The Øresund Link is a 4-lane motorway with a 2-line rail link between Copenhagen, Denmark, and Malmö, Sweden, built across the Øresund Sound between the 2 countries. The environmental concerns associated with construction of the Link were that that it might affect the water and salt flow through the Sound and into and out of the Baltic Sea, lead to destruction of the cod fishery, and affect other biological systems. An International Expert Panel was formed to design mitigation strategies to avoid negative environmental impacts. The Panel designed stringent environmental controls for the construction, of which the most significant was to limit sediment discharges to 5% of the dredged material. This was one-fifth of the discharge limit construction companies usually accept as normal for such a large-scale project. An environmental monitoring program was instituted to monitor compliance with the environmental control measures. As a result of the imposition of tough environmental requirements, the Link was built with little environmental impact and 3 y after it was completed, environmental conditions have returned to normal.
This paper is among 6 peer-reviewed papers published as part of a special series, Ecology in a Cost-Benefit Society. Portions of this paper were presented by the author at an international conference on this topic held at Roskilde University, Denmark, in 2004.
On 23 March 1991 the Governments of Denmark and Sweden signed an agreement to build a fixed transportation link (referred to hereafter as the Link) across the Øresund Sound connecting the metropolitan areas of Copenhagen and Malmö by a 4-lane motorway and a 2-lane railway link. This agreement was ratified by both Governments on 24 August 1991. In making this agreement, the governments required that the Link should lead to unchanged water flow through the Øresund and oxygen and salt supply to the Baltic Sea, the so-called “zero” solution.
The Danish public hearing on the design of the Link, as well as the construction of the Link, became a politically difficult issue because of loud protests from environmental organizations who highlighted the possible negative effects on both the Øresund (the Sound) and, more importantly, the Baltic Sea. The Baltic Sea is particularly vulnerable in that it is atidal with limited water exchange with the North Sea and has shown increased signs of eutrophication, with large areas of the deep basins lacking oxygen (Elmgren 2001). In addition, the important cod fisheries in the Baltic Sea are dependent on saline water entering from the North Sea (Flinkman and Backer 2002). Thus, fears were that any reductions in water or salt flow might upset the delicate balances of the Baltic Sea ecosystems.
The Danish and Swedish Environment Ministers decided that the best way to deal with the problem was to appoint an International Expert Panel, who were given the mandate to minimize the environmental impacts of the construction and to design and oversee the environmental monitoring program. The group comprised 8 oceanographers, ecologists, and fisheries biologists from Norway, Germany, Poland, Estonia, and the UK. This author served as chairman of the panel. The Danish Department of the Environment's marine section had the formal secretariat function and was supported by the Swedish county environmental authorities in Malmö. In most cases, the Panel reported through its chairman to the 2 Ministers directly.
From the constitution of the Panel, it is clear that its mandate was only concerned with the marine environment and did not consider the broader environmental issues, such as effects of the construction on land-based activities (e.g., sand and gravel extraction and gas emissions), nor was the panel concerned with the broader economic socioeconomic effects on the local or national regions.
ENVIRONMENTAL MANAGEMENT OF THE PROJECT
The overall control of the environmental management of the project was the responsibility of the Danish and Swedish environmental authorities, but day-to-day management was the responsibility of a Project Manager from the Øresund-konsortium. Before the Panel began their activities, the authorities had agreed on the following environmental criteria for the Link:
A zero solution (i.e., no changes in water or salt flow to and from the Baltic Sea) was stipulated.
A 25% reduction in eelgrass and mussel beds was permitted in the near area, defined as 500 m either side of the alignment of the Link.
Effects from construction on the flora and fauna were allowed for up to 5 y in the Sound.
The role of the Panel, therefore, was to make recommendations to be implemented by the 2 national environmental authorities.
Much of the Panel's time was spent on the 1st criterion, the zero solution, but that will only be considered briefly here. Figure 1 shows the Link, which starts adjacent to the Copenhagen airport in Denmark and runs to Malmö in Sweden. Near the airport, construction of a bridge would have interfered with aircraft traffic, so a tunnel was built, which emerges at an artificial island and continues as a bridge that is high enough to allow ships moving to and from the Baltic Sea to pass underneath. Although there is a large island in the middle of the Sound, (Saltholm), it is a Danish nature reserve, so the Link was not allowed to cross Saltholm; as a consequence, an artificial island was built south of Saltholm. Originally this island (subsequently called Peberholm, a name suggested by a school student) was designed to be closer to Copenhagen.
At its 1st meeting, the Expert Panel observed that, as initially designed, Peberholm would project further out into the Sound than the island of Saltholm and, thus, would significantly restrict the current flow through the channel near the airport. Because the mandate given paid particular attention to the need to ensure unchanged water flow to and from the Baltic Sea, the Expert Panel recommended that the island be moved into the lee of Saltholm, a distance of 1,100 m. This was immediately approved by the 2 environmental ministers but added an extra cost of $180 million to the project.
Another important aspect was that construction of the bridge pillars would lead to reduction in water and salt flow in the Sound. However, the zero solution could be achieved if the blocking effect of the pillars could be compensated for by dredging. The Sound has a shallow sill that normally restricts the water and salt flow. Therefore, it was planned to dredge the sill to compensate for the blocking effect of the bridge pillars. Yet, it was critical that the amount of compensation dredging should be correct so that a true zero solution was achieved. The blocking effect was calculated to be so small that it could not be measured against the natural day-to-day and seasonal variations in water and salt flow.
As a result, two 3-dimensional baroclinic hydrographic models were constructed—1 by the Swedish Meteorological Institute and one by the Danish Hydraulic Institute—that gave independent assessments of the blocking effect under all natural conditions. These were used successfully to calculate the compensation dredging, and the zero solution was achieved. (For a description of one of these models, see DHI .)
PROTECTING THE BIOLOGICAL SYSTEMS
Transboundary Environmental Impact Assessments (EIAs) are required by European law for any activity that could lead to environmental impacts in >1 country under the Espoo Convention (see United Nations Economic Commission for Europe  for an overview of the Convention). Yet, because Sweden was not at that time a member of the European Union, this requirement was not adopted for the project (see Gullett 2000 for an analysis of the process involved in the decision making associated with the Øresund Fixed Link). As part of the EIA for the Link, a baseline survey was conducted before the Expert Panel began its work. The Danish and Swedish environmental authorities had decided the acceptable and unacceptable effects of construction of the Link. The key criterion specified that a <25% reduction in eelgrass and mussel beds was acceptable 500 m either side of the Link and that no negative effects over the whole of the Sound would be found 5 y after construction was completed. The Expert Panel decided that the EIA must make quantitative predictions of what effects are likely from the construction activities and that these needed to be statistically rigorous and testable. The Expert Panel demanded that a statistical power analysis be done (Cohen 1988) to ensure that a 25% reduction in the key environmental variables associated with eelgrass and mussels could be detected. The results of this analysis are shown in Figure 2.
Figure 2 shows that the criterion of a 25% reduction in the key variables could be detected at best by 66% power for mussel settlement, but with a <30% probability of detecting a 25% reduction for the other variables. A detailed statistical analysis showed that even by increasing the number of replicates 5-fold in subsequent monitoring, the detection limit did in no case approach the 80% criterion that is the usual statistical limit for not committing a type II statistical error. The result of the Expert Panel's analysis was that some variables, such as abundance of soft sediment-living benthos, could not be used as a criterion. The baseline survey had to be repeated and the number of replicates increased 4-fold to get the required power to detect a 25% change in the variables.
The lesson learned from this exercise was that, more often than not, statistical power analyses are not done and, as a consequence, the number of samples taken in environmental monitoring programs is often too low to detect the changes required by environmental authorities. In discussing the power analysis performed for the Link environmental studies, Sanderson and Petersen (2001) claim that possible effects on the deepwater fauna were set 5 times as high as the acceptable risk for a type I error (i.e., β = 0.25). Unfortunately, Sanderson and Petersen (2001) misunderstood the analysis. The Expert Panel assessed the ability to detect a 25% change in environmental variables with α, the risk of committing a type I error, set at 0.05 and β, the risk of committing a type II error, set at 0.10, which are standard practices for acceptable levels of risk.
Next, the Expert Panel had to decide on how best to protect the marine environment from construction of the artificial island because this would involve much dredging and sediment spillage. The baseline survey had established the extent and quality of the eelgrass and mussel beds. Hydrographic models were developed describing where any sediment spill would move and the likely sedimentation areas. Interviewing dredging companies, the Expert Panel learned that under normal operations, a 25% spillage was accepted industry practice. The artificial island was to be constructed by simply piping sediments to the area and then directing the material to different areas.
To reduce the sediment spillage rate, the Expert Panel insisted on construction of a rock retaining wall before sediment was piped ashore. As a consequence of this restriction, the sediment spillage rate was reduced to 10%. The Expert Panel also made detailed enquiries about the dredging processes and machinery to be used and, after careful analysis, set the spillage rate limit at 5%. Although construction companies protested these tight restrictions, the Expert Panel's recommendation to Swedish and Danish Ministers was accepted, setting compliance at a 5% sediment spillage limit.
The next step for the Expert Panel was to ensure that the sediment spillage limit could be monitored and achieved. Together with the Link project manager, the best possible monitoring program was devised with the use of acoustic and other monitoring techniques with direct links to dredgers if the spillage limits were exceeded. Construction of the artificial island at Peberholm was stopped for a 10-d period 1 winter. This was not because the spillage limit was exceeded, but because the monitoring vessel could not operate and so monitoring could not be done. At an overall cost of $0.5 million a day, this work stoppage was not trivial. Other controls included vessels being put on maintenance, rather than dredging, in the event that a dredging vessel exceeded the allowable sediment spillage rate in a given short time interval so that the overall long-term spillage limit would not be exceeded.
The results of these and other stringent construction restrictions contributed to environmental protection. Out of the total dredged quantity of 12 million tonnes of marine sediment, the overall spillage rate was 4.2%, well below the limit set by the Expert Panel.
Because low sediment spillage rates could not be guaranteed during dredging, the Expert Panel implemented a feedback monitoring system with a range of biological variables to monitor the progress of dredging. In North America, feedback monitoring (Gray and Jensen 1993) is referred to as adaptive management (see http://en.wikipedia. org/wiki/Adaptive-management for a good description). The idea behind such a monitoring process is that it is possible to measure statistically quantifiable changes in variables characterizing an environmental system. Should a change be measured in 1 or more variables, actions triggered by a feedback loop are immediately initiated so that the undesirable effect is minimized.
For example, experiments with artificial shading showed that it was possible to detect changes in eelgrass quality measured as changes in shoot density, leaf biomass, and root biomass in disturbed environments compared with undisturbed environments. A feedback system was put into place in the event that the 5% limit in sediment spillage rate was exceeded. Eelgrass monitoring showed that at no time did any of the measured variables fall below the limit, which would lead to a halt in the dredging operations.
Other feedback loops monitored for effects on mussel beds, nesting birds, and fish migration. However, because of the success of the control on spillages from dredging, none of these biological feedback loops resulted in changed construction operations.
Finally, 3 y after the Link opened to rail and automobile traffic, an environmental survey of the biological systems in the Sound did not show any negative effects and indicated that the ecosystem had returned to preconstruction conditions (a list of the environmental reports can be found at Øresundsbro Konsortiet ). In fact, environmental conditions might even have improved because the extensive new areas of hard substrata provided by the bridge pillars provide large areas for recruitment of mussels and other organisms, which now occur in larger populations than before construction of the Link and are food for the large bird populations in the Saltholm bird reserve.
WAS IT WORTH IT?
To the author's knowledge, a detailed cost-benefit analysis of the construction of the Link has not been done. Table 1 shows the construction costs of the bridge according to the Øresundsbro Konsortiet (2004). Of these costs, the environmental management costs are estimated at US$323 million (Table 2).
Table Table 1.. Total construction costs for the Øresund Link in 1990 prices. Data from Øresundsbro Konsortiet (2004)
Cost (Billions US$)
Dredging and reclamation of the artificial island and peninsula
Bridge and toll station
Technical installations, coast to coast
Direct environmental costs
Revenue from EU subsidy, etc.
Total construction costs
Table Table 2.. Direct environmental costs of building the Øresund Link
Cost (Millions US$)
Lengthening the tunnel by 1 km
Extra costs on dredging budget
Total environmental cost
Although the environmental management costs might seem high, at 12% of the total, it should be borne in mind that this was a hugely important environmental issue in Scandinavia. In Sweden, a Minister of the Environment resigned in protest over the decision to build the Link. Throughout Scandinavia, the public is acutely aware of environmental issues and takes great interest in protection of the environment. Thus, it was not surprising when the project was announced that the news media was highly critical of the project on environmental grounds and that environmental organizations mounted public campaigns against the project. Yet, few protests occurred during construction of the Link. Since its completion, no criticisms at all have been made on the effects of the Link on the environment.
The International Expert Panel was praised for its role in successfully managing the environmental effects of one of the largest construction projects of its type in the world. However, the management of environmental effects could not have been done without full support from the management of the Øresundsbro Konsortiet, and especially the Project Manager who followed the Expert Panel's recommendations. Equally important, the Expert Panel was appointed by 2 Ministers from 2 different countries and reported directly to both Ministers. This imparted a unique political leverage that ensured that environmental protection remained a high priority and that the Expert Panel's recommendations were carried out.
I thank the Øresundsbro Konsortiet, the Swedish and Danish environmental authorities, and the members of the Panel for their support during the project. In particular, I thank Claus Dynesen, Øresundsbro Konsortiet Project Manager for the environment, who put in practice the Panel's decisions and so ensured that the project was completed with minimal environmental damage.