Search

Theory and seaborne measurements are presented for the near infrared (NIR: 700-900 nm) water-leaving reflectance in turbid waters. According to theory, the shape of the NIR spectrum is determined largely by pure water absorption and is thus almost invariant. A "similarity" NIR reflectance spectrum is defined by normalization at 780 nm. This spectrum is calculated from seaborne reflectance measurements and is compared with that derived from laboratory water absorption measurements. Factors influencing the shape of the similarity spectrurn are analyzed theoretically and by radiative transfer simulations. These simulations show that the similarity spectrum is valid for waters ranging from moderately turbid (e.g., water-leaving reflectance at 780 nm of order 10(-4) or total suspended matter concentration of order 0.3 g m(-3)) to extremely turbid (e.g., reflectance at 780 nm of order 10(-1) or total suspended matter of order 200 g m(-3)). Measurement uncertainties are analyzed, and the air-sea interface correction is shown to be critical for low reflectances, Applications of the NIR similarity spectrum to atmospheric correction of ocean color data and to the quality control of seaborne, airborne, and spaceborne reflectance measurements in turbid waters are outlined.

A chlorophyll-retrieval algorithm for use with imagery from the Medium Resolution Imaging Spectrometer (MERIS) aboard ENVISAT relying on wavebands centred at 665, 705 and 775 nm was described in a previous paper (Gons et al. 2002, J. Plankton Res., 24, 947-951). The present study reexamined the performance for the current nominal setting to 708.75 nm of the previously envisaged 705 nm band. Validation of the algorithm with reviled coefficients gave the same standard error of estimate for the inland and coastal waters as in the original work. The algorithm has been transcribed for direct application with the MERIS level-2 standard product 'water-leaving reflectance'. By this correction, chlorophyll estimation will generally improve, especially for high concentrations.

A chlorophyll (Chl) a retrieval algorithm, originally developed for spectral subsurface irradiance reflectance determined from above-water shipboard measurements, was adapted for use with satellite imagery to be acquired by the MERIS (Medium Resolution Imaging Spectrometer) instrument. This MERIS algorithm was calibrated for Chl a concentrations in the range 3-185 mg m(-3) using spectral reflectance calculated from shipboard measurements on the IJssel Lagoon (The Netherlands). Next, the algorithm was validated for various inland and coastal waters covering this concentration range. Despite the lower spectral resolution of MERIS as compared to the shipboard spectroradiometer, the standard error of estimate is expected to be similar, i.e. similar to9 mg m(-3) of Chl a in mesotrophic and eutrophic lakes, rivers, estuaries and coastal waters.

Two-band algorithms that use the ratio of reflectances at 672 and 704 nm have already proved successful for chlorophyll a retrieval in a range of coastal and inland waters. An analysis of the effect of reflectance measurement errors on such algorithms is made. It provides important indications of the range of validity of these algorithms and motivates the development of an entirely new type of adaptive two-band algorithm for hyperspectral data, whereby the higher wavelength is chosen for each input spectrum individually. When one selects the wavelength at which reflectance is equal to the reflectance at the red chlorophyll a absorption peak, chlorophyll a retrieval becomes entirely insensitive to spectrally flat reflectance errors, which are typical of imperfect atmospheric correction, and is totally uncoupled from the retrieval or an estimation of backscatter. This new algorithm has been tested for Dutch inland and Belgian coastal waters. (C) 2001 Optical Society of America.

A hand-held spectroradiometer was used for above-water determination of subsurface spectral irradiance reflectance in the Scheldt Estuary (Belgium/The Netherlands), the North Sea off the Belgian coast, and the Hudson/Raritan Estuary (New York/New Jersey). On the North Sea the measurement conditions were adverse, and elsewhere broken cloud caused considerable spectral variation. Despite this variation the retrieval of chlorophyll a (Chl-a) from three reflectance spectra at all sampling stations was stable. The algorithm calibrated for the freshwater IJssel Lagoon (The Netherlands) proved to be applicable to these estuarine and coastal waters(N=30; standard error of estimate = 7 mg m(-3) for corrected Chi-a ranging from 1 to 93 mg m(-3)).

The standard SeaWiFS atmospheric correction algorithm, designed for open ocean water, has been extended for use over turbid coastal and inland waters. Failure of the standard algorithm over turbid waters can be attributed to invalid assumptions of zero water-leaving radiance for the near-infrared bands at 765 and 865 nm. In the present study these assumptions are replaced by the assumptions of spatial homogeneity of the 765:865-nm ratios for aerosol reflectance and for water-leaving reflectance. These two ratios are imposed as calibration parameters after inspection of the Rayleigh-corrected reflectance scatterplot. The performance of the new algorithm is demonstrated for imagery of Belgian coastal waters and yields physically realistic water-leaving radiance spectra. A preliminary comparison with in situ radiance spectra fbr the Dutch Lake Markermeer shows significant improvement over the standard atmospheric correction algorithm. An analysis is made of the sensitivity of results to the choice of calibration parameters, and perspectives for application of the method to other sensors are briefly discussed. (C) 2000 Optical Society of America.

Several coastal ocean models have been used to compute the circulation on the Northwest European Continental Shelf. Five of them, developed within the European Union, are compared in the scope of an idealised three-dimensional test case, dealing with the geostrophic adjustment of a freshwater cylinder. As the central eddy adjusts, unstable baroclinic vortices start to grow. All the models are able to produce such unstable vortices. However, two of them produce an order-two instability, which is in accordance with a previous laboratory experiment, while the others exhibit an order-four instability. Using a simple scaling analysis, it is seen that the azimuthal wavenumber depends on the ratio of the kinetic energy to the available potential energy. It appears that the discrepancy in the azimuthal wavenumber is mainly due to the effect of the discretisation of the horizontal advection of momentum which could produce significant decrease of the total kinetic energy.

Three turbulence closure schemes, designed for stratified shallow water flows, are presented. They are based upon k-epsilon theory and use respectively two, one or zero transport equations for turbulent variables. The models are first tested on the evolution of a wind-driven turbulent layer in a stratified fluid. The results are at least qualitatively in agreement with observational and experimental data. A discussion is given about the existence of self-similar solutions. The models are compared next with the observational data of the Rhine outflow area. The periodic variation in the density structure, forced by wind and tides and which is clearly visible in the data, is predicted by the model. A physical interpretation of the model results is given in the absence of wind forcing. The effects of estuarine circulation, tidal straining and mixing on the development or breakdown of stratification are well represented by the model calculations.

Results are presented of a three-dimensional model study of the tidally-averaged salinity field in the Rhine-Meuse plume. In conditions of low mixing (no wind, neap tide) freshwater emerges from the river as a jet, turns right under the influence of Coriolis force and attaches to the coast as a buoyancy current. Surface residual currents are essentially geostrophic. Within the plume the surface layer is strongly stratified and overlies a bottom mixed layer. Results are strongly sensitive to the parameterization of vertical mixing: models ranging from constant diffusion coefficients through simple algebraic Richardson number dependent formulations to turbulence closure with evolution equations for turbulent kinetic energy and length scale were tested. Turbulence closure with a single evolution equation for turbulent kinetic energy and an algebraic length scale formulation was found to provide a suitable balance between physical realism (assessed by theoretical considerations and the practical ability to represent a well-mixed bottom boundary layer and stratified surface layer) and computational efficiency. Simpler models, which Vary diffusion coefficients as function of Richardson number, may produce similar results, though require more careful calibration. Constant diffusion coefficients are clearly inadequate for the application considered. Even the preferred ''k'' model requires some calibration as a ''background'' or ''ambient'' mixing coefficient had to be introduced to avoid unrealistically strong stratification. The tidally-averaged salinity field was found to be qualitatively similar at neap and spring tide, though cross-shore penetration of the outflow jet was reduced, bottom-mixed layer thickness increased and overall stratification reduced at springs. In contrast, the salinity field was found to be strongly dependent on wind strength and direction, both through wind-induced surface mixing and advection by wind-driven surface currents.

The Rhine is the largest river discharge of the North Sea, and has considerable ecological importance as a source of nutrients and pollutants. While much of the Dutch coastal zone is intensively monitored, it is difficult to perform measurements within the first few km of discharge, a region of very high horizontal and vertical gradients, which influences strongly the fate of freshwater and riverborne material. A fine resolution 3D hydrodynamic model, MU-ROFI, has been developed to simulate the flow in this initial discharge zone. The model is described briefly and a first set of model results is presented and discussed. The process of front formation and associated downwelling on the upstream (with respect to the tide) boundary of the freshwater plume, switching from the South to the North boundary as the tide turns, is illustrated. Qualitative comparison with CTD and airborne remotely sensed measurement indicates that such a behaviour is realistic, at least in conditions of low wind and Neap tide.

The aim of this study was to determine the feasibility of using airborne remote sensing to locate the Rhine plume front. Interest in fronts arises from the desire to predict the fate of pollutants and biological nutrients discharged from rivers into the open sea. Observations were made during flights over the Dutch coastal waters using a vertically-mounted video camera and a side-looking airborne radar (SLAR) designed for oil slick detection. Comparison of radar images with visual observations of the sea colour discontinuity and foam line establish that fronts can indeed be detected by SLAR because of high radar backscatter along the convergence line, where the fresh water jet impinges on saltier water. This provides a sound basis for future investigations using Synthetic Aperture Radar as mounted on ERS-1. An estimation of errors is given, identifying priorities for improvement of the technique. The accuracy achieved is considered sufficient for the validation of hydrodynamic models.