A four-dimensional scheme based on singular value decomposition (4DSVD) for chaotic-attractor-theory-oriented data assimilation Export

@article{citeulike:3979139, abstract = {The chaotic-attractor-theory-oriented data assimilation (CDA) method is reviewed. A scheme based on the singular value decomposition (SVD) analysis, called the 4DSVD, is then updated to a real four-dimensional scheme for the CDA. This algorithm employs the SVD to extract the base vectors that span the phase spaces of both model and observation chaotic attractors. Four groups of experiments are carried out under the perfect model assumption by using the Lorenz 28-variable model to test the performance of this scheme in assimilating data. Meanwhile, this scheme is compared with the four-dimensional variational data assimilation (4DVAR). The results show that the 4DSVD is effective and efficient in assimilating ” observations” to obtain analysis states that are consistent with model dynamics. The accuracy of the analysis state of the 4DSVD is similar to that of the 4DVAR; however, the required computational time of the 4DSVD is much less than that of the 4DVAR, and the 4DSVD also avoids having to estimate the background error covariance matrix. The results of observation system simulation experiments carried out using the Weather Research and Forecasting Modeling System show that the method could generate good analyses by assimilating incomplete observations. They dalso exhibit the ability of the 4DSVD to constrain the assimilation using the full model dynamics. }, author = {Wang, Jincheng and Li, Jianping}, citeulike-article-id = {3979139}, citeulike-linkout-0 = {http://www.agu.org/pubs/crossref/2009/2008JD010916.shtml}, citeulike-linkout-1 = {http://dx.doi.org/10.1029/2008JD010916}, day = {29}, doi = {10.1029/2008JD010916}, issn = {0148-0227}, journal = {Journal of Geophysical Research - Atmospheres}, keywords = {4dvar, assimilation, dynamics, nonlinear, svd, weather}, month = {January}, number = {D2}, pages = {D02114+}, posted-at = {2009-01-29 15:07:46}, priority = {2}, title = {A four-dimensional scheme based on singular value decomposition (4DSVD) for chaotic-attractor-theory-oriented data assimilation}, url = {http://dx.doi.org/10.1029/2008JD010916}, volume = {114}, year = {2009} }

TY - JOUR ID - citeulike:3979139 L3 - citeulike-article-id:3979139 N2 - The chaotic-attractor-theory-oriented data assimilation (CDA) method is reviewed. A scheme based on the singular value decomposition (SVD) analysis, called the 4DSVD, is then updated to a real four-dimensional scheme for the CDA. This algorithm employs the SVD to extract the base vectors that span the phase spaces of both model and observation chaotic attractors. Four groups of experiments are carried out under the perfect model assumption by using the Lorenz 28-variable model to test the performance of this scheme in assimilating data. Meanwhile, this scheme is compared with the four-dimensional variational data assimilation (4DVAR). The results show that the 4DSVD is effective and efficient in assimilating “observations” to obtain analysis states that are consistent with model dynamics. The accuracy of the analysis state of the 4DSVD is similar to that of the 4DVAR; however, the required computational time of the 4DSVD is much less than that of the 4DVAR, and the 4DSVD also avoids having to estimate the background error covariance matrix. The results of observation system simulation experiments carried out using the Weather Research and Forecasting Modeling System show that the method could generate good analyses by assimilating incomplete observations. They dalso exhibit the ability of the 4DSVD to constrain the assimilation using the full model dynamics. IS - D2 JF - Journal of Geophysical Research - Atmospheres SN - 0148-0227 TI - A four-dimensional scheme based on singular value decomposition (4DSVD) for chaotic-attractor-theory-oriented data assimilation VL - 114 SP - D02114 KW - 4dvar KW - assimilation KW - dynamics KW - nonlinear KW - svd KW - weather AU - Wang, Jincheng AU - Li, Jianping PY - 2009/01/29/ UR - http://dx.doi.org/10.1029/2008JD010916 ER -