On the regularity of Hamiltonian stationary Lagrangian submanifolds (Q1711915)

In this interesting paper the authors study regularity of Hamiltonian stationary Lagrangian submanifolds of $\mathbb{C}^n$. The main result of this paper is a Morrey-type theorem for such submanifolds: if a $C^1$ Lagrangian submanifold is a critical point of the volume functional under Hamiltonian variations, then it must be real analytic. Their result shows that Morrey's theorem for minimal submanifolds admits a complete fourth order analogue. \par More specifically, let $\Omega\subset\mathbb{R}^n$ be a bounded domain and $u:\Omega\rightarrow\mathbb{R}$ be smooth. The volume of the gradient graph $\Gamma_u=\{(x,Du(x)) : x\in\Omega\}$ is given by \[ F_{\Omega}(u)=\int_{\Omega}\sqrt{\text{det}(I+(D^2u)^TD^2u)}dx. \] The corresponding Euler-Lagrange equation is \[ \int_{\Omega}\sqrt{\text{det}g}g^{ij}\delta^{kl}u_{ik}\eta_{jl}dx=0, \] where $g$ is the induced metric from the Euclidean metric on $\mathbb{R}^{2n}$. When $u\in C^4(\Omega)$, this equation is equivalent to $\Delta_{g}\theta=0$, where $\Delta_g$ is the Laplace-Beltrami operator on $\Gamma_u$ for the induced metric $g$ and $\theta$ is the so-called Lagrangian phase function for the gradient graph $\Gamma_u$, defined by \[ \theta=\text{Im}\log\text{det}(I+\sqrt{-1}D^2u). \] The first goal of this paper is the study of the regularity of submanifolds that locally are described by potentials solving the Euler-Lagrange equation. In particular, the authors show that if $D^2u$ does not have large discontinuities, then the potential $u$ must be smooth. Next, the authors prove that in certain cases, weak solutions to $\Delta_g\theta=0$ enjoy full regularity. \par The main geometric result of this paper, the Morrey-type theorem, is obtained by combining the two previous results.