensor, will be shown for the case of anti-de Sitter spacetimes.