Components of Springer fibers associated to closed orbits for the symmetric pairs \((\mathrm{Sp}(2n), \mathrm{Sp}(2p)\times\mathrm{Sp}(2q))\) and \((\mathrm{SO}(2n), \mathrm{GL}(n))\). I. (Q392120)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Components of Springer fibers associated to closed orbits for the symmetric pairs \((\mathrm{Sp}(2n), \mathrm{Sp}(2p)\times\mathrm{Sp}(2q))\) and \((\mathrm{SO}(2n), \mathrm{GL}(n))\). I.
scientific article

    Statements

    Components of Springer fibers associated to closed orbits for the symmetric pairs \((\mathrm{Sp}(2n), \mathrm{Sp}(2p)\times\mathrm{Sp}(2q))\) and \((\mathrm{SO}(2n), \mathrm{GL}(n))\). I. (English)
    0 references
    13 January 2014
    0 references
    This paper yields a description of the structure of certain (irreducible) components of Springer fibers associated to closed \(K\)-orbits in the flag variety of a complex classical group \(G\) for the pairs \[ (G,K) = (\mathrm{Sp}(2n), \mathrm{Sp}(2p)\times\mathrm{Sp}(2q)) \quad \text{and} \quad (G,K) = (\mathrm{SO}(2n), \mathrm{GL}(n)), \tag{*} \] where \(p + q = n\), which arise from the real Lie groups \(\mathrm{Sp}(p,q)\) and \(\mathrm{SO}^*(2n)\) and fall, respectively, in the two cases referred to as type C and type D. Analogous results for other pairs were established in two previous papers by the same authors [Represent. Theory 12, 403--434 (2008; Zbl 1186.22017); J. Algebra 345, No. 1, 109--136 (2011; Zbl 1254.22007)] and here we use the terminology and notation of the review of the second one. The authors give an algorithm to construct a generic element \(f\) in \(\mathfrak n^-\cap\mathfrak p\) for the pairs \((G,K)\) in (*) and define a sequence \(((G_0,K_0),(G_2,K_2),\dots,(G_{2m},K_{2m}))\) of group pairs, where \(G_0 = G\), \(K_0 = K\), \(G_{2j}\subset G_{2(j-1)}\), \(K_{2j} = K\cap\,G_{2j}\) for \(1\leq j\leq m\), as well as sequences \((Q_0,Q_2,\dots,Q_{2m})\), \((L_0,L_2,\dots,L_{2m})\), where \(Q_{2j}\) is a parabolic subgroup of \(K_{2j}\) and \(L_{2j}\) is the (reductive) Levi factor of \(Q_{2j}\); they also define a suitable reductive subgroup \(L_{2j+1}\) of \(K_{2j}\) and, only for type C, reductive subgroups \(\widehat{L}_{2j+1}, \widehat{\widehat{L}}_{2j+1}\) of \(L_{2j+1}\) for \(0\leqslant j\leqslant m\). Then they prove the main theorem of this paper: \(\gamma_{\mathcal Q}^{-1}(f)\) is the closure of \[ \left(\prod_{j=0}^mZ^{(j)}\right)M_mM_{m-1}\dots M_1M_0\cdot\mathfrak b, \; \text{where} \; M_j = \begin{cases}\widehat{L}_{2j+1}\widehat{\widehat{L}}_{2j+1}L_{2j} &\; \text{for type C}\\ L_{2j+1}L_{2j} &\; \text{for type D}\end{cases} \] and \(Z^{(j)}\) is a suited subgroup of \(Z_K(f)\).
    0 references
    flag variety
    0 references
    nilpotent cone
    0 references
    Springer fiber
    0 references
    associated cycle
    0 references
    discrete series representation
    0 references
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references