A reduced adjoint approach to variational data assimilation

The adjoint method has been used very often for variational data assimilation. The computational cost to run the adjoint model often exceeds several original model runs and the method needs significant programming efforts to implement the adjoint model code. The work proposed here is variational data assimilation based on proper orthogonal decomposition (POD) which avoids the implementation of the adjoint of the tangent linear approximation of the original nonlinear model. An ensemble of the forward model simulations is used to determine the approximation of the covariance matrix and only the dominant eigenvectors of this matrix are used to define a model subspace. The adjoint of the tangent linear model is replaced by the reduced adjoint based on this reduced space. Thus the adjoint model is run in reduced space with negligible computational cost. Once the gradient is obtained in reduced space it is projected back in full space and the minimization process is carried in full space. In the paper the reduced adjoint approach to variational data assimilation is introduced. The characteristics and performance of the method are illustrated with a number of data assimilation experiments in a ground water subsurface contaminant model.

[1]  Jérôme Vialard,et al.  Three- and Four-Dimensional Variational Assimilation with a General Circulation Model of the Tropical Pacific Ocean. Part I: Formulation, Internal Diagnostics, and Consistency Checks , 2003 .

[2]  M. Verlaan,et al.  Efficient identification of uncertain parameters in a large‐scale tidal model of the European continental shelf by proper orthogonal decomposition , 2012 .

[3]  A. Heemink,et al.  IDENTIFICATION OF SHALLOW SEA MODELS , 1993 .

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

[5]  J. C. Ramírez,et al.  Estimation of aquifer parameters under transient and steady-state conditions , 1984 .

[6]  Carl Wunsch,et al.  State estimation improves prospects for ocean research , 2002 .

[7]  D. Ullman,et al.  Model parameter estimation from data assimilation modeling: Temporal and spatial variability of the bottom drag coefficient , 1998 .

[8]  William Carlisle Thacker,et al.  Fitting models to inadequate data by enforcing spatial and temporal smoothness , 1988 .

[9]  Pierre F. J. Lermusiaux,et al.  Estimation and study of mesoscale variability in the strait of Sicily , 1999 .

[10]  P. Delecluse,et al.  Three- and Four-Dimensional Variational Assimilation with a General Circulation Model of the Tropical Pacific Ocean. Part II: Physical Validation , 2003 .

[11]  Thomas Kaminski,et al.  An Example of an Automatic Differentiation-Based Modelling System , 2003, ICCSA.

[12]  Ionel M. Navon,et al.  A reduced‐order approach to four‐dimensional variational data assimilation using proper orthogonal decomposition , 2007 .

[13]  Ibrahim Hoteit,et al.  Treating strong adjoint sensitivities in tropical eddy‐permitting variational data assimilation , 2005 .

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

[15]  Ionel M. Navon,et al.  A Dual-Weighted Approach to Order Reduction in 4DVAR Data Assimilation , 2008, Monthly Weather Review.

[16]  F. L. Dimet,et al.  Variational algorithms for analysis and assimilation of meteorological observations: theoretical aspects , 1986 .

[17]  Athanasios C. Antoulas,et al.  Approximation of Large-Scale Dynamical Systems , 2005, Advances in Design and Control.

[18]  R. Courant,et al.  Methods of Mathematical Physics , 1962 .

[19]  Andreas Schiller,et al.  A technique for the determination of surface heat and freshwater fluxes from hydrographic observations, using an approximate adjoint ocean circulation model , 1995 .

[20]  B. Cornuelle,et al.  Impact of Resolution and Optimized ECCO Forcing on Simulations of the Tropical Pacific , 2008 .

[21]  Nancy Nichols,et al.  Using Model Reduction Methods within Incremental Four-Dimensional Variational Data Assimilation , 2008 .

[22]  J. Jansen,et al.  Model-reduced gradient-based history matching , 2011 .

[23]  Ibrahim Hoteit,et al.  Efficiency of reduced-order, time-dependent adjoint data assimilation approaches , 2006 .

[24]  Andreas Griewank,et al.  Achieving logarithmic growth of temporal and spatial complexity in reverse automatic differentiation , 1992 .

[25]  V. Hurai,et al.  BASIC program for interpretation of microthermometric data from H 2 O and H 2 O-NaCl fluid inclusions , 1989 .

[26]  A. H. Al-Rabeh,et al.  Optimal estimation of parameters for a two‐dimensional hydrodynamical model of the Arabian Gulf , 1993 .

[27]  P. Courtier,et al.  Variational assimilation of meteorological observations with the direct and adjoint shallow-water equations , 1990 .

[28]  Patrick Heimbach,et al.  An eddy-permitting, dynamically consistent adjoint-based assimilation system for the tropical Pacific: Hindcast experiments in 2000 , 2010 .

[29]  Arnold W. Heemink,et al.  Inverse 3D shallow water flow modelling of the continental shelf , 2002 .

[30]  Arnold W. Heemink,et al.  Model-Reduced Variational Data Assimilation , 2006 .

[31]  M. Piggott,et al.  International Journal for Numerical Methods in Fluids a Fully Non-linear Model for Three-dimensional Overturning Waves over an Arbitrary Bottom , 2009 .

[32]  I. Hoteit,et al.  Simultaneous perturbation stochastic approximation for tidal models , 2011 .

[33]  J. M. Lewis,et al.  The use of adjoint equations to solve a variational adjustment problem with advective constraints , 1985 .

[34]  S. P. Neuman,et al.  Estimation of Aquifer Parameters Under Transient and Steady State Conditions: 1. Maximum Likelihood Method Incorporating Prior Information , 1986 .