A theorem of Mœglin and Waldspurger for covering groups (Q2256310)

The author proves an analog of a well-known result for reductive \(p\)-adic groups (characteristic zero) for \(r\)-fold central extensions of such reductive \(p\)-adic groups. Let us describe this more precisely. The Harish-Chandra theorem gives an expansion of the character of an irreducible admissible representation of a reductive \(p\)-adic group near the identity. We remind that this expansion is a sum which varies over the set of nilpotent orbits and the summands (with complex coefficients) are Fourier transforms of suitably chosen measures on these orbits. In 1987, Mœglin and Waldspurger proved (generalizing to this context a result of Rodier) that the coefficients in this sum are the dimensions of certain spaces of degenerate Whittaker forms. Moreover, the leading non-zero terms (i.e. the maximal nilpotent orbits) appearing in the character expansion are precisely the maximal nilpotent orbits for which these spaces of the degenerate Whittaker forms are non-zero. Varma extended this result to the even residual characteristic. Recently, Li proved the existence of the so-called Harish-Chandra-Howe expansion of the character of an irreducible admissible representation of this kind of covering group of a reductive \(p\)-adic group near the identity; it looks the same as in the linear case. Now the author, using various splittings and careful choices of certain characters, manages to use Li's expansion to prove the Mœglin-Waldspurger theorem in the case of covering groups. We briefly describe the content of the paper: the first section is the Introduction. In the second section the author introduces a certain sequence of subgroups of a reductive \(p\)-adic group \(G,\) which form a basis of neighborhoods of the identity, and certain characters of these groups (which were introduced by Mœglin and Waldspurger and Varma). In the third section the author introduces the covering group \(\widetilde{G}\) which is of the main interest. He proves that there is a splitting of the cover over the subgroups introduced in the previous section. In the fourth section the degenerate Whittaker forms for \(\widetilde{G}\) are introduced. This is the adaptation of the results in the linear case using splitting. In the fifth section the author proves the main theorem using the subgroups introduced in the second section and certain subspaces of the representation space related to these subgroups, whose dimension is equal to the dimension of the space of the corresponding degenerate Whittaker forms.