The geometry of optimally transported meshes on the sphere

Mesh redistribution methods provide meshes with increased resolution in particular regions of interest, while requiring that the connectivity of the mesh remains fixed. Some state of the art methods attempt to explicitly prescribe both the local cell size and their alignment to features of the solution. However, the resulting problem is overdetermined, necessitating a compromise between these conflicting requirements. An alternative approach, extolled by the current authors, is to prescribe only the local cell size and augment this with an optimal transport condition. It is well-known that the resulting problem is well-posed and has a unique solution, not just in Euclidean space but also on the sphere. Although the optimal transport approach abandons explicit control over mesh anisotropy, previous work has shown that, on the plane, the meshes produced are naturally aligned to various simple features. In this paper, we show that the same is true on the sphere. Furthermore, formal results on the regularity of solutions to optimal transport problems imply the sphere is actually more benign than the plane for mesh generation, due to its intrinsic curvature. We also see informal evidence of this. We provide a wide range of examples to showcase the behaviour of optimally transported meshes on the sphere, from axisymmetric cases that can be solved analytically to more general examples that are tackled numerically. Evaluation of the singular values and singular vectors of the mesh transformation provides a quantitative measure of the mesh anisotropy, and this is shown to match analytic predictions.

[1]  Cristian E. Gutiérrez,et al.  The Monge―Ampère Equation , 2001 .

[2]  David A. Ham,et al.  A Parallel Edge Orientation Algorithm for Quadrilateral Meshes , 2015, SIAM J. Sci. Comput..

[3]  Philip Browne,et al.  Fast three dimensional r-adaptive mesh redistribution , 2014, J. Comput. Phys..

[4]  Andrew T. T. McRae,et al.  Firedrake: automating the finite element method by composing abstractions , 2015, ACM Trans. Math. Softw..

[5]  Lisandro Dalcin,et al.  Parallel distributed computing using Python , 2011 .

[6]  Andrew T. T. McRae,et al.  Automating the solution of PDEs on the sphere and other manifolds in FEniCS 1.2 , 2013 .

[7]  Eric Jones,et al.  SciPy: Open Source Scientific Tools for Python , 2001 .

[8]  Robert D. Russell,et al.  The geometry of r-adaptive meshes generated using optimal transport methods , 2015, J. Comput. Phys..

[9]  John Thuburn,et al.  Horizontal grids for global weather and climate prediction models: a review , 2012 .

[10]  Xu Wang Some counterexamples to the regularity of Monge-Ampère equations , 1995 .

[11]  William Gropp,et al.  Efficient Management of Parallelism in Object-Oriented Numerical Software Libraries , 1997, SciTools.

[12]  C. Villani Topics in Optimal Transportation , 2003 .

[13]  C. Villani Optimal Transport: Old and New , 2008 .

[14]  Gian Luca Delzanno,et al.  An optimal robust equidistribution method for two-dimensional grid adaptation based on Monge-Kantorovich optimization , 2008, J. Comput. Phys..

[15]  G. Loeper,et al.  Gradient estimates for potentials of invertible gradient–mappings on the sphere , 2006 .

[16]  David A. Ham,et al.  An Algorithm for the Optimization of Finite Element Integration Loops , 2016, ACM Trans. Math. Softw..

[17]  Bharat K. Soni,et al.  Handbook of Grid Generation , 1998 .

[18]  Gian Luca Delzanno,et al.  Robust, multidimensional mesh-motion based on Monge-Kantorovich equidistribution , 2011, J. Comput. Phys..

[19]  Andrew T. T. McRae,et al.  PyOP2: Framework for performance-portable parallel computations on unstructured meshes , 2016 .

[20]  Andrew T. T. McRae,et al.  Automated Generation and Symbolic Manipulation of Tensor Product Finite Elements , 2014, SIAM J. Sci. Comput..

[21]  Andrew T. T. McRae,et al.  Optimal-Transport-Based Mesh Adaptivity on the Plane and Sphere Using Finite Elements , 2016, SIAM J. Sci. Comput..

[22]  L. Caffarelli Interior $W^{2,p}$ estimates for solutions of the Monge-Ampère equation , 1990 .

[23]  Andrew T. T. McRae,et al.  fiat: The Finite Element Automated Tabulator , 2016 .

[24]  Andrew T. T. McRae,et al.  A structure-exploiting numbering algorithm for finite elements on extruded meshes, and its performance evaluation in Firedrake , 2016 .

[25]  L. Caffarelli,et al.  On the Regularity of Reflector Antennas , 2008 .

[26]  R. McCann Polar factorization of maps on Riemannian manifolds , 2001 .

[27]  P. Swarztrauber,et al.  A standard test set for numerical approximations to the shallow water equations in spherical geometry , 1992 .

[28]  Matthew G. Knepley,et al.  petsc4py: The Python interface to PETSc , 2016 .

[29]  Anders Logg,et al.  Unified form language: A domain-specific language for weak formulations of partial differential equations , 2012, TOMS.

[30]  Tom Henderson,et al.  A general method for modeling on irregular grids , 2011, Int. J. High Perform. Comput. Appl..

[31]  Peter H. Lauritzen,et al.  A class of deformational flow test cases for linear transport problems on the sphere , 2010, J. Comput. Phys..

[32]  Lawrence Mitchell,et al.  TSFC: a structure-preserving form compiler , 2017, SIAM J. Sci. Comput..

[33]  J. F. Williams,et al.  MOVING MESH GENERATION USING THE PARABOLIC MONGE–AMPÈRE EQUATION∗ , 2008 .

[34]  G. Loeper On the regularity of solutions of optimal transportation problems , 2009 .

[35]  J. F. Williams,et al.  Parabolic Monge-Ampère methods for blow-up problems in several spatial dimensions , 2006 .

[36]  Robert D. Russell,et al.  Adaptive Moving Mesh Methods , 2010 .

[37]  G. Burton TOPICS IN OPTIMAL TRANSPORTATION (Graduate Studies in Mathematics 58) By CÉDRIC VILLANI: 370 pp., US$59.00, ISBN 0-8218-3312-X (American Mathematical Society, Providence, RI, 2003) , 2004 .

[38]  D. Williamson The Evolution of Dynamical Cores for Global Atmospheric Models(125th Anniversary Issue of the Meteorological Society of Japan) , 2007 .

[39]  N. Trudinger,et al.  On second derivative estimates for equations of Monge-Ampère type , 1984, Bulletin of the Australian Mathematical Society.

[40]  G. Loeper Regularity of Optimal Maps on the Sphere: the Quadratic Cost and the Reflector Antenna , 2013, 1301.6229.

[41]  Todd D. Ringler,et al.  Exploring a Multiresolution Modeling Approach within the Shallow-Water Equations , 2011 .

[42]  Y. Brenier Polar Factorization and Monotone Rearrangement of Vector-Valued Functions , 1991 .

[43]  Philip Browne,et al.  Mesh adaptation on the sphere using optimal transport and the numerical solution of a Monge-Ampère type equation , 2015, J. Comput. Phys..