Stylianos Flampouris

Determination of Bathymetric and Current Maps by the Method DiSC Based on the Analysis of Nautical X-Band Radar Image Sequences of the Sea Surface (November 2007)

Morphodynamic processes in coastal areas are affected by tidal currents and sea state. The continuous observation of near coastal areas is important in order to monitor dangerous current-regime and bathymetry changes. Therefore, there is an urgent need for remote sensing techniques delivering the important hydrographic parameters with a high spatial resolution. Dispersive surface classificator (DiSC) is a newly developed method based on the analysis of nautical X-band radar image sequences of sea surface waves to determine spatial maps of hydrographic parameters, e.g., spatial maps of the bathymetry and the ocean current field. The method DiSC is described and is illustrated by the presentation of results based on a dataset acquired with a ground-based X-band radar installation mounted on the Island of Sylt in the German Bight. The calculated bathymetric maps are verified by multibeam echo sounder observations.

Accuracy of Bathymetric Assessment by Locally Analyzing Radar Ocean Wave Imagery (February 2008)

In this paper, the error source in assessing the bathymetry by a recently presented method, the dispersive surface classificator, is discussed. This method is based on the analysis of X-band radar image sequences of sea-surface waves to determine spatial maps of hydrographic parameters. To implement this objective, the radar-deduced bathymetry is validating by multibeam echosounder data. The accuracy of the method is high in homogeneous areas and reduced at the areas of bathymetric gradient and lower but comparable with multibeam echosounder data, under the assumption of the spatial resolution. The identification of systematic correlation of the absolute value of the error with the slope is significant and insignificant with the actual depth itself. The spatial correlation of the error illustrates that the direction of the wave field influences the neighboring grid cell in the same direction. The application of the method during crucial weather conditions is the main advantage and permits the accurate operational nearshore monitoring for several applications.

A Novel Approach Based on Marine Radar Data Analysis for High-Resolution Bathymetry Map Generation

This letter deals with the analysis of a novel data processing approach to estimate the local depth in a shallow coastal area starting from data collected by an in situ X-band radar system. The reconstruction approach is based on the maximization of the normalized scalar product (NSP) between the measured and the theoretical wave dispersion relation, which embeds the dependence on the searched for parameter (local depth). The use of NSP approach allows obtaining a high-resolution spatial map of the investigated area, and a thorough statistical analysis is carried out by comparing the local depth estimation results with the ground-truth data made available by a multi-beam echo-sounder survey. In addition, the improved accuracy of the NSP approach is demonstrated with respect to other methods previously used for the analysis of the same area.

The Influence of the Inverted Sea Wave Theories on the Derivation of Coastal Bathymetry

The estimation of nearshore bathymetry by inverting the wave dispersion function is an accepted method. The performance of four different wave theories inverted with dispersive surface classificator for the derivation of the bathymetry is examined during gale oceanographic conditions. The physical and technical limitations for the inversion are taken into consideration. The analyzed data are radar image sequences, and ground truth is an in situ echosounders bathymetry data set. All geomorphological features detectable with a 41-m spatial resolution have been identified, but there is an underestimation of the absolute depth. The applicable theories have mean error less than 7%. The nonlinear theories are compared with the linear theory, and the experimental difference of their performance is on the order of the theoretical expectation O (6%-8%).

Optimizing spectral wave estimates with adjoint-based sensitivity maps

A discrete numerical adjoint has recently been developed for the stochastic wave model SWAN. In the present study, this adjoint code is used to construct spectral sensitivity maps for two nearshore domains. The maps display the correlations of spectral energy levels throughout the domain with the observed energy levels at a selected location or region of interest (LOI/ROI), providing a full spectrum of values at all locations in the domain. We investigate the effectiveness of sensitivity maps based on significant wave height (Hs) in determining alternate offshore instrument deployment sites when a chosen nearshore location or region is inaccessible. Wave and bathymetry datasets are employed from one shallower, small-scale domain (Duck, NC) and one deeper, larger-scale domain (San Diego, CA). The effects of seasonal changes in wave climate, errors in bathymetry, and multiple assimilation points on sensitivity map shapes and model performance are investigated. Model accuracy is evaluated by comparing spectral statistics as well as with an RMS skill score, which estimates a mean model–data error across all spectral bins. Results indicate that data assimilation from identified high-sensitivity alternate locations consistently improves model performance at nearshore LOIs, while assimilation from low-sensitivity locations results in lesser or no improvement. Use of sub-sampled or alongshore-averaged bathymetry has a domain-specific effect on model performance when assimilating from a high-sensitivity alternate location. When multiple alternate assimilation locations are used from areas of lower sensitivity, model performance may be worse than with a single, high-sensitivity assimilation point.

Survey of bathymetry and current fields by radar image series acquired by shore based x-band radar

The error source in assessing the bathymetry by the Dispersive Surface Classificator method is discussed in this paper. The accuracy of the method is high in the deeper areas and is reduced behind the slopes. The identification of systematic correlation of the absolute value of the error with the slope was not possible. The spatial correlation of the error illustrates that the direction of the wave field influences the two neighboring cells in the same direction.

Corrections to “A Novel Approach Based on Marine Radar Data Analysis for High-Resolution Bathymetry Map Generation”

THIS document is an errata corrige of the above paper [1]. In [1], we described a novel approach to deal with the local depth estimation in shallow coastal areas through marine radar data. The reconstruction technique is based on the maximization of the normalized scalar product (NSP) between the theoretical and the measured wave dispersion relation, which depends on the desired local depth. Accordingly, the use of the NSP approach allows us to obtain a more detailed estimation of the bathymetric maps, wherein each pixel of the reconstructed bathymetric map is achieved as the average of a larger area where the assumption of the spatial homogeneity is verified. This letter [1] has a twofold aim. First of all, in [1], the validation of the NSP approach has been carried out by means of in situ data acquired at the coastal area of Sylt Island in Germany. Moreover, the estimation accuracy of the bathymetry provided by the NSP method has been compared with the one achieved by a benchmark, i.e., the linear version of the DIspersive Surface Classification (DiSC) [2]. To validate the bathymetry estimates provided by the NSP method, the 12 hourly-acquired X-band image sequences have been analyzed to generate 12 high-resolution bathymetric maps; the singly estimated bathymetry maps have been averaged in order to reduce the statistical uncertainty of the outcome. In situ measurements collected by a multibeam echo-sounder and represented in the same spatial grid of the NSP estimates (pixel spacing of 3.4 m) have been used as ground truth. The performance of the NSP approach has been evaluated through a statistical analysis in terms of scatter plot and histogram between the NSP outcomes and the echo sounder data. The histogram in [1] shows that the radar estimates suffer from a bias of about 1 m. In order to compare the DiSC and the NSP results in [1] we provided a statistics analysis in terms of, mean difference (Mean), relative mean difference (Rel. Mean), root mean square (RMS) and square of the correlation coefficient (R2).

Analysis of nautical X-band radar images for the generation of bathymetric map by the NSP method

This work presents the experimental validation of a novel data processing approach to estimate the local depth in a shallow coastal area starting from X-band radar data. The approach is based on the maximization of the Normalized Scalar Product (NSP) between the measured and the theoretical wave dispersion relations, which embed the dependence on the searched for parameters (current vector and depth). The use of NSP approach allows obtaining a high resolution spatial map of the investigated area and a thorough statistical analysis of this approach is carried out by comparison with the ground truth data collected by a multibeam echo-sounder. Finally, the accuracy of the NSP approach is compared with the one achieved by other methods and prove to provide better results.

Observing littoral waves by Doppler radar

The offshore observation of the wavefield with coherent radar systems consist a common practice for the study of the electromagnetic waves probing the sea surface waves and for the extraction of information of the sea waves. In this investigation, a Dopplerized, horizontally polarized, nautical radar is utilized for the monitoring of the wavefield evolution in the littoral zone. The radar datasets are globally unique and cover different geophysical conditions; therefore the impact of the bathymetry on the Doppler spectra is discussed, the horizontal velocity towards the shore is calculated and the properties of the radar deduced quantities are compared with in situ measurements.

Generation of bathymetric maps with high resolution through the analysis of nautical X-band radar images

This work deals with the analysis of a novel data processing approach to estimate the local depth in a shallow coastal area starting from X-band radar data. The approach is based on the maximization of the Normalized Scalar Product (NSP) between the measured and the theoretical wave dispersion relations, which embed the dependence on the searched for parameters (current vector and depth). In the present work, we presented an improvement version of NSP approach able to estimate the local depth, which allows to obtain a high resolution spatial map (subpixel) of the investigated area. The performances of the NSP approach is evaluated by comparing the estimated bathymetry with the ground truth provided the multi-beam echo sounder reported on the same grid resolution of the bathymetric maps of the X-band radar images.