On a free boundary value problem of physical geodesy, II (existence)

From MaRDI portal
Publication:4726764





[For part I see ibid. 7, 269-289 (1985; Zbl 0582.35114).] This paper, which is part of the author's Habilitationsschrift (Univ. GH Essen, 1982) continues a previous paper (part I). The determination of the figure of the earth is considered as a free boundary value problem for the Poisson equation \(\Delta u=\omega^ 2\), \(\omega\) denoting the rotational velocity of the earth. With a Hölder exponent \(\epsilon\in (0,1)\), it is proved that within a \(C^{1,\epsilon}\)-neighbourhood of some approximative boundary a boundary can be found such that u and \(\nabla u\) attain prescribed (measured) data there (which are supposed to be in \(C^{1,\epsilon}\) and \(C^{2,\epsilon}\) resp.). Increasing the regularity of the data, higher regularity of the solution can be shown. These results are comparable to those of \textit{L. Hörmander} [Arch. Ration. Mech. Anal. 62, 1-52 (1976; Zbl 0331.35020)], however by a Legendre transform, the use of a hard implicit function theorem can be avoided. The following slip was not detected while proof reading: from page 10 on, \(\epsilon\in (0,1]\) has to be replaced by \(\epsilon\in (0,1)\) of course.











This page was built for publication: On a free boundary value problem of physical geodesy, II (existence)

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4726764)