AN ERROR SUBSPACE PERSPECTIVE ON DATA ASSIMILATION

Two families of methods are widely used in data assimilation: the four-dimensional variational (4D-Var) approach, and the ensemble Kalman filter (EnKF) approach. The two families have been developed largely through parallel research efforts. Each method has its advantages and disadvantages. It is of interest to develop hybrid data assimilation algorithms that can combine the relative strengths of the two approaches. This paper proposes a subspace approach to investigate the theoretical equivalence between the suboptimal 4D-Var method (where only a small number of optimization iterations are performed) and the practical EnKF method (where only a small number of ensemble members are used) in a linear setting. The analysis motivates a new hybrid algorithm: the optimization directions obtained from a short window 4D-Var run are used to construct the EnKF initial ensemble. The proposed hybrid method is computationally less expensive than a full 4D-Var, as only short assimilation windows are considered. The hybrid method has the potential to perform better than the regular EnKF due to its look-ahead property. Numerical results show that the proposed hybrid ensemble filter method performs better than the regular EnKF method for the test problem considered.

[1]  Matthew Fisher,et al.  On the equivalence between Kalman smoothing and weak‐constraint four‐dimensional variational data assimilation , 2005, Quarterly Journal of the Royal Meteorological Society.

[2]  Adrian Sandu,et al.  A hybrid approach to estimating error covariances in variational data assimilation , 2010 .

[3]  Serge Gratton,et al.  Approximate Gauss-Newton Methods for Nonlinear Least Squares Problems , 2007, SIAM J. Optim..

[4]  Alberto Carrassi,et al.  Lyapunov vectors and assimilation in the unstable subspace: theory and applications , 2013 .

[5]  Roger Daley,et al.  The application of Kalman smoother theory to the estimation of 4DVAR error statistics , 1996 .

[6]  Ming Cai,et al.  Bred Vectors of the Zebiak–Cane Model and Their Potential Application to ENSO Predictions , 2003 .

[7]  Jeffrey L. Anderson.  A Local Least Squares Framework for Ensemble Filtering , 2003 .

[8]  J. Whitaker,et al.  Ensemble Square Root Filters , 2003, Statistical Methods for Climate Scientists.

[9]  P. Courtier,et al.  Variational Assimilation of Meteorological Observations With the Adjoint Vorticity Equation. I: Theory , 2007 .

[10]  Jonathan D. Beezley,et al.  PREDICTOR-CORRECTOR AND MORPHING ENSEMBLE FILTERS FOR THE ASSIMILATION OF SPARSE DATA INTO HIGH-DIMENSIONAL NONLINEAR SYSTEMS , 2006 .

[11]  Xiangjun Tian,et al.  An ensemble-based explicit four-dimensional variational assimilation method , 2008 .

[12]  David D. Parrish,et al.  GSI 3DVar-Based Ensemble-Variational Hybrid Data Assimilation for NCEP Global Forecast System: Single-Resolution Experiments , 2013 .

[13]  Marc Bocquet,et al.  Diagnosis and impacts of non-Gaussianity of innovations in data assimilation , 2010 .

[14]  R. Daley.  Atmospheric Data Analysis , 1991 .

[15]  Eric Blayo,et al.  Reduced‐order 4D‐Var: A preconditioner for the Incremental 4D‐Var data assimilation method , 2006, 0709.2814.

[16]  G. Evensen,et al.  Analysis Scheme in the Ensemble Kalman Filter , 1998 .

[17]  J. Yorke,et al.  Chaos: An Introduction to Dynamical Systems , 1997 .

[18]  Clive D Rodgers,et al.  Inverse Methods for Atmospheric Sounding: Theory and Practice , 2000 .

[19]  Ionel M. Navon,et al.  Optimality of variational data assimilation and its relationship with the Kalman filter and smoother , 2001 .

[20]  G. Evensen.  Data Assimilation: The Ensemble Kalman Filter , 2006 .

[21]  Istvan Szunyogh,et al.  A Local Ensemble Kalman Filter for Atmospheric Data Assimilation , 2002 .

[22]  J. Yorke,et al.  Four-dimensional ensemble Kalman filtering , 2004 .

[23]  Adrian Sandu,et al.  A Hybrid Variational/Ensemble Filter Approach to Data Assimilation , 2009 .

[24]  I. Yu. Gejadze,et al.  On Analysis Error Covariances in Variational Data Assimilation , 2008, SIAM J. Sci. Comput..

[25]  Fuqing Zhang,et al.  Coupling ensemble Kalman filter with four-dimensional variational data assimilation , 2009 .

[26]  Daryl T. Kleist,et al.  Forecast System: Single Resolution Experiments , 2013 .

[27]  Philippe Courtier,et al.  Four‐Dimensional Assimilation In the Presence of Baroclinic Instability , 1992 .

[28]  A C C Gibbs,et al.  Data Analysis , 2009, Encyclopedia of Database Systems.

[29]  A. Bennett,et al.  Inverse Modeling of the Ocean and Atmosphere , 2002 .

[30]  M. Buehner,et al.  Intercomparison of Variational Data Assimilation and the Ensemble Kalman Filter for Global Deterministic NWP. Part II: One-Month Experiments with Real Observations , 2010 .

[31]  Li Wei,et al.  Fitting model fields to observations by using singular value decomposition: An ensemble‐based 4DVar approach , 2007 .

[32]  Zhiquan Liu,et al.  Hybrid Variational/Ensemble Data Assimilation , 2011 .

[33]  Andrew C. Lorenc,et al.  The potential of the ensemble Kalman filter for NWP—a comparison with 4D‐Var , 2003 .

[34]  Zhijin Li,et al.  Optimality of 4D-Var and its relationship with the Kalman lter and Kalman smoother , 1999 .

[35]  Geir Evensen,et al.  The Ensemble Kalman Filter: theoretical formulation and practical implementation , 2003 .

[36]  E. Kalnay,et al.  C ○ 2007 The Authors , 2006 .

[37]  J. Sherman,et al.  Adjustment of an Inverse Matrix Corresponding to a Change in One Element of a Given Matrix , 1950 .

[38]  Dacian N. Daescu,et al.  The Impact of Background Error on Incomplete Observations for 4D-Var Data Assimilation with the FSU GSM , 2005, International Conference on Computational Science.

[39]  B. Hunt,et al.  Four-dimensional local ensemble transform Kalman filter: numerical experiments with a global circulation model , 2007 .

[40]  A. Robinson,et al.  Data Assimilation via Error Subspace Statistical Estimation.Part I: Theory and Schemes , 1999, Monthly Weather Review.

[41]  Marc Bocquet,et al.  Joint state and parameter estimation with an iterative ensemble Kalman smoother , 2013 .

[42]  A. Lorenc,et al.  Operational implementation of a hybrid ensemble/4D‐Var global data assimilation system at the Met Office , 2013 .

[43]  M. Fisher,et al.  Assimilation Techniques (5): Approximate Kalman Filters and Singular Vectors April 2001 , 2002 .

[44]  Fuqing Zhang,et al.  Dynamics and Structure of Mesoscale Error Covariance of a Winter Cyclone Estimated through Short-Range Ensemble Forecasts , 2005 .

[45]  Pierre FJ Lermusiaux.  Data Assimilation via Error Subspace Statistical Estimation. , 1999 .

[46]  E. Kalnay,et al.  Ensemble Forecasting at NCEP and the Breeding Method , 1997 .

[47]  M. Buehner,et al.  Intercomparison of Variational Data Assimilation and the Ensemble Kalman Filter for Global Deterministic NWP. Part I: Description and Single-Observation Experiments , 2010 .

[48]  Adrian Sandu,et al.  Singular Vector Analysis for Atmospheric Chemical Transport Models , 2006 .

[49]  C. Qiu,et al.  A hybrid space approach for ensemble‐based 4‐D variational data assimilation , 2009 .

[50]  T. Başar,et al.  A New Approach to Linear Filtering and Prediction Problems , 2001 .

[51]  J. M. Lewis,et al.  Dynamic Data Assimilation: A Least Squares Approach , 2006 .

[52]  R. E. Kalman,et al.  A New Approach to Linear Filtering and Prediction Problems , 2002 .

[53]  Q. Xiao,et al.  An Ensemble-Based Four-Dimensional Variational Data Assimilation Scheme. Part I: Technical Formulation and Preliminary Test , 2008 .