Suspended sediment modelling in a shelf sea (North Sea)

This paper extends the modelling of suspended particulate matter (SPM) on t he local coastal scale (described in preceding papers) to SPM modelling on the scale of the North Sea, focusing on representing SPM patterns and their seasonal distribution. The modelling includes a sensitivity study, in whic h model results are assessed using surface SPM concentration patterns extra cted from NOAA reflectance imagery, as well as North Sea Project in situ da ta. Over the past decade or so, first-order estimates of the net suspended load and its associated sources and sinks have been available and are generally substantiated. However, developments in the simulation of large-scale SPM behaviour are still severely restricted by the available descriptions of av ailable sediment sources and sediment erosion and deposition processes. Thi s paper indicates how remotely sensed reflectance images can provide additi onal information on the spatial distribution of (sea surface) suspended sed iments. A primary objective of this paper is to examine sensitivities of SPM simula tions in 2D (vertically averaged) and 3D models. A boundary-fitted coordina te modelling approach with intra-tidal resolution and synoptic meteorology is applied, as well as more schematic approaches. A related objective is to examine how both limited in situ observational data and reflectance imager y can be used to assess and improve such simulations. An integrated modelling-monitoring approach, using inverse and 'Goodness-of -Fit' (GoF) approaches applied to remotely sensed reflectance imagery, is u sed to derive a structured sensitivity analysis providing a quantified asse ssment of the strengths and weaknesses of modelling and input data. It is s hown that, especially in the coastal zone where salinity stratification may occur, 3D modelling is required while much of the sensitivity analysis can be based on a 2D modelling approach. This quantification of the effects of uncertainties of inputs and erosion/deposition parameters improves underst anding of the sediment distribution and budgets on the North Sea scale. It is concluded that whilst process studies are likely to contribute to imp roving erosion/deposition algorithms, and model developments will provide e nhanced dynamical descriptions, accurate overall simulation will remain dep endent on some (inverse) processes to reduce the uncertainty in sediment so urces. (C) 2000 Elsevier Science B.V. All rights reserved.