Peter Damm

An investigation into the variability of circulation and transport on the north-west european shelf using three hydrodynamic models

Recent modelling studies of the circulation of the north-west European shelf have been undertaken at three institutes; Proudman Oceanographic Laboratory (POL), Institut fur Meereskunde (IfM), and Institute of Marine Research (IMR) in which long-term meteorological forcing data have been applied to drive each institute’s own hydrodynamic model. These models have been run for periods of up to 39 years (1955–1993) — made possible by the use of a long set of good quality meteorological data from the Norwegian Meteorological Institute (DNMI). These studies have provided a valuable insight into the long-term shelf circulation and the nature of its variability. The quantification of volume transports through key shelf sections has been completed and, where possible, comparisons have been made with other modelling and observational studies. Calculated transports for North Sea sections confirm the generally accepted circulation pattern although difficulties arise in the Irish Sea due to model limitations (low resolution and unsuitable advection schemes). The three models show similar patterns of variability in the water volume transports calculated for the sections. Mean values compare well in the southern North Sea where the water is well mixed, however large differences in mean transport values occur for northern North Sea sections due to the differences in model physics. IfM and IMR models are baroclinic and as such provide a more realistic representation of the transport through these deep water sections where baroclinic processes are important. Spectral analysis of the 39 year model runs shows a dominant annual cycle and less significant longer period signals of 7–10 years.

Time series analysis of monthly mean data of temperature, salinity, nutrients, suspended matter, phyto- and zooplankton at eight locations on the northwest european shelf

In this study an overview is given of the time series analysis of monthly mean data of physical, chemical and biological parameters. The time series are available at eight locations on the Northwest European Shelf. The integrated evaluation of those time series gives the opportunity to look for connections between the different parts of the shelf. Temperature and salinity seem to be externally forced. For the nutrients and biological parameters the local forcing is dominating the time series. It is concluded that there are areas that are comparable to each other (freshwater dominated boxes along the Belgian and Dutch coasts and German Bight; Atlantic dominated boxes in the English Channel and off the Scottish coast), although significant cross-correlations are hardly found. The Irish Sea can be regarded as a separate ecosystem.

Decadal variability on the northwest european shelf

The seasonal cycles of sea surface temperature and other parameters at eight sites on the Northwest European Shelf show an interannual variability in the range of years to decades for the years 1960–1995. Statistical methods like spectral analysis or phase diagrams prove that only a few frequency peaks beyond the annual cycle are of real significance. Most prominent signals for temperature and salinity on the shelf are the periods around 8 and 17 years. In general the time series are too short to identify definitely long-term trends or transitions between different states of equilibrium. However, there is an indication of a shift in the thermohaline regime along the Belgian-Dutch coast from the eighties to the nineties. Explanations for the decadal variability can be given by the intrinsic time scales of the internal shelf dynamics as well as by external periods. A spectral analysis of calculated mass fluxes at sections across the Northern and the Channel entrances of the North Sea show significant peaks beyond the seasonal cycle.This suggests an advective influence from the Atlantic Ocean. Another external source of decadal variability on the Northwest European Shelf is the atmosphere. Here no phase shift can be expected between corresponding signals of time series at different North Sea sites as it is indeed observed. The article is substantially a description and analysis of decadal variability based on observational and computational data as well as on statistical methods complied in the «Northwest European Shelf Programme” (NOWESP). The respective details are explained in other papers of this volume.

The NOWESP research data base

In the NOWESP project historical data from the Northwest European Shelf were compiled and evaluated to estimate the variability and trends in water movements, concentrations of dissolved and particulate constituents, and fluxes of the relevant substances across the shelf. As an integral part of the project, the NOWESP Research Data Base was created as a research tool to provide the data and data products needed for the analyses within the project. The tasks of the NOWESP Research Data Base group were the acquisition of the relevant data sets, with the intensive support of all partners, organization of the data sets in the NOWESP Research Data Base, merging of the specific data sets for the ten main state variables used in NOWESP, and the provision of data products for analysis within NOWESP. The data compiled during NOWESP represent a unique data set for the Northwest European Shelf. The data set is sufficiently comprehensive to allow the definition of long time series at about 14 sites in eight areas. It further enables the derivation of mean annual cycles of horizontal distributions of nine main state variables. NOWESP thus has provided valuable data sets for estimating budgets and fluxes across the shelf and, in addition, important data sets for the forcing and validation of ecological shelf sea models. An overview of the NOWESP data set is given. The organization of the data base is described in some detail, and examples of the products obtained for NOWESP are displayed.

Variability in fluxes of nutrients (N, P, Si) into the North Sea from the Atlantic Ocean and skagerrak caused by variability in water flow

Fluxes of nutrients (dissolved phosphate, nitrate and silicate) from the Atlantic Ocean (Northern Atlantic and Channel) and Skagerrak to the North Sea were calculated for the upper 30 meters and for the total water depths. Monthly average water flows and their standard deviations were taken from the model calculations of the If M (HAMSOM model) and IMR (NORWECOM model) for the period 1955–1993 and 1976–1995, respectively. Climatological monthly mean nutrient data were collected for the same transects from the NOWESP Research Data Base, and extracted from additional literature. With the water flow from the models and the available nutrient data rather precise fluxes can be calculated. The aim is to compare these fluxes with those published in the literature which are often based on yearly averaged water flow and some nutrient concentrations. The northern Atlantic inflow (total depth) of the mean fluxes of nitrate, phosphate and silicate is 3,972 ±1,604, 713 ±310 and 3,847 ±1,527 ktonnes/year. About 20% of these fluxes transported in the top 30 m water layer (IMR model). For the upper 30 m the IfM model calculated higher fluxes for these nutrients: 1,654 ±621, 301 ±111 and 1,549 ±565 ktonnes/year, respectively. For the Skagerrak the IMR model estimated a net mean outflow to the Baltic, while the IfM model estimated a net mean inflow of nutrients to the North Sea. Both models agree rather well in the mean inflow of nutrients through the Channel, but with rather high standard deviations as compared to the other sections. For example, for the IMR total depth estimation: 162 ±236, 27 ±39 and 161 ±237 ktonnes/year for N, P, Si respectively. In the northern sections, literature data on nutrient fluxes are at the lower end of the range estimated in this study, while in the Channel the literature data are in the upper range. It will be discussed that the largest improvements in nutrient flux estimations are to be expected when the precision and accuracy of water flow data become better.

The northwest european shelf temperature and salinity variability

From long temperature and salinity time series of the Northwest European shelf and 39 years of numerical model transport calculations, it is possible to derive the long-term hydrographic variability of the North Sea system and its possible causes. This is shown by means of trend analyses, cross-correlations, and power spectra. Initially there is a weak linear long-term positive trend in the SST series, which is in agreement with the global warming trend. In all series investigated, several periodicities are found on corresponding band widths, and we argue that this is due to long-term oscillations in the North Atlantic circulation system and interaction with the atmosphere. Cross-correlations between Sea Surface Temperature (SST)/ Sea Surface Salinity (SSS) and some transports are significant, with time lags of about 5 to 6 years. These time lags are unexplained; we assume that differences in the subpolar and subtropical gyre circulation might be responsible. A two-year time lag between the SSS in the Rockall Channel and east of Scotland points to lateral exchanges by shear dispersion, not to advective transports. It seems that the North Sea SST longterm fluctuations are coupled to the Atlantic subpolar gyre SST variability. In contrast, the Bay of Biscay behaves very similarly to the Atlantic subtropical gyre system. The ratio between the inflow of saline Atlantic water into the North Sea and continental run-off seems to have been rather constant in the last 100 years.