The current state of understanding the North Sea’s physical system is presented. First, basic phenomena like astronomical tides and general circulation will be described and analysed with respect to their physical nature and respective interactions. There will be special focus on fundamental dynamic balances. Next, some specific topics relevant to the marine ecosystem, the economy and society will be considered: among them, spreading and transport processes, the fresh water budget, the heat budget and storm surges. A separate section is dedicated to the North Sea of Tomorrow, i.e. the prospective variations of the physical environment resulting from global changes in future decades. The statements are based on the long experience of the authors and their groups and include findings that are The complete text of the paper is available at http://www.iopan.gda.pl/oceanologia/ 664 J. Sündermann, T. Pohlmann little known if at all. The review finishes with a list of open questions and the corresponding research demands.

Mathematical Background and Methods of Ocean Modelling.- Simple Linear Models for Diagnostic Calculation of Ocean Climate Characteristics.- Nonlinear Models for Diagnostic, Prognostic and Adjustment Calculations of Ocean Climate Characteristics.- Synthesis of Models and Observed Data.- Modelling of Climate Variability in Selected Ocean Basins.- Modelling Climate Variability of Selected Shelf Seas.

In addition to natural risks such as rising sea levels, storm surges, or tsunamiwaves, the pollution of the marine environment represents a serious threat tocoastal inhabitants. However, contrary to the first factors mentioned, pollutiondoes not present a direct danger to human life. Societies have recently started tobecome aware of marine pollution, however, awareness is increasing rather slowlyand is not sufficiently developed everywhere. Nevertheless, pollution of the seadamages the marine ecosystem irreversibly over long time scales, endangering abroad spectrum of resources, from seafood to recreational spaces. The struggleagainst marine pollution requires environmental knowledge within society, politicalresolve, and money. Industrial nations, which are also the biggest polluters,are meeting these criteria to some extent, but among them there is no uniformposition on marine pollution. Developing countries are in danger of repeating theenvironmental mistakes of previous decades, but on a greater magnitude. Thecurrent state of the sea is characterized by considerable pollutant load in the shelfregions of the most industrialized nations (e.g., the Northwest European shelf), butthere is also a trend of improvement in these areas. Sea shelves of less developedcountries (e.g., the East Asian waters) exhibit an increasing burden. Industrialnations must share their experiences and provide conceptual and financial help toless developed countries.

Within the KUSTOS program (Coastal Mass and Energy Fluxes-the Land-Sea Transition in the Southeastern North Sea) 28 to 36 German Bight stations were seasonally surveyed (summer 1994, spring 1995, winter 1995–1996) for light conditions, dissolved inorganic nutrient concentrations, chlorophylla (chla), and photosynthesis versus light intensity (P:E) parameters. Combining P:E curve characteristics with irradiance, attenuation, and chlorophyll data resulted in seasonal estimates of the spatial distribution of total primary production. These data were used for an annual estimate of the total primary production in the Bight. In winter 1996 the water throughout the German Bight was well mixed. Dissolved inorganic nutrient concentrations were relatively high (nitrogen [DIN], soluble reactive phosphorus [SRP], and silicate [Si]: 23, 1, and 10 μM, respectively). Chla levels generally were low (< 2 μg l−1) with higher concentrations (4–16 μg l−1) in North Frisian coastal waters. Phytoplankton was limited by light. Total primary production averaged 0.2 g C m−2 d−1. Two surveys in April and May 1995 captured the buildup of a strong seasonal thermo-cline accompained by the development of a typical spring diatom bloom. High nutrient levels in the mixed layer during the first survey (DIN, SRP, and Si: 46, 0.45, and 11 μM, respectively) decreased towards the second survey (DIN, SRP, and Si: 30.5, 0.12, and 1.5 μM, respectively) and average nutrient ratios shifted further towards highly imbalanced values (DIN:SRP: 136 in survey 1, 580 in survey 2; DIN:Si: 13.5 in survey 1, 96 in survey 2). Chla ranged from 2 to 16 μg l−1 for the first survey and rose to 12–50 μg l−1 in the second survey. Phytoplankton in nearshore areas continued to be light limited during the second survey, while data from the stratified regions in the open German Bight indicates SRP and Si limitation. Total primary production ranged from 4.0 to 6.3 g C m−2 d−1. During summer 1994 a strong thermal stratification was present in the German Bight proper and shallow coastal areas showed unusually warm (up to 22°C), mixed waters. Chla concentrations ranged from 2 to 18 μg l−1. P:E characteristics were relatively high despite the low nutrient regime (DIN, SRP, and Si: 2, 0.2, and 1.5 μM, respectively), resulting in overall high total primary production values with an average of 7.7 g C m−2 d−1. Based on the seasonal primary production estimates of the described surveys a budget calculation yielded a total annual production of 430 g C m−2 yr−1 for the German Bight.

Marine Turbulence: Theories, Observations, and Models is the first book to give a comprehensive overview of measurement techniques and theories for marine turbulence and mixing processes. It describes the processes which control the mixing of greenhouse gases, nutrients, trace elements, and hazardous substances in our oceans and shelf seas – from local to planetary scales. These processes buffer climate changes and are centrally important for regional to global ecosystem dynamics. The book is divided into eight parts. Part I introduces the nature of turbulence in relation to stratification, waves, and intermittence. Part II describes observational techniques for field studies. Part III presents selected computational means for the study of turbulence. Part IV introduces details of boundary layers. Parts V and VI present practical case studies in estuaries, fjords, lakes, and shelf seas, and at the shelf edge. Part VII bridges the smallscale three-dimensional turbulence and quasi-two-dimensional turbulence occurring on the planetary scale. Part VIII concludes the book with an overview of comprehensive data sets and models codes. The publication also contains source codes of turbulence models and models of the upper-ocean mixing layer (COHERENS and GOTM), and observational data sets of turbulence charac teristics or corresponding proxies of waters from all over the world.

In a joint project (1997–2000), the University of Hamburg (Germany) and the Ocean University of Qingdao (China) have investigated the circulation, water mass structure, nutrient fluxes and phytoplankton dynamics in the Bohai Sea. A hierarchy of coupled circulation, transport and production models for the seasonal scale has been developed and tested against field data from two ship cruises in autumn and spring. The models are available now for scenario calculations under the impact of environmental change.

The circulation and the hydrography of the Bohai Sea are simulated with the Hamburg Shelf Ocean Model (HAMSOM). The model is three-dimensional, prognostic baroclinic and has a resolution of 5 min in latitude and longitude and 10 layers in the vertical. It is initialised and forced with the five main tidal constituents, temperature and salinity distributions taken from the Levitus database, monthly mean river run-off values and European Centre for Medium Range Weather Forecast (ECMWF) reanalysed data of air pressure, wind stress and of those parameters relevant for the calculation of heat fluxes. The simulation period covers 14 years from 1980 to 1993 due to the availability of the time-dependent ECMWF forcing.The results are analysed by means of time series and EOFs focussing on the interpretation of fluctuations with periods above the tidal cycle. Furthermore, tracer simulations are carried out and turnover times are calculated in order to evaluate the importance of these fluctuations on the renewal and transport of water masses in the Bohai Sea.One of the major outcomes of the investigation is the overall dominance of the annual cycle in all hydrographic parameters and the importance of stochastic weather fluctuations on the transport of water masses in the Bohai Sea. (C) 2003 Elsevier B.V. All rights reserved.

Papers in this and a companion issue report on an interdisciplinary study of the vertical fluxes of properties throughout the water column from the sea surface to the seabed. The project was centred on measuring and modelling for two contrasting sites in the North Sea of turbulence properties and their effects on particles, zooplankton and nutrient cycling, particularly the relative importance of cycling in the water column, the seabed fluff layer and the sediments. Turbulence activity was weaker at the northern site, which stratified in summer, and where measurements were obtained during the start of the autumnal breakdown of stratification. The southern site, where measurements were taken during and after the spring bloom, was much more dynamic both in terms of turbulence and of particles. The site was close to the Dutch coast and was well mixed throughout the year, except for the intermittent influence of the Rhine plume. The study contributes towards the long-term goal of developing robust water column plankton models applicable in the full range of turbulence environments encountered in continental shelf seas.

Dissipation rate measurements in the northern North Sea from two independent observations are compared with various numerical models. The turbulence was characterised by tidal forcing in the bottom boundary layer and atmospheric forcing in the surface boundary layer. The observations were carried out by using free-falling profilers equipped with shear probes and fast CTD sensors. The models are based on Reynolds averaging and range from simple one-equation models to two-equation models with algebraic second-moment closures. Several error measures are applied for comparison of observations and model results. It is shown that the differences between the two observations are significantly larger than the equivalent measures between the model results. This is caused by the stochastic character of turbulent microstructure in connection with under-sampling, but also by the distance between the two observational sites, the movements of the vessels, instrument errors and so forth. The models on the other hand, although closed on different levels, are all based on the same assumptions and driven by the same external forcing, thus showing only relatively small differences between each other.