Coupling Tangent-Linear and Adjoint Models

We consider the solution of a (generalized) eigenvalue problem arising in physical oceanography that involves the evaluation of both the tangent-linear and adjoint versions of the underlying numerical model. Two different approaches are discussed. First, tangent-linear and adjoint models are generated by the software tool TAF and used separately. Second, the two models are combined into a single derivative model based on optimally preaccumulated local gradients of all scalar assignments. The coupled tangent-linear / adjoint model promises to be a good solution for small or medium sized problems. However, the simplicity of the example code at hand prevents us from observing considerable run time differences between the two approaches.

[1]  Peter U. Clark,et al.  The role of the thermohaline circulation in abrupt climate change , 2002, Nature.

[2]  Alfred V. Aho,et al.  Compilers: Principles, Techniques, and Tools , 1986, Addison-Wesley series in computer science / World student series edition.

[3]  Patrick Heimbach,et al.  Automatic Generation of Efficient Adjoint Code for a Parallel Navier-Stokes Solver , 2002, International Conference on Computational Science.

[4]  H. Stommel,et al.  Thermohaline Convection with Two Stable Regimes of Flow , 1961 .

[5]  Carl Wunsch,et al.  What Is the Thermohaline Circulation? , 2002, Science.

[6]  Carl Wunsch,et al.  Volume, heat, and freshwater transports of the global ocean circulation 1993-2000, estimated from a general circulation model constrained by World Ocean Circulation Experiment (WOCE) data , 2003 .

[7]  E. Tziperman.  Inherently unstable climate behaviour due to weak thermohaline ocean circulation , 1997, Nature.

[8]  Chao Yang,et al.  ARPACK users' guide - solution of large-scale eigenvalue problems with implicitly restarted Arnoldi methods , 1998, Software, environments, tools.

[9]  P. Hovland,et al.  An XML-Based Platform for Semantic Transformation of Numerical Programs , 2002 .

[10]  B. Farrell,et al.  Perturbation Growth and Structure in Uncertain Flows , 2022 .

[11]  Thomas Kaminski,et al.  Recipes for adjoint code construction , 1998, TOMS.

[12]  Uwe Naumann,et al.  Optimal accumulation of Jacobian matrices by elimination methods on the dual computational graph , 2004, Math. Program..

[13]  Andreas Griewank,et al.  Circumventing Storage Limitations in Variational Data Assimilation Studies , 1997, SIAM J. Sci. Comput..

[14]  Patrick Heimbach,et al.  An efficient exact adjoint of the parallel MIT General Circulation Model, generated via automatic differentiation , 2005, Future Gener. Comput. Syst..

[15]  B. Farrell,et al.  An Adjoint Method for Obtaining the Most Rapidly Growing Perturbation to Oceanic Flows , 1992 .

[16]  Jack Dongarra,et al.  Computational Science — ICCS 2002 , 2002, Lecture Notes in Computer Science.

[17]  A. Griewank,et al.  On the calculation of Jacobian matrices by the Markowitz rule , 1991 .

[18]  P. Courtier,et al.  A strategy for operational implementation of 4D‐Var, using an incremental approach , 1994 .

[19]  Carl Wunsch,et al.  Global ocean circulation during 1992-1997, estimated from ocean observations and a general circulation model , 2002 .

[20]  R. Giering,et al.  Construction of the adjoint MIT ocean general circulation model and application to Atlantic heat transport sensitivity , 1999 .

[21]  P. Ioannou,et al.  Perturbation Growth and Structure in Uncertain Flows. Part II , 2002 .

[22]  U. Naumann,et al.  TBR Analysis in Reverse-Mode Automatic Differentiation , 2003 .

[23]  E. Tziperman,et al.  Linear versus Self-Sustained Interdecadal Thermohaline Variability in a Coupled Box Model , 1997 .

[24]  E. Tziperman,et al.  Transient Growth and Optimal Excitation of Thermohaline Variability , 2002 .

[25]  A. Gordon.  The Role of Thermohaline Circulation in Global Climate Change , 1990 .

[26]  Andreas Griewank,et al.  Evaluating derivatives - principles and techniques of algorithmic differentiation, Second Edition , 2000, Frontiers in applied mathematics.