Prediction of nontidal sea level variations in the Persian Gulf using data assimilation techniques. Issue 3 (3rd July 2018)

• Record Type: Journal Article
• Authors: Afshar-Kaveh, Naghmeh; Ghaheri, Abbas; Chegini, Vahid; Nazarali, Mostafa
• Is Part Of: Coastal engineering journal. Volume 60:Issue 3(2018)
• Page Start: 340
• Page End: 355
• Publication Date: 2018-07-03
• Subjects: Data assimilation -- non-tidal sea level -- residual -- Persian Gulf -- FVCOM; Coastal engineering -- Periodicals; Coasts -- Periodicals; Hydraulic engineering -- Periodicals; Coast changes -- Periodicals; Shore protection -- Periodicals; Coast changes; Coastal engineering; Coasts; Hydraulic engineering; Shore protection; Periodicals; 627.58
• DOI: 10.1080/21664250.2018.1530852
• Languages: English
• ISSNs: 2166-4250

Abstract

ABSTRACT: The regular pattern of tidal water level fluctuation in the open seas is often deviated by the effect of weather systems. The stochastic behavior of these systems makes it difficult to predict the nontidal residuals (NTR) precisely. The hydrodynamic models as efficient tools for the simulation of NTR may suffer from lack of accuracy due to approximate boundary conditions and forcings. Data Assimilation (DA) techniques combine hydrodynamic model predictions with available field observations to improve the predictions. In this study, a number of DA methods were employed to improve the NTR predictions in a semienclosed basin. The results have shown the studied DA techniques, which are originally developed for weather prediction problems, can be improved by considering the physical mechanisms of NTR generation. A new formula was introduced for the calculation of weight factors of the existing methods. Application of the proposed method increased the accuracy of NTR predictions significantly so that the correlation coefficient of the NTR predictions increased from 0.660 to 0.896 and the root mean square error decreased from 5.22 cm to 2.51 cm. In addition, so it is concluded among the studied DA schemes, Barnes method has better efficiency for prediction of NTR in the region.