Here are some references for my talks at the Nancy scalpotmas2026 Masterclass.
Papers I will be discussing the content of:
Harmonic Functions and The Mass of 3-Dimensional Asymptotically Flat Riemannian Manifolds
Rigid comparison geometry for Riemannian bands and open incomplete manifolds
Stability of Euclidean 3-space for the positive mass theorem
Here are supplementary resources:
Bartnik's paper on the mass of asymptotically flat manifolds
Gromov and Lawson's big IHES paper on positive scalar curvature manifolds
Dan Lee's textbook on mathematical (Riemannian) General Relativity
Daniel Stern's paper on scalar curvature of 3-manifolds and harmonic maps to S^1
My paper with Allen-Bryden-Khuri on the suboptimal Penrose inequality
Schedule of talks, subject to change:
Day 1: Introduction to scalar curvature, overview of minimal surface and Dirac operator methods, fundamental scalar curvature level set formula
Day 2: Introduction to mass in General Relativity, the dominant energy condition and constraint equations, sketch of proof of the Riemannian positive mass theorem by harmonic functions
Day 3: Spacetime harmonic functions, the spacetime positive mass theorem, application to band inequalities, application to Llarull's teorem
Day 4: Sketch of the C^0 Dong-Song mass stability result. Sketch of application to the suboptimal Penrose inequality.
Day 1 Exercises:
[In the proof of the rigidity of the systole inequality using harmonic maps to S^1] Show that the Gauss curvatures of the level sets of \tilde{u} are constant.
Prove that the second fundamental form of a regular level set is the tangential part of the Hessian divided by |\nabla u|.
Prove X(|\nabla u|)=Hess u (X,\nu)
Prove the decomposition formula for the Laplacian over a hypersurface S with unit normal \eta: \Delta u=\Delta^S u + H^S \eta(u) + \nabla_{\eta \eta} u.
Read Daniel Stern's article, particularly the part that deals with critical points of a harmonic map.
Show that, for a harmonic function u on a domain \Omega (and no critical points on \partial \Omega), that
\int_{\partial \Omega} [ n(|\nabla u|)-|\nabla u|\kappa_t ]dA
is equal to
-\int_{\partial \Omega} [ |\nabla ^T u|/|\nabla u| ] <\nabla ^T u ,\nabla (n(u)/|\nabla^T u|)>dA
where n is the outward unit normal and \nabla ^T u denotes the tangential part of \nabla u on \partial Omega.