Ringkøbing Fjord is a shallow estuary connected to the North Sea through a canal and a sluice. A simple model is presented with which the water level and the stratification in the estuary may be calculated from observations of wind, input of freshwater, water level in the North Sea, and data for the operation of the sluice. Two main components are the use of open-channel, hydraulic theories to determine the exchange through the canal and sluice and a potential energy concept to calculate the vertical mixing. One major simplifying assumption is to disregard the baroclinic dynamics, which works well except in case of weak or strong winds. The model is run for 1999, and the results compare reasonably well with observations of water levels and salinity profiles at several locations. The primary objectives of the model are to be able to estimate, as a function of the operational practice of the sluice, a) the salinity of the main body of water and b) the amounts and the lives of the intermittently inflowing volumes of seawater, which could lead to oxygen depletion in the deep parts of the estuary due to eutrophication. This is demonstrated in a period in September 1999 during which a volume of seawater is first admitted to the estuary and then gradually eroded, while the oxygen in it is rapidly consumed. The present model provides an efficient and computationally inexpensive tool to study some estuarine properties, other alternatives being three-dimensional models of high spatial and temporal resolution and complex turbulence description. The hydraulic part of the model, describing the flow through the canal and sluice, can be modified, and so the model can be adapted to other estuaries that are characterized by a large degree of horizontal homogeneity and a well-defined connection to the sea.

The monthly advective supply of nutrients and nutrient pool sizes in the Kattegat and the Belt Sea are estimated for the period 1989-1996 from a combination of time-dependent hydrodynamic model results and concentration measurements of dissolved inorganic nitrogen (DIN), phosphate (DIP) and total nitrogen and phosphorous (TN and TP). The net supply of nutrients to these seas is computed as the sum of advection and nutrient loads from atmosphere and land to the Kattegat and the Belt Sea. The average net supplies are 5000 t N per month and 500 t P per month in both areas, corresponding to approximately 10% of the amounts required for the annual primary production in the area. The net supply of N displays a small seasonal variation, while the net supply of P exhibits both a spring maximum and a winter minimum. The seasonal nutrient turnover cycle is estimated from the residual of the net supply and the seasonal variation in nutrient pools. Net release of DIP from the sediment takes place during late summer and winter, while net release of DIN occurs mainly during winter. Thus, the net release from the sediment adds to the pool of nutrients available for the spring bloom in the Baltic entrance the following year. Both advection and the autumn-winter release of P exceed the land load of P. Thus, on a short time-scale, reductions in P load cannot be expected to significantly change the availability of P for biogeochemical processes. However, the monthly N load from land and atmosphere exceeds the advective supply. Reductions in N load thus have a significant influence on availability of N for biogeochemical processes. (C) 2002 Elsevier Science B.V. All rights reserved.

Monthly time series of nutrient pools and transport in the Baltic Entrance were computed for the period 1974–1999 from a combination of hydrodynamic model results and observational data obtained in national monitoring programmes. Significant seasonal variations in both nutrient pools and transport were found. For example, inflowing dense nutrient-rich Skagerrak water causes net transport of dissolved inorganic phosphorus (DIP) and inorganic nitrogen (DIN) towards the Baltic during summer where surface concentrations are low. Surface-water concentrations of DIN and DIP increase during winter and the net transport may reverse. The computations indicated that the pools of dissolved inorganic phosphorus and total phosphorus (TP) were larger in the 1980s than in previous and following decades. Variations in pools of dissolved inorganic nitrogen (DIN) and total nitrogen (TN) were more irregular throughout the period. Computed nutrient transports likewise exhibited long-term variations. Net transport of DIN and DIP was southward, from the Skagerrak to the Kattegat and further into the Belt Sea, during the 1970s. In the same period, the transport of DIN and DIP from the Baltic Sea to the Baltic entrance was below average. In both the Kattegat and the Belt Sea transports reversed during the 1980s. In the 1990s the southward transport of DIP was restored, but the transport of DIN was not restored to the same extent. Between the Kattegat and the Skagerrak the transport diminished while the Belt Sea continued to export DIN to the Kattegat. Northward transports of TN and TP persisted throughout the period. A notable maximum TP transport was found to have taken place during the 1980s.

This paper presents data on physical conditions and carbon transport in a typical northeast Greenland fjord along with predictions of expected changes in the area due to climate change. The fjord has an average depth of 100 m; the maximum depth is 360 m, and a sill at a depth of 45 m is found at its entrance. Sea ice covers the fjord from early October to late July. The freshwater input to the fjord, occurring from June to September, is 1063x10(6) m(3) from the catchment area (3109 km(2)) and 440x10(6) m(3) from melting of sea ice. During the ice-free period this buoyancy input and mixing by wind and tides results in an estuarine circulation in which lighter low-salinity water is moved seaward above denser water from the Greenland Sea. The tidal amplitude is 0.8 to 1.5 m, and the transport of tides from the outer parts of the fjord to the inner parts is delayed less than 15 min due to low friction in the fjord system. During the ice-free period, a net carbon input of 15-50 t C d(-1) occurs in the outer region of the fjord due to transport from land and the adjacent Greenland Sea. A regional atmosphere-ocean model predicts a temperature increase of 6-8degreesC at the end of this century (2071-2100) that will lead to increase in freshwater runoff, thinning of the sea ice, and an increase in ice-free conditions from 2.5 mo to 4.7-5.3 mo in Young Sound. The increased freshwater input will greatly enhance the estuarine circulation and nutrient input to the fjord and is expected to increase biological productivity.

The oxygen level in the deep layers (20–40 m) of the Baltic entrance is modelled as a function of an oxygen consumption rate constant and climate forcing. The oxygen level is calculated as median values of the observed oxygen concentrations. Climate forcing is expressed by water temperature and consumption period, representing the time during which the water was isolated below the halocline and experienced net respiration. First- and zero-order reaction terms are considered, and it is found that the first-order reaction yields an optimal representation of the natural range of oxygen concentration observations in the area. However, neither model performance nor conclusions are strongly dependent on the choice of reaction order. Using the first-order model, the modelled oxygen level declines exponentially with increasing consumption period, where the averaged rate constant for the oxygen consumption at 5 °C is estimated at 0.0045 day−1. The consumption rate increases exponentially by a factor of 3 with a 10 °C temperature increase. Climate forcing accounts for 64% of the total variation in oxygen level and 95% of its seasonal variation. The gross consumption rate is enhanced by 25% after the spring and autumn phytoplankton blooms, while it is reduced during winter, when surface-water primary production is light limited. This seasonal variation supports the argument that gross consumption can be limited by the availability of organic matter. The averaged seasonal variation in gross consumption rate is included in the model. In this way, we have accounted for both atmospheric forcing and seasonal variations in the bottom-water oxygen level at the Baltic entrance. The averaged error of the model estimate thus becomes 9%. This residual variation arises primarily from modelling uncertainties and variations in the gross consumption forced by interannual changes in the availability of organic matter for respiration. Our results support the hypothesis that the decline in oxygen conditions observed between 1960 and 2000 cannot be explained without considering changes in the export production in surface waters of the entrance to the Baltic Sea. The observed reduction in the oxygen level cannot be explained by the variability in atmospheric forcing on water exchange and bottom-water temperature as resolved by the present transport model.

This chapter describes the operational hydrodynamical model of the Danish waters—a Danish national program for monitoring the water environment. DHI Water & Environment is operating a regional 3D hydrodynamical model of the North Sea, the Danish Waters, and the Baltic Sea for the Danish Ministry of Environment and Energy. The objective is to determine the transports of water, salinity, and nutrients through sections dividing different seas. An interactive two-way nesting technique is used in the 3D model, MIKE 3, for 5 grids of horizontal resolution that varies from 9–1/3 nm. The model is forced by astronomical tide and by meteorological data provided by the Danish Meteorological Institute. The calibration, validation, and operational use of the model are based on both on-line and off-line measurements. The dissemination of the results is a key issue. A database containing model results and measurements has been established and made accessible through the Internet.

Water exchange and residence time are calculated for 31 small Danish estuaries to assess the spatial variability of estuarine processes and biogeochemical properties. To identify the uncertainty of the residence time estimates, three different model types have been applied to the estuaries. The dynamic models applied comprise hydrodynamic (HD) models, and a well-mixed batch reactor model for the winter-nitrate concentration. Residence times of the dynamic models range from 0·3 to 127 d. The median value of the deviation between the results of these two model types is 30%. Furthermore, a morphological model is formulated. It includes entrance width as the independent variable, and the approximation that the saltwater flow per unit entrance width is equal for investigated Danish estuaries. This model yields a fair representation of water exchange and residence time over three and two orders of magnitude, respectively. The deviation from the dynamic model results is 40%. Hence, in comparison to entrance width, differences in mixing and forcing appear to be of limited importance to the water exchange variability between Danish estuaries. The morphological model may thus be used to give a sound estimate of the water exchange for Danish estuaries, where more detailed modelling is lacking. However, in either model comparison, the deviation between model results is less than the residence time variability between the estuaries. The models may thus supplement one another for making quantitatively acceptable analysis of processes and bio-geochemical properties in Danish estuaries.

The response of macrobenthic faunal abundance and biomass to nutrient load, and factors that may modify this response are examined by means of inter-estuary comparisons of 14 shallow Danish estuaries. Data for this analysis are the physicochemical and biological variables monitored by local authorities mainly during the period 1989–95. A clear positive effect of nutrient load is demonstrated on benthic biomass, over a wide range of total N-load as the model substance from 2–200gm−2year−1. The relationship was curvilinear with a levelling off or even depression of biomass at high total N-load (above c. 35gm−2year−1). A mixed chemostat model using total load and hydraulic residence time for estuaries was applied to estimate the load that could be realized into primary production, and consequently enter into benthic production. Two measures, the load corrected for winter export (the realized N-load) and the nutrient pool available for the spring bloom (SBNP) were calculated. The benthic metabolic demand inferred from biomass, assuming an annual P:B ratio of 1 (P, secondary production; B, benthic biomass), was approximately of the same magnitude as both total N-load and realized N-load. A positive correlation was still found between benthic biomass/production and the realized N-load, but the linearity of the regression between them was not improved compared to the corresponding relationship with total load. The best linearly proportional relationship was obtained with the spring bloom N-pool (SBNP) calculated from the chemostat model. Stoichiometry suggested, however, that the spring bloom is of little importance for supporting benthic standing stock in these well flushed estuaries. To explain the strong statistical relationship, despite poor causality, with SBNP, it is suggested that the algorithm behind SBNP reflects the ability of the estuary to retain nutrients in the water mass in the productive period, both as free molecules and included into biological particles. These findings strongly indicate that benthic standing stock system-wide is food limited and indicate the importance of interaction between loading and estuary residence time (flushing) for the outcome of eutrophication. The findings are in agreement with reports that high estuary flushing rate may modify effects of eutrophication, and they deviate from previous studies in shallow coastal areas reporting either no effect, or negative effects, of eutrophication on benthic biomass.

We review various aspects of the structure and functioning of Danish estuaries from data collected by the National Monitoring Program and from information in published sources. We present data on the physical, chemical, and biological characteristics of estuaries in Denmark, we evaluate the functioning of these systems as filters and transformers of nutrients and we evaluate the outlook for Danish estuaries in the future. Danish estuarine systems are for the most part shallow (<3 m deep), have short residence times, and tend to be heavily loaded with nutrients primarily from agricultural sources. Total average loads from land per unit watershed area are 112 kg P km −2 yr −1 and 2,400 kg N km −2 yr −1 during the period 1989–1995. The total phosphorus (TP) load in estuaries has been significantly reduced over the last decade, following implementation of the 1987 Action Plan for the Aquatic Environment (Vandmiljøplan in Danish) that prescribed that nitrogen loads to the total aquatic environment should be reduced by 50% and phosphorus loads by 80%. Reductions in the total nitrogen (TN) load have been more modest. Nutrient loading is one of the primary determinants of estuarine nutrient concentration with 70% of the annual variation in TN concentration and 55% of the annual variation in TP concentration explained by variation in the load. Many Danish estuaries have rich communities of macrophytes and benthic filter feeders, such as Mytlis edulis and Ciona intestinalis , that can control water column chlorophyll concentrations by their filter feeding activities. Many of the estuaries experience hypoxia and anoxia, especially during warm and calm summer months. Further reductions in nutrient loading are expected following implementation of the Action Plan for the Aquatic Environment II, with predicted improvements in oxygen concentrations and in the functioning of these shallow, dynamic estuarine systems.

A one-dimensional model was used to evaluate the role of wind mixing and a horizontal buoyancy flux on the stratification of the upper 5–10m in Kattegat at the entrance to the Baltic Sea estuary. The horizontal buoyancy flux in the upper 5–10m is shown to cause a stabilization of an order of magnitude equal to wind mixing. This buoyancy flux is caused by pulses of large volumes of less salty Baltic Proper water periodically entering Kattegat. When spread over Kattegat, these pulses create stratification within the upper 5m that typically lasts for 4 days. After stratified periods, the upper 5m become homogeneous for a day or two before the stratification is re-established. Stratification of the upper 10m typically lasts for 10–15 days.

Changes in salt stratification were investigated by using temperature and fluorescence measurements as tracers. By doing this, information on the short term history of the water column was revealed. Further the dominant processes in mixing and stratification of the upper metres of the water column in the southern Kattegat were identified.Due to the salt content, density increased with depth over most of the water column during the period of measurement, A 50 km intrusion of warm, phytoplankton-rich water from the Great Belt outlet was identified.The density increase was maintained due to the buoyancy input given by the combined effect of a horizontal density gradient and vertical shear. Hence the vertical temperature inhomogeneities lasted for 3-7 days and were not mixed even under an event of strong wing mixing. In this period the nutrient and salt transport are uncoupled, since most of the mixing energy is used to redistribute salt within the nutrient-poor surface layer.A reduction of the horizontal density gradient reduced the buoyancy input and allowed for the homogenization of the upper 15 metres within a few hours. One sharp interface was created. After the formation of the sharp interface salt and nutrient transport can be calculated by using the same entrainment function for both salt and nutrient transport. Only during approximately 5% of the time there was a direct entrainment of bottom water and only 30% of the available mixing energy could be used to give this entrainment.