Reducing the memory requirements of parameter estimation using model order reduction

Figure 2: Error [m] for reconstructed model output and observations with truncated projection U in different size Figure 3: Cost function (upper left); Final correction factor difference in bathymetry for estimation without and with SVD (upper right); RMSE difference between initial model and estimated model without (lower left) and with ( lower right) SVD. Color blue shows the improvement. Introduction Previous development of a parameter estimation scheme for a Global Tide and Surge Model (GTSM) showed that accurate estimation of the parameters is currently limited by the memory use of the analysis step. Singular values decomposition (SVD) is a useful technique to reduce the high dimension system with a smaller linear subspace. In this study, we focus on the application of SVD in time patterns to reduce the dimension of model output and observations. As expected, the time patterns show a strong resemblance to the tidal constituents, which indicates that the memory requirements can be reduced dramatically by projection of the model output and observations onto the time-SVD patterns.