WSEAS Transactions on Systems
Print ISSN: 1109-2777, E-ISSN: 2224-2678
Volume 14, 2015
Estimating the Impact of Observational Data on Forecasts Made by Coupled Nonlinear Dynamical System
Authors: ,
Abstract: In this paper, the adjoint-base technique is applied to estimate the observational impact on the forecast quality using the coupled nonlinear chaotic dynamical system obtained by combining two (fast and slow) versions of the Lorenz system. The exploration of forecast sensitivity with respect to observations and the estimation of the observational impact on the forecast quality require considerable computational resources. For simple-enough low-order models, the computational cost is minor and therefore models of this class can be used as simple test instruments to emulate more complex systems. The method is incorporated into the existing four-dimensional variational data assimilation system. The objective function represents the difference between total energy norms of the background and calculated forecast error. Since data assimilation procedure shadows the “true” state of a system given by the observational data, the shadowing property of coupled nonlinear chaotic dynamical system is considered for weak and strong coupling between fast and slow subsystems. The discussed approach can be used in designing the optimal observing network and estimating the value of various observations.
Search Articles
Keywords: Dynamical system, chaotic dynamics, data assimilation, adjoint model, observation impact, sensitivity analysis
Pages: 33-44
WSEAS Transactions on Systems, ISSN / E-ISSN: 1109-2777 / 2224-2678, Volume 14, 2015, Art. #4