Publications
Following publications have been announced by our department Hydrodynamics and Data Assimilation. For further information please contact the marked authors of the publications:
Nam, P.T., Staneva, J., Bonaduce, A., Stanev, E., & Grayek, S. (2024): Interannual sea level variability in the North and Baltic seas and net flux through the Danish straits. Ocean Dynamics, doi:10.1007/s10236-024-01626-7
Abstract:
The paper presents the reconstruction of sea levels in the North Sea and Baltic Sea using Kalman filter approach. Based on the statistical characteristics of one year of daily maps of sea level from the Geesthacht COAstal model SysTem (GCOAST) and daily data at tide gauges along the coastline of two basins, the method can reconstruct effectively and accurately the multidecadal sea level anomalies. The high accuracy reconstruction data were then used to investigate the interannual variability in both basins and to estimate the difference between outflows and inflows (net flux) through the Danish Straits. The highest mean, standard deviation, and extreme values of sea level anomalies appear in winter and are well reproduced in different regions, such as the German Bight, the Southern North Sea, the Bothnian Bay, the Gulfs of Finland and Riga. The sea level variability is highly correlated with the mean sea level pressure and the zonal wind, particularly in the German Bight and in the winter months. The contributions of river runoff and net precipitation on the net flux are significant in the spring. The local wind has a greater influence on the net flux than the remote drivers.
Yuan, B., Jacob, B., Chen, W,, & Staneva, J. (2024): Downscaling sea surface height and currents in coastal regions using convolutional neural network. Applied Ocean Research, Vol 151, 104153, doi:10.1016/j.apor.2024.104153
Abstract:
Downscaling is the process to obtain high-resolution data from low-resolution data. Recently statistical models using convolutional neural networks have gained popularity for fast downscaling of environmental fields, while their application to the coastal sea surface height and currents is lacking. This research aims to downscale sea surface height and depth-averaged current to a resolution of hundreds of meters in coastal regions with dynamic shorelines using convolutional neural networks. Hourly outputs over one year from a physical numerical model for a coastal region in the German Bight are used as the low-resolution input and high-resolution ground truth for the network. The results show that the network effectively reconstructs sea surface height and current in the region to a resolution of hundreds of meters with a scale factor of 16 or even 64, and accurately traces the moving sea surface and shorelines. The global mean absolute error and root mean square error for the sea surface height are found to be less than 0.03 m and 0.07 m, respectively, and for the current less than 0.03 m/s and 0.05 m/s, respectively. These values are around ten times smaller than those obtained from interpolation methods including nearest neighbor, bilinear and bicubic. The network also effectively replicates the distribution of high-resolution data. The errors in the reconstructed time average, 1st percentile and 99th percentile are significantly smaller than those from interpolation methods, especially for the current. These results highlight the ability of the network to downscale sea surface height and currents in regions with complex shorelines, and have implications for downscaling other coastal fields and shoreline tracking.
