Local well-posedness for the Einstein-Vlasov system (Q2821658)

The Vlasov equation gives a statistical description of a collection of some particles. It is characterized by the fact that there is no direct interaction between particles, i.e. no collisions are included in the model. The Vlasov equation can be coupled to the Einstein equations giving rise to the Einstein-Vlasov system. Solutions to this system model space-times with collisionless self-gravitating particles described by a distribution function \(f\). Some introduction to the Einstein-Vlasov models could be seen in [\textit{A. D. Rendall}, Banach Cent. Publ. 41, 35--68 (1997; Zbl 0892.35148)].NEWLINENEWLINEThe author of the present paper analyses the local well-posedness of the Einstein-Vlasov system in the mean curvature-spatially harmonic gauge (\(CMCSH\) gauge) introduced by \textit{L. Andersson} and \textit{V. Moncrief} [Ann. Henri Poincaré 4, No. 1, 1--34 (2003; Zbl 1028.83005)]. The Einstein-Vlasov system is represented in the form \(\partial_tg_{ab}=-2Nk_{ab}+{\mathcal{L}}_{X}g_{ab}\), \( \partial_tk_{ab}=-\nabla_a\nabla_bN+N(R_{ab}-\delta_{ab}+ \text{tr}kk_{ab}-2k_{ai}k_b^i-\hat{T}_{ab})+{\mathcal{L}}_Xk_{ab}\), \( \triangle N=-\partial_t\tau + N(|k|_g^2+\eta )\), \( \triangle X^i+R^{i}_{m}X^m-{\mathcal{L}}_XV^i= 2\nabla_jNk^{ji}-\nabla^iN\text{tr}_gk+2Nj^i- \) \(-\bigl(2Nk^{mn}-({\mathcal{L}}_Xg)^{mn}\bigr)(\Gamma^{i}_{mn} -\hat{\Gamma}^{i}_{mn})\), \({\mathcal{X}}f=0\), where \(M\) is an orientable, connected, smooth manifold of dimension \(n \geq 2\), that admits a smooth Riemannian-Einstein metric with negative Einstein constant, \(\Gamma \), \(\hat{\Gamma }\) denote the corresponding Christoffel symbols, \(R_{ij}\) is the Ricci tensor of \(g\) (metric tensor), \(\delta_{ij}=(1/2)(\nabla_iV_j+\nabla_jV_i)\), \(N\) is the lapse function, \(X\) is the shift vector field, tangent to each Cauchy surfaces \(M_t = M\times\{t\}\), \({\mathcal{L}}_X\) denotes the Lie-derivative w.r.t. a vector field \(X\), \(k_{ab}\) is the second fundamental form, \(T\) is a time-like field normal to \(M_t\), \(V^{k}=g^{ab}e^k(\nabla_ae_b-\hat{\nabla }_ae_b)\), \(\hat{\nabla }\) is the Levi-Civita connection. Actually, this is an elliptic-hyperbolic system of PDEs, and in some cases \(\delta_{ij}=0\). The first above stated pair of equations in case \(\delta_{ij}=0\) are Einstein evolutions equations. Note that the additional \(\delta \)-term in the second equation is used when is taking into account that the constraints and gauge conditions are propagated. The Vlasov equation here is \({\mathcal{X}}f=0\). The CMCSH gauge is defined by the equalities: NEWLINE\[NEWLINE \tau\equiv \text{tr}_gk=-nt^{-1}, \;\;V^k=0. NEWLINE\]NEWLINENEWLINENEWLINEThe main statement here concerns the local existence of the solution on a compact interval \(I\subset\mathbb{R}\) for the above stated Einstein-Vlasov system in CMCSH gauge. This solution belongs to the class \(C_{b}^{2}(J\times M)\), and \(k\in C_{b}^{2}(J\times M)\), \(J=(T_0-T,T_0+T)\) (\(T>0\), \(T_0\in I\)).