Its Role in Realising the Bridge and Tunnel Connection Between Denmark and Sweden Mette Thorkilsen and Claus Dynesen

105 © IWA Publishing 2001 Journal of Hydroinformatics | 03.2 | 2001

An owner’s view of hydroinformatics: its role in realising the bridge and tunnel connection between Denmark and Sweden Mette Thorkilsen and Claus Dynesen

ABSTRACT

The motorway and railway connection between Denmark and Sweden, opened on 1 July 2000, when Mette Thorkilsen Project Manager, taken together with the connection across the Great Belt between the largest Danish islands, now Environment, provides a direct link between the Scandinavian peninsular and the rest of Europe. At a total cost of Øresundsbro Konsortiet some 8 billion US dollars, these projects represented the largest infrastructural investments of their Claus Dynesen Project Director, kind in Europe. Although backed by strong political and economic interests, these projects were also Environment & Authorities, Øresundsbro Konsortiet opposed by a part of the public and especially by political and environmental interest groups. This opposition was particularly pronounced in the case of the Denmark-Sweden link, partly owing to its location in a densely populated area and partly due to the potential impacts of the proposed link on the very sensitive local and regional marine environment. Thus, alongside the task of designing and constructing the physical link, the consortium that was responsible for its realisation, Øresundsbro Konsortiet, had to find ways to satisfy these many diverse interests. This paper describes how Øresundsbro Konsortiet, being an owner that valued constructive partnership, took up these challenges in their management, and how the environmental concerns were accommodated in the design and construction methods. Furthermore, it describes how the socio-technical approaches already taken up and developed within hydroinformatics in earlier projects were taken much further in the case of the Denmark-Sweden link. Finally, the paper describes the role of hydroinformatics in the various phases of the project and its significance in achieving the successful completion. The role of hydroinformatics as an important technology in facilitating stakeholder dialogue is thereby also clearly illustrated. Key words | dredging and reclamation, environmental management, environmental modelling, the Øresund Fixed Link, public-private partnership project, sociotechnology

INTRODUCTION The 16.4 km long motorway and railway connection across the Great Belt between the two principal islands across the Sound between Denmark and Sweden was of Denmark, provides a direct connection between the opened on 1 July 2000 by H. M. Queen Margrethe II of Scandinavian peninsular and the rest of Europe. The Denmark and H. M. King Carl XVI Gustav of Sweden project was completed six months ahead of schedule and jointly with Prime Ministers Poul Nyrup Rasmussen within its original budget of 3.7 billion US dollars. Its com- (Denmark) and Go¨ ran Persson (Sweden). Since this par- pletion largely justiﬁed the construction of the 4.2 billion ticular Sound is called Øresund in Danish, we shall refer US dollar Great Belt transport link that opened in 1997–98. to it as Øresund in this paper. These links, viewed together, were charged with the The Øresund Fixed Link (Figure 1), together with the greatest geopolitical signiﬁcance within the context of the technically somewhat similar but 1.1 kilometre longer link European Union, and were thus driven and supported by

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 106 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Øresundsbro Konsortiet chose to take on the project’s political sensitivity as a challenge. The surrounding world’s requirements, for instance, with regard to the design of the facility and the link’s environmental impact, were incorporated into the detailed planning as technical problems to be resolved. This gave the proposed solutions a high degree of visibility and allowed them to be com- municated to and discussed with the organisations that had presented them. Without the accommodation of these concerns in the design and construction methods, it is almost certain that this project could not have been realised. This choice was further reﬂected throughout the con- Figure 1 | Aerial view of the completed Øresund Fixed Link (Summer 2000). struction period by adopting an open information policy, which, for example, enabled all parties to follow closely the progression of the works and how the environment strong political pressures both within Scandinavia and actually responded to these. within Europe as a whole. A common feature of all the different phases of the The proposal to construct the Øresund link aroused project was the widespread uses of ICT solutions, not only substantial opposition, however, both in Denmark and in functioning in the role of technical problem solvers, but Sweden, and certainly much more than had been experi- also as facilitators for communicating complex decision enced in the case of the Great Belt link. In both cases, the problems to all interested persons and parties. This was marine environment was a major concern because of the done, for example, by simulating work procedures, threat that these constructions clearly posed to the local providing visualisation of design features and especially of marine environments, and even beyond, to the environ- information about environmental aspects related to the ment of the Baltic. However, owing to the bi-national ﬂow of water. nature of the Øresund Link project, its location in a The task of informing the politicians, the responsible densely populated region with 3 million people using the authorities and the public as a whole in a truthful and Øresund for recreation, and within an area of great realistic way about the consequences of the construction environmental sensitivity, the environmental debate and, inseparably from this, the empowering of these to became very intense and the environmental issue indeed act as genuine stakeholders in the management of came to have the biggest impact. the natural environment, was a task of hydroinformatics The task of constructing this link thus became not over and above its tasks within the physical project only one of the constructing of the physical works them- itself. selves, but also, and even essentially, one of answering to In order to perform this task within such a Public- the concerns, expectations and requirements of a diverse Private type of project, hydroinformatics had to employ range of formal and informal stakeholders. Thus, when existing technologies as well as introduce new tech- the consortium that was formed to construct the link nologies and adapt to new social-institutional arrange- (Øresundsbro Konsortiet, or ØSK) received permission to ments, even while contributing to elaborating the proceed with the construction, it was faced with a number strategies by which the project had to be designed of choices regarding how to address the wide range of and managed. Some of the technologies could be taken requirements from the authorities, interested parties and over from experience on other very large projects, but the general public—often against the background of an as some were quite original. Some of the strategies were yet undecided code of governance and regulation. hitherto untried. This paper describes the corresponding

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 107 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

developments, which made the construction of the link between Denmark and Sweden possible.

1 HISTORICAL BACKGROUND

1.1 The social breakthrough realised by the application of hydroinformatics

As with many other great engineering enterprises, the projects for constructing road and rail links between the two largest Danish islands and the European mainland, and between Denmark and Sweden, have been discussed for well over a century. Problems of technical feasibility, financial viability and political desirability have largely frustrated attempts to bring these projects to fruition up until quite recent times. With the greater economic and increasing political integration of Europe, many of these more traditional problems have been largely overcome. At Figure 2 | The location of the Great Belt Link and the Øresund Link. the same time, however, they have been joined by new problems, and in the first place by problems concerning the sustainability of the natural environment, both during and following the execution of the projects. In many cases, this new class of problems concerning the natural environ- As concerns the subject of the present paper, the road ment, commonly engaging the attention of large parts of and rail link between Denmark and Sweden, the way to its the general public, has made it much more difficult to realisation through the use of hydroinformatics was pre- realise projects that have an obvious and direct impact pared, for a large part, by two earlier projects. The first of upon the natural environment. As a result, persons these was that of the protection of the city of Venice responsible for the execution of many, and probably most, against flooding, a project that, although now extensively great engineering works can no longer restrict themselves investigated, has still not been realised. The second such to the physical realisation of the works, but must increas- project undertaken on the way to the present one was that ingly direct their attention to the political and public of the road and rail link across the Great Belt, between the acceptability of these works, and especially with respect to Danish island of Sjælland (and the many other islands the natural environment. This is to say that works that already connected to it) and the island of Fyn (which had were originally conceived in terms of steel and concrete been connected to mainland Denmark in the 1930s). The have now also to be conceived in terms of people and a first of these projects, that of Venice if executed, would be wide range of natural organisms. It was the emergence of of the same order of magnitude in financial outlay as the this new kind of socio-political/institutional environment project considered here, which was of the order of that led to the emergence of a new discipline, which was 3.7 billion US dollars, while the Great Belt connection was that of hydroinformatics (Abbott 1991). As will be somewhat larger, coming out at about 4.2 billion US explained here, this new discipline has played and will dollars, with both amounts being at the then-current continue to play an important, and sometimes essential, exchange rates. The positions of these links are shown role in making such projects possible. in Figure 2.

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 108 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

The protection of Venice also represents the type of chemical, biological and ecological modelling complexes project where the issue of the environment becomes employed (see Figure 3). determinant for the main design criteria. The proposed The Great Belt project overlapped with the Venice storm surge barriers in each of the three entrances to the project and continued after the Venice work had been Venice lagoon were specified (in Law No. 798 passed by placed in abeyance. The Danish Great Belt Link crosses the Italian parliament in 1984) to allow the free passage one of the most complex water bodies in the world. Due to of ships, but should at the same time not be detrimental its location between the brackish Baltic Sea and the saline to the lagoon ecosystem. The works should not cause the North Sea, the Belt is characterised by a persistent, but famous bathing beaches of Cavallino and Lido to be very dynamic, predominantly two-layer stratification. eroded and the flushing of the lagoon by the Adriatic Being the main connection between the Baltic and the should not be reduced. During the course of the project North Sea, the flow through the Belt is crucial for the investigations, it became clear that the pollution of the aquatic environment of the already heavily stressed Baltic. Venice lagoon was a major problem in itself and studies When the Danish Parliament passed the Great Belt Link were conducted to find methods of stopping a still ongo- Act in 1987, a key requirement was that the flow through ing morphological degradation and for restoring the eco- the Belt and, related to this, the salt balance in the logical environment. The severity and importance of the Baltic, should remain unchanged. This was popularly problems to be studied dictated not only the application summarised as a demand for a ‘zero solution’. of novel techniques, which involved combinations of The experience gained and the calibrations obtained physical and numerical modelling together with field between data and modelling results in the Venice project investigations, but necessitated some further develop- provided such a level of confidence in the technical ments of these. For example, to prevent Venice from capabilities of hydroinformatics that it could be used being flooded in the future, accurate and reliable fore- responsibly in the Great Belt project. In particular, it made casts of water levels will be required in order to operate it possible to set conditions on the design and construction the storm surge barriers in an efficient and environ- procedures that would guarantee that flows between the mentally friendly way. This required a modelling of the North Sea and the Baltic would stay almost unchanged Adriatic area, providing means to generate boundary if remedial dredging was used to compensate for the conditions at Venice for both long-period wave (tidal/ blocking effects of an artificial island and the connecting storm surge/seiching) and short-period wave motions. structures. In order to satisfy these requirements, the The long-period components were subsequently used for nearly-horizontal flow modelling was extended to two- on-line forecasting purposes together with measured layer flows, thus making it possible to represent most of water levels. Today a further refinement of this model the strong stratification effects. The introduction of such a encapsulated in artificial neural networks is used to warn vertical resolution in the modelled flows justified and the citizens of Venice when increased water levels will eventually necessitated the introduction of means for occur. Other developments that took place involved the measuring flow and density distributions in the vertical. use of nearly-horizontal flow modelling integrated with Following the use of chains of salinity- and velocity- chemical, biological and ecological modelling for study- measuring instruments in each vertical, acoustic Doppler ing of nutrient pollution and eutrophication in the equipment was also increasingly introduced. The use of lagoon. Morphological studies were carried out along vessel-mounted and towed equipment of this type the littorals and within the lagoon, and these required extended the sea areas covered and also the range of the coupling of wave, current and sediment transport phenomena that could be identified. These increasingly modelling covering cohesive as well as non-cohesive came to include physical and biological features of the sediments. Hydrographical and biological surveying of seabed. The introduction of radio links from instrument considerable areas of the lagoon was carried out, provid- networks through shore stations provided real-time data ing new means to calibrate and verify several of the for the control of numerical computations for real-time

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 109 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 3 | Venice—protecting a cultural pearl.

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 110 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

various parties involved to proceed with this project. It was at this point of persuasion of the various stakeholders to grant their consents, and to incorporate these into the enabling legislation and contractual agreements, that the social side of the sociotechnical equation of hydroinformatics ﬁrst became satisﬁed.

2 THE SOCIAL/LEGAL PREPARATIONS FOR THE DENMARK–SWEDEN TRANSPORT LINK

2.1 The political background and the Government treaty

As mentioned above, the governments of Sweden and Denmark have on various occasions over the last century Figure 4 | On-line monitoring stations in the Great Belt. been discussing the construction of a fixed link across Øresund. The institutional foundation for the now existing link was laid in 1985 (Dynesen 2000) when the two long-period and short-period wave characterisations and governments established a working group with the task of for other purposes. The combination of modelling results preparing the basis for a new Government agreement. The and data, much of it on-line, made it possible to predict group’s results were presented later the same year. How- many flow and short-period wave conditions during the ever, events were initially driven by a political desire to design and construction phases of the project, and thus the link Denmark together before it was linked to its neigh- on-line prediction of sediment plumes, this last by using bour, so that the Danish parliament decided to postpone the PARTICLE model developed during the initial phase the decision concerning Øresund and give instead the of the project. These provided the means to safeguard the highest priority to a fixed link across the Great Belt. marine operations and to allow construction works to However, a Danish political agreement was achieved in proceed more efficiently in space and time while satisfying 1986, stating that the government would work to establish strict environmental controls (see Figure 4). a fixed link across Øresund. On that basis, the Danish– In effect, this amalgamation of different technologies Swedish working group resumed its work and updated the provided such synergies that a new confidence was gener- technical and economic investigations. Further investiga- ated in the capacity of this new kind of integrating tech- tions into the environmental consequences of construct- nology, called hydroinformatics, to provide solutions ing a fixed link for the Baltic Sea and for Øresund were not only to unprecedented technical problems, but also included in the work. In 1989 it was concluded that there to hitherto intractable social problems concerning the was a general political will in both countries to establish integrity of the natural environment. either a bored railway tunnel or a combined rail and road As and by themselves, of course, these earlier develop- link between Copenhagen and Malmo¨ . Further studies ments did not suffice to justify the construction of the road and political negotiations in 1990 and 1991 finally led to and rail link between Denmark and Sweden. However, the government agreement regarding the construction of a without these developments it would have been much combined rail and road link. The original proposed layout more difficult, and perhaps impossible, to persuade the is shown in Figure 5.

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 111 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

and oxygen between the North Sea and the Baltic passing over the ‘sill’ of relatively shallow water, called the Drogden-Limhamn sill, between Copenhagen and Malmo¨. Many kinds of model simulations were run in order to illustrate the nature of the changes that might be expected and how the negative consequences of these could be almost entirely compensated by dredging the seabed to provide deeper water in the still-open areas. It was very clear to everyone, however, that the dredging processes themselves could just as well lead to serious subsea environmental damage, especially to eelgrass and mussel beds, with corresponding losses of the fish species living, breeding and feeding there. These processes, together with the general disturbance caused by the works, could also have a negative influence upon the extensive wildlife in Figure 5 | The original proposed layout of the Øresund Link. the area, such as ducks, geese, swans and seals. In Article 5 of the intergovernmental agreement it was stated that the final design of the Øresund Fixed Link should be selected with due consideration to what was 2.2 The physical nature of the transport link and its ‘ecologically motivated, technically feasible and economi- consequences cally reasonable in order to prevent detrimental impact on Despite the fact that there was a very broad political the environment’. The first part of the Danish enabling law support in the parliaments in both Denmark and Sweden and an English translation of two of its technically more to establish the fixed link, there was a very strong opposi- important paragraphs are presented in Figure 6. tion both from public opinion generally, and especially from political and environmental interest groups. The opposition against the construction of the link was based 2.3 The formation of the public–private partnership on two arguments. Firstly, it was claimed that it would project never be economically feasible to construct and operate the Fixed Link. Secondly, the opponents claimed that the The political challenge that then arose was one of construction of the link would have detrimental ecological identifying an operational way to accommodate these effects on the environment of the Baltic Sea due to immediate concerns as expressed by the opposition. This changes in the water exchange between the North Sea and accommodation had itself to be anchored in the enabling the Baltic. They were also alarmed about the effect on legislation, and this legislation had to be of a kind Øresund of the foreseen enormous releases of clay and that could be accepted by the Danish and the Swedish limestone sediments during dredging and reclamation parliaments on behalf of their respective constituencies. works. Being then of such a sociotechnical nature, this initiative It was obvious that the original proposed layout called for new social/institutional arrangements. The shown in Figure 5, with the artificial peninsula at the central institutional innovation was the formation of a Copenhagen International airport, the artificial island, the Danish–Swedish body, called in Danish the Øresundsbro pillars and the protective bases of the approach and high Konsortiet, charged with designing, building, operating bridges, would cause a large increase in resistance to flow. and financing the link and for informing and consulting The fixed structures would alter the structure of the flow as with the stakeholders in such a way as to obviate their well, and would thus alter further the exchanges of salt active opposition to the project (see Figure 7).

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 112 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 7 | Stakeholder diagram.

expectations expressed in the framework set out by the politicians in the two countries. Thus, right from the beginning, the project had to co-exist with ‘political-consumers’, i.e. stakeholders outside its immediate group of owners, users and employees, These ‘political consumers’ all wished to exert their influence in order to protect their own interests and exert influence on the link’s design and its operation. The enormous public attention given to the project and the lively environmental debate surrounding it meant that the environmental authorities in particular had to demonstrate that they were being independent and strong in all their dealings with Øresundsbro Konsortiet. A diverse range of organisations and private individuals all demanded influence, too. They were, and continue to be, active in the public debate, putting pressure on politicians and the authorities as well as on the Øresundsbro Konsortiet itself. Local residents, the sailing fraternity,

Figure 6 | The Government’s requirements. and companies located in the vicinity of Øresund, for example, have insisted on the adoption of particularly responsible attitudes during the construction period. The realities of the political ownership, the expectations within local communities, the requirements of the When the Øresundsbro Konsortiet was established in authorities, and a critical press, all led very early to the 1992 the project’s overall political, economic and environ- formulation of the first two management principles, as mental parameters were clear from the outset (Lundhus follows: (1) The participants in the process must be treated 2000). Yet the project had still to define its own visions, as partners (this also includes informal stakeholders). All targets and standards in order to fulfil successfully the involved authorities, consultants, and contractors are

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 113 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

‘information’ as a strategic management tool through the establishment of a Public Affairs Department.

2.4 The final permission and the authorities’ requirements

According to the Intergovernmental Treaty, the ordinary legislation of each of the two countries should be followed, meaning that Øresundsbro Konsortiet was obliged to prepare the necessary applications and documentation for obtaining approval from the relevant authorities in both countries. Although Sweden and Denmark are close neighbours, the legislation in the two countries is based on Figure 8 | The owner’s organisation. very different legal and administrative principles, rules and procedures. The very different nature of the approval procedures, and different political circumstances in the two countries, placed particularly high demands on the management of Øresundsbro Konsortiet in the years also regarded as partners and are included in the overall 1992–94. However, the very time consuming process management process for the project through common for obtaining the approvals also gave the Consortium a objectives. (2) Øresundsbro Konsortiet must always take good opportunity to develop the necessary organis- the lead in the negotiating processes and itself propose ation, contractual strategies and quality systems, as well solutions. as to optimise the design and develop appropriate At the early stages of the project—integrally with administrative and other procedures. the formation of Øresundsbro Konsortiet itself—it was As with the Great Belt Link, which also crosses waters necessary to identify the most important issues that designated as an international waterway (the so-called would require the formulation of deliberate policies and ‘T-Route’), the Øresund Link crosses, albeit within Danish strategies. This was important for both the establishment and Swedish territorial waters, navigation channels of the organisation and to ensure ease and efficiency in that possess international strait status (after the abolition managing corporate communications. The most important of the Øresund toll following the Copenhagen Treaty of these were (not in order of importance): time, economy, of 1857). Furthermore, according to the Espoo environmental quality, traffic expectations, working Convention, all states riparian upon the Baltic had to be environments, positioning of construction sites, labour given opportunity to express their environmental con- markets, end-users, contractors, technologies, risk cerns and requirements relative to the design of and per- management, aesthetics/architecture and safety (both mission to construct the link. Also, the passage through during construction and operation). Of these issues, Øresund during the construction period had to remain the one with the biggest impact on the project was open to navigation subject to international maritime regu- consistently that of the environment. lations regarding navigational aids and announcements. The organisation adopted (see Figure 8) reflected the These circumstances were taken up as challenges as well. importance given to the key issue through the formation of The approval process involved 10 public hearings the staff function ‘Environment & Authorities’ that was addressing different components of the project and more assigned the responsibility for controlling and reporting than 150 authorities were engaged in this process in compliance with the authorities’ requirements. It (Dynesen 1999). The basis for the hearings were environ- also reflected the high priority that was put on using mental impact assessments based on comprehensive

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 114 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

environmental investigations of all relevant subjects, leading to background material prepared for the consideration restricting the flow of highly saline and oxygen-rich of the authorities, ‘green’ organisations and the general water to the Baltic’. public. The material included more than 75 reports, com- The authorities specified in detail how the over- prising more than 10,000 pages. The media, and particu- all objective for the Baltic Sea should be achieved by larly the national and local press and television, Øresundsbro Konsortiet using the environmentally monitored, analysed, popularised, published and broad- optimised design and by undertaking compensatory cast the proceedings of the public hearings. For these dredging calculated using 3D models, i.e. the events and for the benefit of the media, many simulations so-called ‘zero-solution’, securing unchanged trans- of flow and environmental events were made and their port of water, salt and oxygen through the Sound. results were commonly presented in coloured-dynamic perspective formats. In Sweden, the establishment of the Øresund Link The environmental impact assessment for the was tested under the terms of the National Resources Act, Øresund area showed that sediment spill in connection the Water Act and the Environmental Protection Act. On with dredging and reclamation would be the most 16 June 1994 the Swedish Government approved the important cause of potential environmental problems construction of the Øresund Link under the terms of (Figure 9). The benthic communities in Øresund are the National Resources Act and the Water Act and in 1995 adapted to a natural high transparency of the water and a the licensing Board for Environmental Protection gave its low level of sedimentation. The most important and sen- permission to build and operate the fixed link on Swedish sitive benthic communities consist of large seagrass beds territory under the terms of the Environmental Protection dominated by eelgrass in shallow water around Saltholm Act. Then, in a verdict in July, the Swedish Water and large populations of common mussels on the hard Court gave permission to construct the Øresund Link bottom along the alignment of the Øresund Fixed Link. under the terms of the Water Act. At the same time the The benthic communities house an abundant invertebrate Water Court set out the environmental requirements fauna and are important as spawning, nursery and feeding which Øresundsbro Konsortiet had to meet in connec- grounds for a number of species of fish. Saltholm and its tion with construction activities in Swedish territorial surrounding waters constitute an EU Special Protection waters. Area and are waters of great importance as feeding, rest- On 8 July 1994, the Danish Government approved the ing, nesting and moulting places for many bird species and project. In March 1995, the Danish Minister of Transport a small population of common seals and a few grey seals approved, in consultation with the Minister of Energy (see Figure 10). and Environment, the Øresund project’s environmental In connection with the stipulation of the environ- quality objectives as well as the criteria and requirements mental requirements, the authorities defined an ‘inner for the control and monitoring programme for the link’s impact zone’, covering an area of 500 m either side of the Danish sections. Link trajectory, including compensation dredging areas, The authorities’ environmental quality objectives and and an ‘outer impact zone’ that extended 7 km on either requirements are summarised below. side of the inner impact zone.

Environmental Quality Objectives—Baltic Sea Environmental Quality Objective—Øresund

‘The Øresund Fixed Link must not cause any ‘The permanent loss of areas resulting from the change to the water ﬂow through Øresund, thereby establishment of the artiﬁcial peninsula, the

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 115 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

artificial island and the bridge piers, and permanent In addition, the Danish authorities established a effects resulting from local changes in hydrographic series of restrictions on benthic flora and fauna, conditions can be accepted.In the inner impact bathing water quality, bird foraging in shallow zone more extensive effects are acceptable, although waters, and other issues. special consideration must be shown to the area around Saltholm.In the outer impact zone, temporary changes (less than 5 years) are acceptable.Out- It was essential to the social acceptability of the side the outer impact zone the effects of the project that the extent to which these criteria were satis- construction work must not hinder the fulfilment of fied could be controlled by all the principal stakeholders the objectives and criteria for coastal waters as and that these should accordingly be fully informed in real stipulated in regional environmental plans.As far time of all relevant developments and empowered to inter- as concerns the open parts of the Øresund, the vene in all ongoing and planned processes if they consid- construction work must not reduce the possibilities ered that these conflicted with the pre-established for establishing an indigenous natural flora and agreements. The Øresundsbro Konsortiet had therefore to fauna.’ take account of stakeholder power at all times and In order to secure that the sediment plumes respond accordingly, if necessary closing down whole generated by the dredging and reclamation activities processes if these were found unacceptable to one of the would only give rise to temporary impacts, at levels stakeholders. The means by which this condition was fulfilling the objectives for Øresund the authorities realised are described later in this paper. established two types of environmental requirements:

3 INTRODUCTION TO THE HYDRAULIC DESIGN Technical requirements STAGE: ENVIRONMENTAL MANAGEMENT AND – dredged materials must not exceed 7.5 million STRATEGIES m3. The design of the link on its hydraulic and water environ- – sediment spillage had to be kept below 5% of mental sides followed from, partially overlapped and con- the total dredging volume. tinuously interacted with, the environmental debate. The – total spillage must be calculated with an design stage was itself composed of two parts, the one that accuracy of ± 20%. in its final form would meet the overall requirements and – spillage must be limited in intensity, time and the other concerned with elaborating construction pro- space with due regard to fishing and the flora cedures that would satisfy the above criteria during the and fauna. construction period. The following examples may serve to illustrate how Biological requirements environmental management was perceived and performed – the impact on biomass of eelgrass and by the owner so as to ensure that these requirements and common mussel had to be below 25% of the expectations were met. Later, the hydroinformatics tools total biomass in the impact zone during the used and developed to achieve these goals will be described. construction of the link. – dredging in both channels was not allowed to 3.1 Environmental optimisation of the design be carried out simultaneously, thus avoiding The Intergovernmental Treaty had left one very important the migration of herring. question open for later decision: should the Øresund Link be constructed as a so-called ‘zero solution’ or not?

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 116 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 9 | Example of simulated dispersal of dredged material spill and sedimentation using the PARTICLE model.

Øresundsbro Konsortiet very soon realised that it general criteria of an ecologically motivated, technically would not be politically acceptable—especially consider- feasible and economically reasonable design of the Link. ing the fact that the Great Belt Link under construction Three main alternatives were chosen and presented to was designed as a ‘zero solution’—to deliver an optimised the authorities, to an international expert panel and to the design of the Øresund Link that was not optimised in general public through the public hearings. The aim was to relation to the Baltic Sea. As a consequence, Øresundsbro find a solution that reduced the blocking effect of the Konsortiet developed a large number of alternative constructions to a minimum, while providing substantial designs in order to find a solution that would fulfil the reductions in the compensatory dredging volumes and at

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 117 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 10 | The environmental concerns for the Øresund were related to potential effects from sediment spill on shading on macro algae, impact on mussels, fish migration, birds and bathing water quality.

the same time holding the investment cost at a reasonable artiﬁcial island and peninsula. Approximately 2.5 million level in order to fulﬁl the governments’ requirement that m3 of these were associated with deepening and realign- the Link should be paid back within a return period of ment of navigation channels. Compared with the link approximately 30 years. The consequence of this environmentally optimised design (see Table 1) of the link was that the strongest Table 1 | The result of the environmental optimisation* process political argument against the Link was eliminated, leading to the Swedish government’s approval of the con- 1991 project Final design

struction of the link on Swedish territory in June 1994 Peninsula 1.0 km 0.5 km (under the condition that a number of environmental Tunnel 2.0 km 3.8 km demands were met). In addition, the risk of long-term environmental effects on the Baltic Sea was minimised. Island 2.5 km 4.1 km In order to fulﬁl the zero solution for the optimised Bridge 10.0 km 8.0 km

layout described above, a design was prepared in which Dredging volumes 13.7 million m3 7.5 million m3 approximately 4 million m3 of soil had to be excavated Blocking 2.3–2.5% 0.3–0.5% from the seabed in the region of the link alignment to compensate for the blocking by the bridge pillars, the *Total length of the Link cannot be compared due to changed alignment.

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 118 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 11 | The final layout of the Øresund Fixed Link.

described in the government agreement, which was During the initial phase 1990–91 and in the design calculated to have a blocking effect of approximately optimisation phase 1992–94 the environmental conse- 2.3% and would, to obtain a ‘zero solution’, require quences of the proposed link were studied using a number compensation dredging of approximately 11.7 million of 2D deterministic free-surface flow models. These cubic metres of seabed materials, a considerable part models, which were all generated using the MIKE 21 of the risk of local environmental impact was thereby modelling system, were applied to calculate the flow- eliminated. blocking of the proposed structures and the approximate A considerable effort was also put into the aesthetics compensation dredging volumes required to obtain of the project in order to increase its acceptability further. a neutral effect (‘zero solution’) on the flow through the Thus, for example, the artificial island was ‘sculpted’ into a Øresund. Video animations of the general flow pattern in pleasing form for the millions of persons flying over it into the area and how this would be modified by the presence Copenhagen airport, while its stone wave protection was of the link formed part of the reporting, including simu- carefully chosen in a uniform dark-red colour in order to lations of sediment spreading from dredging operations provide visual pleasure for those sailing past it. The final using the Particle model. Although in the feasibility phase layout of the link is shown in Figure 11. this was a relatively simple model set-up, the animations

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 119 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

thereby provided the understanding of the complex the impacts to acceptable levels. However, because of the local and regional flow phenomena. These also provided unpredictability of the hydrographical and meteorological the confidence that solutions that would satisfy the regime and natural variations in the ecosystems, the actual requirements would be obtainable. state of the environment during the construction works had to be monitored in order to ensure the fulfilment of the general environmental criteria. 3.2 Management and execution of the works in an The feedback procedures included: environmental context 1. the assessment and approval of equipment, work Early in the process of preparing the environmental plans and other control tools, prior to the initiation impact assessment, or EIA, the potential major impact of operations; in the Øresund area was identified as originating in 2. the application of threshold criteria and feedback the dispersal of spilled sediments from the dredging and loops with an agreed code of action, and reclamation activities. Consequently, a great effort was put 3. a clear definition of the respective responsibilities of into organising the environmental monitoring in order to the involved parties. ensure that the criteria for the emissions would be met. 1999 However, the fulfilment of the framework type A robust modelling tool (Dynesen & Dietrich ) requirements set out by the Danish and Swedish authori- was required for an optimal planning of the distribution in ties demanded the development of a completely new time and space of the total allowable (5%) spill that would environmental strategy. It became clear that the philoso- ensure the fulfilment of the biological requirements. The phy of the strategy should build on feedback principles development of such a strong and reliable mathematical combining detailed planning of the dredging operations tool relating the environmental load (spill rates) to the online with monitoring and modelling activities. The com- biological impacts proved to be an essential element bination of monitoring and models assisted in keeping in assessing the impacts of the requirements, and a track of the process of impact development, thus enabling major challenge for Øresundsbro Konsortiet and the adjustments to be implemented in the construction work consultants. in due time to ensure fulfilment of the objectives. The manner in which this model was integrated into The integration of the technical and environmental the feedback monitoring/modelling system so as to follow project aspects and contractual commitments of the con- the actual impact from the construction works will be tractor in the fulfilment of the environmental require- described later. However, because the statutory requirements became very important elements in this strategy. ments could not be implemented directly in terms of a Two major tools were introduced to ensure that the contractual and construction practice, the model became spill was kept below the limits necessary to fulfil the very important in the process of developing contractual objectives and criteria for all the variables: requirements, operational criteria and specifications for the control and monitoring programmes. Thus the model The contractor was held responsible through his • provided criteria making it possible to decide whether or contract for keeping the spill below specified limits not the works would be in compliance with the statutory varying in time and space, taking into consideration requirements, within an acceptable statistical margin of the environmentally sensitive periods and areas. uncertainty. A feedback-monitoring programme was • By using this model it was possible to combine the implemented to covering sediment spill, the spill rates with the most important biological impacts on dispersal hereof and biological key variables the vegetation caused by the spill. All other biological representing the most sensitive benthic communities. criteria were accordingly transformed into criteria which In principle, the contractor’s fulfilment of the criteria conformed to the model output, such as sediment concen- of maximum spillage alone should have sufficed to reduce tration and/or sedimentation. For example, fish migration

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 120 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

criteria were interpreted as a maximum allowable sediment concentration of 10 mg/l in the waters in periods of herring migration. The aesthetic bathing criteria and the bird foraging criteria were also transformed into simple sediment concentration criteria representing a minimum visibility of the water in the area of the public bathing season and bird foraging season. Also the settling and conditions of the common mussels and larvae were related back to maximum allowable rates of sedimentation in the sea bed area covered by the mussels banks. The impact area around the link (<7 km) was divided into six eelgrass impact zones reflecting the intensity of the dredging operations, and the model was used to define the geographical and chronological distribution of spill in the different impact zones. The resulting spill rate distribution was integrated into the dredging contract between the owner and the dredging contractor. This transformation of general ‘political’ requirements to contractual requirements and specifications for the Control & Monitoring Programme can be illustrated in the following way (see Figure 12). The division of responsibilities between the owner and the contractor meant in general terms that the contractor had to take full responsibility for the environmental effects Figure 12 | Transformation of the authorities’ requirements. of his own choice of detailed design, construction methods, choice of materials, etc., while the owner took full responsibility for the establishment of the Fixed Link and for securing that the total consequences of more The dredging and reclamation activities required in than one contractor’s work did not result in environ- connection with both temporary and permanent works mental effects exceeding the bounds set by the overall comprised 17 separate work plans. For each of these, environmental requirements. detailed EIA’s were made and reported in a ‘Dredging The owner used several different tools in order to Instruction’ submitted to the authorities for approval. make sure that this distribution of responsibilities The authorities performed an independent control, functioned adequately in practice. based on an annual survey of the environmental con- The first of these was incorporated in the ‘Dredging ditions in Øresund in general. The authorities were, on Instructions’ (Graveinstruks), which served as a quality this basis and on the basis of the owner’s reports, respon- document for administering the requirement set up by sible for reporting to the parliaments and governments Øresundsbro Konsortiet to reduce the spill to 5% on of the two countries. average and to reduce the spill in time, extreme intensity and space. The second tool was the Feedback Monitoring 3.3 Information management Programme, which served both as a planning tool and as a tool for a timely control of the environmental In order to keep the public actively informed about the consequences of the ongoing works. plans and progress relating to the environment as well

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 121 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

as to the other aspects of the project, Øresundsbro ‘sharing and dissemination of knowledge’ events material- Konsortiet already in its early days established a Public ised in conference proceeding volumes and at present three Affairs Department. books about the Øresund Link Project featuring the same This department also functioned as the project’s aspects are in the process of being published. own sounding board and was central for picking up ‘sig- nals’ from the surrounding world. A principal aim was to establish a common identity for the many thousands of employees on the project by keeping them updated and 4 THE HYDROGRAPHIC MONITORING AND in turn encouraging them to become the project’s MODELLING EMPLOYED AT THE DESIGN AND ambassadors and to inﬂuence attitudes. Open and CONSTRUCTION STAGES transparent information policies were applied to all aspects of the link and through all project phases. The hydrographic conditions in Øresund were investi- Communication took place through many different gated during a comprehensive survey programme that types of media, such as press releases, TV interviews and continued over the period from February 1992 (when the broadcasts, newsletters, brochures, annual reports, video Øresundsbro Konsortiet was formed) to June 2000. The taking and participation and presentations in local meet- scope of this programme was to contribute to the data and ing places. Exhibition centres on both side of the Sound knowledge base used for the design of the link. In order to functioned as important information centres where the provide data for the environmental situation in Øresund many facets of the infrastructure project from computer and to prepare the general environmental impact assess- graphics dreams to physical reality were narrated and ment and to target the control and monitoring pro- illustrated by means of miniature models and multimedia gramme, a number of environmental studies were shows. Information also became generally available undertaken over the period 1992–1995. Detailed baseline through the Internet when the consortium’s web site studies were conducted for establishing the natural opened some time after the actual construction works turbidity and for mapping the distribution of the dominant were initiated. Similar kinds of information were pro- communities in Øresund. These communities included vided, especially for the printed media, with more detailed eelgrass, shallow water vegetation, macroalgae com- descriptions made available as downloads. The actual munities, mussels and shallow and deepwater fauna. The progression of the works could be followed directly by a aim of the baseline studies was that of describing the number of on-site web cameras and a photo archive with spatial and seasonal variation of selected key variables. stills and aerial views was maintained. Alongside the surveys, investigations were carried out and As described later, a central environmental information data collected aimed at determining the individual plant and management system, called EAGLE, was developed and animal communities’ vulnerability to the products of linking the control and monitoring programmes together. the construction work. This system was also accessible via the Internet, being primarily targeted at the formal stakeholders but with links 4.1 Calculation of the zero solution by numerical to computers at the Exhibition centres allowing a broader modelling audience as well to observe and learn that the environment was being well cared for in real time. During the last year As explained above, the Danish and Swedish authorities before completion, three international conferences, jointly deﬁned the overall environmental requirement that the organised by Øresundsbro Konsortiet, the main contractors Fixed Link across Øresund must not have any longer term and international associations, were held, one in Malmo¨ impact on the marine environment in the Baltic Sea: the focusing on bridges and two in Copenhagen about the chal- so-called ‘zero solution’. lenges, solutions and lessons learned within the subject area Øresundsbro Konsortiet initiated the arrangement of Dredging, Reclamation and Submerged Tunnels. These that the Swedish Meteorological and Hydrological

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 122 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Institute (SMHI) and the Danish Hydraulic Institute (DHI) independently should employ suites of two different advanced ‘state-of-the-art’ 3D models for the calculation of the zero solution. The two suites of models were calibrated and validated by the use of data collected in the hydrographic monitoring programme. SMHI used Phoenics-based models for this purpose, but these were not employed for all 3D modelling. All the DHI models of this kind, which were used throughout the project, were generated from the third-generation, System 3, and subsequently from its fourth-generation successor, the MIKE 3, modelling packages. The models used were thus fully capable of resolving the dynamics of the flow processes in the vertical, as well as in the horizontal, planes. All models of this kind were of second-order accuracy in their dynamic and between third and fourth order in their advective terms. These modelling systems were themselves Figure 13 | Still from MIKE 3 dynamic simulation of a salt wedge on its way to the further developed as a result of their use in this project, Dragør-Limhamn Sill. and especially in the direction of the description of turbu- lent mixing in the vertical in the presence of strong density gradients on the basis of acoustic Doppler measurements. For the model calculations, a 72-day design period was The basic types of data recorded in the hydrographic chosen among data collected through the hydrographic survey, that covered the entire period from February 1992 monitoring programme. It was documented that this to June 2000, were the following: period could be considered representative for the hydro- speeds and directions of currents graphic conditions on the Drogden-Limhamn sill. The • water temperatures design period also included an extreme inflow event, • salinities which was important for the transport of salt and oxygen • turbidities to the Baltic, and this was suitable for assessing the • water levels achievement of the zero solution requirements set by • wind speeds and directions the Danish and Swedish authorities. The selection of • short-period wave heights. the design period was based on data from the Drogden • Channel by considering the surface current speed, the The permanent monitoring network utilised equipment current duration period and the surface salinity. with on-line data transmission facilities. Figures 13 and 17 show the movement of saltier Great attention was given in this project to the accu- water from the Sound into the Baltic. Figures 14, 15 and racy of positioning of all components, whether these were 16 illustrate the measurements made to control and anchored or towed instruments, model grid points or update the numerical simulations already shown in constructional elements. GPS equipment was employed Figure 13. The layout of the permanent instrument array extensively and played a major part in ensuring the suc- used at the design and construction stages is shown to cess of the entire project. Within this context, it was the left in Figures 14 and 15. This was augmented shown again in this project that a much greater attention by predefined transits with vessel-mounted acoustic than hitherto had to be given to the chart and map Doppler and other measuring equipment, as exemplified projections that were employed for positioning and in Figure 16. recording the positions of instruments and for setting up

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 123 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 14 | Comparison between observed and simulated currents at a depth of 15 m below mean sea level. The observed currents were obtained by the permanent on-line monitoring system and the simulated currents were computed with the MIKE 3 modelling system.

the models, as well as during the construction process. Blocking estimates for water and salt transport were Considerable development has already occurred earlier in calculated with both the MIKE 3-generated models and this area also (see Abbott 1997). In particular, systems had the Phoenics model of SMHI, while only MIKE 3 was used to be set up to facilitate the transformation from one for calculating the oxygen transport. On the basis of the projection and one instrument survey model grid to model results for salt and water transport, combined another. The models set up for this project used a blocking estimates with associated uncertainties were Universal Transverse Mercator (UTM) projection for the calculated. These combined results were calculated as numerical grid generations, and transformations were pro- weighted average values of the results from the two models. vided from other isometric projections, from Lambert’s Ultimately, the model results were applied to demonstrate conic orthomorphic projection, from the Mercator projec- that all acceptance criteria were fulfilled for the final tion most commonly used on sea charts, and from a range layout of the link, including the compensation dredging. of stereographic projections, and especially the polar stereographic projection. These facilities were of course supported by the usual range of graphics tools. 4.2 The feedback monitoring programme A particular significance accrued in this respect to the then rapidly ongoing release of nautical charts in digital Being responsible for the control of the marine biological form conforming to the DX90 standard as defined by the impacts of the different contractors’ construction International Hydrographic Association. works, Øresundsbro Konsortiet developed a Feedback

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 124 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 15 | Comparison between observed and simulated salinities at a depth of 30 m below mean sea level over the same period as used in Figure 14. The observed currents were obtained by the fixed on-line monitoring system and the simulated were computed with the MIKE 3 modelling system.

Monitoring Programme. The overall objective of the moni- and reclamation operations as part of the feedback monitoring programme was to ensure that the environmental toring programme. The bases for the model complex were objectives and related design criteria set up by the authori- then frequently updated with the actual measured spill ties in the two countries were fulfilled both during and and the hydrodynamic boundary conditions. The integrity after the construction work. of the calibration and verification procedures used in The primary means to ensure compliance with the the individual models were also checked regularly by environmental requirements were provided by the plan- comparing model results with hydrographic field data. ning of the dredging and reclamation works and the sub- The two-dimensional, one-layer, free-surface-flow sequent preparation of environmental assessments based modelling system, MIKE 21, was used to compute most of on numerical model calculations and the introduction of the environmental consequences of the marine construc- data assimilation principles. The feedback made it poss- tion activities. The model complex consisted of hydro- ible to plan the execution of the construction work and dynamic modules, sediment spreading and sediment implement adjustments of the ongoing activities in due transport modules and an ecological module. The model time, thus reducing the risk of violating the criteria while set-up is illustrated in Figure 18. at the same time minimising the risk of stopping the The model complex was applied to describe the future ongoing activities. (forecast modelling) and the past (hindcast modelling) in A comprehensive numerical model complex was used the Øresund project. In the hindcast modelling the actual to simulate the environmental impact from the dredging dispersal of spill and the ecological impact were modelled

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 125 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 16 | Comparison between observed currents measured in a surveyed transect and MIKE 3 simulated currents.

using measured hydrodynamic and meteorological bound- field measurements, and that field measurements and their ary conditions. The hindcasts resulted in updated bases databases were usually of even less value unless their for forecast modelling, upon which even more accurate products were integrated in and related through numerical forecasts could be constructed. modelling studies (Abbott 1991). The Sound project As introduced earlier, the Feedback Monitoring extended this understanding further by integrating many Programme provided content for both formal stakeholders different kinds and sources of data to provide the bases for (owner, contractor and authorities) and informal stake- a more complete knowledge of environmental develop- holder (public, media, etc.) consumption, as well as for ments, and even as a means to promote a better under- technical and environmental design and construction standing of events. Thus, for example, the development of purposes. A further institutional innovation was then submerged grass meadows (covering some 132 km2 of the necessary as a means of linking to and mediating between seabed in the immediate vicinity of the link) was followed this programme and its ‘market’. This innovation took the using satellite observations, aerial photography, towed form of a Feedback Monitoring Centre (FBC), which will acoustic Doppler surveys and numerical and data-driven also be described later in this paper. Figure 19 illustrates modelling. This combination of facilities was later aug- the layout of the system employed. mented by advanced data mining techniques for automat- By the time that the Sound project was initiated it had ing the classification of the extent and type of bed been clearly demonstrated, and especially from the Venice vegetation (a system that is presently being further devel- and Great Belt investigations, that numerical modelling oped by DHI into a general sea bed mapping tool). Simi- was of limited value unless it was supported by adequate larly, the positioning of each of the 50,000-ton submerged

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 126 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 17 | 3-D modelling of salt intrusion into the Baltic Sea. Animations of this kind have made the regional oceanography generally well known to the public, the NGO’s, and the politicians (Moeller 2000).

Figure 19 | Principles of Feedback Monitoring (Adaptive Monitoring).

that stirred up sediment, from silts and clays deposited Figure 18 | The MIKE 21 set-up used in the Feedback Monitoring Programme: coupling during the works themselves to limestone deposits laid of the model complex and the data flow between the modules. down 60 million years ago, and from the most primitive algae to the most intelligent and sociable of birds and mammals. tunnel elements, carrying a four-line highway and a two- Some notion of the coverage of the data-collecting track railway, was steered from on-line data sources con- programme may be provided from Figures 20–23. necting into encapsulations into neural networks of Figure 20 shows the extent of eelgrass in the Sound in the numerical modelling simulations and other, historic, data vicinity of the link, the transects along which observations sources (Babovic et al. 2000). were conducted and the areas that were subject to aerial The measurement programme also covered unprec- photography. Eelgrass (Zostera marina) formed one edented scales, ranging from tides and storm surges in the important part of the bed vegetation generally and its North Sea and the Baltic to local short-period wave ﬁelds condition and extent were employed accordingly as an

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 127 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 21 | Distribution of common mussel (Mytilus edulis)inØresund before the construction of the Øresund Fixed Link (Øresundsbro Konsortiet 2000).

Figure 20 | Distribution of eelgrass in Øresund. The indicated transects were surveyed in the compliance investigations in 1997–1999 (Øresundsbro Konsortiet 2000).

indicator of the condition and extent of the bed vegetation generally. Figure 21 shows the 15 km2 extent of blue mussels (Mytilus edulis) and the distribution of mussel-dedicated measuring stations in the immediate proximity of the link. Altogether there were 46 km2 of mussel beds in the area Figure 22 | Coverage of common mussels in Øresund, November 1999, based on a underwater photo survey using DHI’s vessel operated photo sampler inﬂuenced by the link construction (where a mussel bed ensuring high productivity and accurate positioning enabling precise was deﬁned as an area of the seabed that was covered by monitoring of potential effects (Øresundsbro Konsortiet 2000). more than 40% by mussels). It was estimated that there were on average some 92,000 tons of live mussels on these beds. The extent to which mussels were distributed in the immediate neighbourhood of the link is shown in Figure 22. Observations of this kind continue, as outlined also had to be investigated and monitored before and later in this paper. during the construction works. The distribution of the There are of course many other forms of seabed fauna observation stations for the monitoring of fauna in rela- than eelgrass and mussels and the development of these tively low water (as indicated by Macema) and in the

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 128 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 24 | The dipper dredger Chicago.

Figure 23 | Monitoring stations included in the assessment of the impacts on the benthic fauna communities in Øresund (Øresundsbro Konsortiet 2000). sometimes traceable over distances of 100 km under strong and unidirectional current inﬂuences and short- period wave induced disturbances in shallower waters.

deeper water (as indicated by Abra-Amphibera) is shown in Figure 23. 4.3 Dredging and reclamation Such species-specific monitoring programmes were connected to other more general and often even more During construction of the Øresund Fixed Link intensive chemical and biochemical investigations, such approximately 7.5 million m3 material in total was as those directed to the release of deposits of heavy metals dredged from the seabed. The dredging was in fact neces- during dredging operations, the release and distribution of sary not only to compensate for the reduction in water waste water around the Sound, the release of nutrients in flow caused by the link structures but also for construction general from surrounding land areas and the distributions purposes. and variations in oxygen balances in the waters influenced The dredged material was utilised for the construc- by the link. From both the instrumentation and the tion of the artificial island south of Saltholm (called modelling sides, all of these investigations had been Peberholm) and for the peninsula located east of Kastrup thoroughly prepared and rehearsed in earlier projects, and Airport. These reclamation operations took place behind especially from the Venice and Great Belt studies. closed bunds made of gravel dikes with a protective layer This long period of preparation and testing, and the of rocks outside. A mechanical dipper dredger and a associated elaboration of the required instrumentation hydraulic cutter-suction dredger carried out most of the networks, suitable (e.g. SQL) databases and compu- major dredging operations. The dimensions of these pieces tational and graphics software, was also essential to the of equipment may perhaps be gauged from Figures 24 rapidity and reliability with which such resources could be and 25. A number of smaller backhoes were used for mobilised in the Sound project. the dredging of pier foundations, the realignment of All of these studies were in turn related back to the navigation channels and for other more local operations. influences of the link-construction activities, and in par- The introduction during this period of Version 4.0 of ticular to the plumes of fine materials that these released, MIKE 21, with its seamless integration of GIS and

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 129 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

dredging materials in combination with measurements of turbidity and current speed. Further to this, the contractor had to observe the daily and weekly spill rates, even while keeping the overall spill below the stipulated 5%. This implied that the contractor sometimes had to move around with the high capacity dredging equipment (principally the cutter-suction dredger) from impact area to impact area to exploit the environmentally available spill rate capacities. In situations with strong currents in the Øresund it was sometimes necessary to stop dredging Figure 25 | The cutter suction dredger Castor. temporarily with the cutter-suction dredger during the peak of the currents. This was done approximately 20 times (Jansen 1999). The times at which this occurred were determined by implementing an additional facility (Early hydraulic/environmental modelling facilities, together Warning System) to the spill monitoring system. This with its facilities to rescale and reorientate the aerial transferred the resulting momentary spill value after com- photographic views and to incorporate these into GIS pletion of a monitoring line automatically to the early contexts, provided the means to introduce the influences warning computer at the site office, thus allowing almost of the ongoing construction works on the environment in on-line calculation of the cumulative spill for the day. If real time. At the same time, the use of on-line GPS this appeared to threaten the daily spill rate it could be positioning of construction works and the rapid transla- decided to stop the dredging. Stopping at 16:00 hrs and tion of these positionings between different cartographic then starting again at 20:00 hrs, for example, proved to be projections provided the assurance that these updatings of acceptable when knowing the trend of the spill and the flow fields and their environmental consequences were existing current data and extrapolating with forecasted properly founded. current data applicable to the areas influenced by the The conditions for the dredging and reclamation dredging. As a further example, a situation typically work, such as the maximum allowable dredging volume occurred where the contractor applied for permission to and sediment spill, were set out by both the Swedish and exceed the spill rates stated in the Dredging Instruction. the Danish authorities. As explained earlier, the overall This request was then analysed with the feedback condition in relation to the spill from dredging and recla- model, using the actual experienced current conditions mation was that the total spill in connection with the (hindcast) and these results compared with measure- construction of the Øresund Link must not exceed 5% of ments of the environmental variables combined with a the total amount of the material extracted in the alignment forecast of the contractor’s proposed work plan. Based zone, corresponding to a maximum of approximately on this evaluation, permission could often be given to 375,000 m3. proceed and complete the work within the specified Material was defined as spill if it could be transported timeframe. away from the work zones or the reclamation areas by the Further to this, a probabilistic analysis of the spill prevalent currents and waves. Work zones were defined, monitoring system was carried out to establish methods as explained above, as regions with a radius of 200 m for more rapidly estimating the amount of spill. This around all dredging and reclamation areas. The quantity approach was used to reduce the frequencies of the ship- of spill had to be documented by measurements by ship based spill measurements. Thus, the estimation approach (see Figures 26 and 27) and fixed stations at outlets from was applied in order to predict the spill in periods where the sedimentation basins. The spill from the dredging no spill measurements were carried out based on informa- operations was also calculated, based on data about the tion from periods where spill measurements were carried

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 130 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 26 | The Contractor’s spill monitoring—vessel set-up.

out, as influenced by incoming real-time hydrographic and 3 and 4 mm that occurred in the worst conditions was dredging data. comparable with the background rate of sedimentation from natural bed material, such as could be accommodated environmentally by the bioturbation activities of 4.4 Results achieved with the Control and Monitoring bed-colonising organisms. Programmes and related hydroinformatics activities Extensive and still-continuing investigations of fish breeding, development and migration have shown that the Since a particular importance accrued to the conse- period of the construction activities was marked by very quences of the dredging activities, a major part of the different variations for different species. For example, hydrographic investigations were directed to this aspect. Figure 29 shows extreme variations in the concentration The sediment plumes of dredging spill extended at least of elvers and small eels at five locations around the site. 80 km north of the dredging zone, up into the Kattegat, Other species show less and different—and even and some 120 km south into the Baltic. The distribution of opposite—variations both with regard to numbers and age spilt sediment in the region of the works is shown in distribution. However, a causal relation between the Figure 28. It is seen that there was no substantial impact in observed (and opposite) variations of fish stocks and the the particularly environmentally sensitive area of the ‘sill’, construction works is not likely and is not supported by with the greater part of the spill accumulating in deeper the observed impacts on bottom vegetation and benthic and less sensitive areas. The sedimentation of between communities. Moreover, very considerable variations in

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 131 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 27 | Monitoring vessel moving through a sediment plume on a survey line 200 m outside the work zone.

fish stocks have been increasingly observed in recent years Figure 28 | Spilled sediment accumulated on the seabed as per 1 January 2000 (hindcast modelled) in the Øresund model (Øresundsbro Konsortiet 2000). in most sea areas. Indeed, this situation has led to the introduction of severe European Union imposed quotas on fish landings within the Union. The effects of the works on bird life appear to have 4.5 EAGLE been less pronounced. Figure 30 shows the estimated and observed number of eider ducks nesting on the island of From the sociotechnical point of view of hydroinfor- Saltholm, in the immediate vicinity of the works, indicat- matics, the key to the success of the Denmark–Sweden ing a reallocation of nesting. Figure 31 shows the variation road and rail link was the introduction of new means in the number of grey geese moulting on Saltholm and to ensure the active participation of stakeholders in all the number hatching on the Swedish side of the works. environmentally sensitive decision-making processes. Figure 32 shows the variation in the number of swans These means were provided by an Internet-enabled system moulting around Saltholm. A considerable effort was for the real-time provision of information about all expended and continues to be made to avoid the intro- environmentally critical processes occurring during the duction of predators, such as foxes, rats and cats, onto construction process. This system, which provided an Saltholm and Peberholm. overview of all significant events in text and graphical The number of seals observed on and around Saltholm form, was called EAGLE. It was maintained by the FBC, actually increased during the construction period. The which was itself initiated and operated by a consortium of population of the seal reservation on Saltholm remained four organisations (including DHI and SMHI) on behalf constant, at between 20 and 30 individuals. of Øresundsbro Konsortiet. The FBC was itself one of the

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 132 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Figure 31 | Number of greylag geese moulting on Saltholm and breeding at the Scanian lakes. Data before 1993 were derived from other survey programmes (Øresundsbro Konsortiet 2000).

Figure 29 | Abundance of elvers and small eels in shallow water (0.4–1.2 m) at five locations in Øresund in 1992–1999. The results from 1992–95 represent the baseline (Øresundsbro Konsortiet 2000).

Figure 32 | Number of mute swans moulting on the shallows of Saltholm, 1993–1999 (Øresundsbro Konsortiet 2000).

Figure 30 | Estimated and observed number of eider nests on Saltholm. Estimates are based on breeding parameters recorded during the period 1993–1999 (Øresundsbro Konsortiet 2000).

A second database was that run independently of the FBC by DHI and SMHI themselves, which contained immense major suppliers of environmental data and model results amounts of material from surveys, simulations and physi- to EAGLE and at the same time acted as Øresundsbro cal model tests. The third source of data was Øresundsbro Konsortiet’s feedback adviser and information manager. Konsortiet’s database of the earthworks, tunnel and bridge All feedback management and advice came in time to be constructions, positioning arrangements, surveys, con- based upon the EAGLE system. In the words of Jensen & struction quantities, soil samples and other such sources Lyngby (1999): ‘The backbone of the environmental con- of information of an engineering nature. trol and management during the construction phase [was] These three databases were linked together to provide the environmental information system EAGLE.’ a unified EAGLE database. The expert staff of the FBC The EAGLE system drew upon three databases. The were responsible for assembling and formatting the first was provided, as already explained, by the FBC itself material so provided so as to make it ‘digestible’ to its on the basis of its own feedback monitoring programme. ‘end-users’, or ‘consumers’. The material was most com- This contained data collected during the monitoring sur- monly presented within a standard geographical informa- veys as well as data from the baseline studies, satellite tion system (GIS) environment. This type of content was images, coastal morphology photographs and other items. accompanied by other types, such as graphs, figures,

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 133 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

The construction of the link further increases the accessibility between Southern Sweden and Eastern Denmark and brings together a new conurbation of Copenhagen, Malmo¨ and other population centres in the Swedish province of Scania. This potential for an increase in economic activity in consequence was—and is still—seen as an additional attraction of the link and has provided it with a more local political support. The vision is that the ﬁxed link will span economic, administrative, institutional, technical, cultural and other barriers (Matthiessen 1999) initiating new social-economic dynamics in the Øresund area—a new metropolis in a high position in the European urban hierarchy. However, these various geopolitical ambitions and Figure 33 | Example of a Feedback menu from EAGLE showing work zones, weekly spill rates and measured spill (Thorkildsen 1999). more local political aspirations were strongly opposed by a large part of the general public, environmental interest groups and the political opposition, which was concerned about the impact of the link upon the natural environ- images and reports. The latter included incident reports ment. Given the proven ability of such concerns to stop or describing conclusions of investigations and corrective reverse infrastructural projects, there was then every actions taken as a result of the feedback procedure. An expectation that the environmental movement could example of a menu from EAGLE is shown in Figure 33. easily lead to a large number of confrontations, ultimately During the course of the project, EAGLE evolved into preventing the realisation of this project too, unless a powerful tool for decision making. Øresundsbro this was tackled in a new way. Øresundsbro Konsortiet’s Konsortiet’s management depended upon it for the latest basic concept therefore was to be a client that valued environmental information, while the Danish and constructive partnership with all the parties involved. Swedish authorities and other stakeholders were provided Consequently, a strong emphasis was put on quality with the same level of information at the same time. and transparency. This overall tendency away from an exclusively representative democracy and towards direct participatory democracy (itself now being more and more catalysed by the technologies of the Internet) made it clear that public and political support could only be 5 CONCLUSION regained through allowing a much wider dissemination of The construction of the road and rail connection between environmental information and, in particular, information Denmark and Sweden, when taken together with the on the potential consequences in a readily understandable construction of a very similar link across the Great form. In the present case the accommodation of these Belt, completed the direct land connection between the concerns was achieved by the owner through adjusting the Scandinavian peninsular and mainland Europe. It original alignment as well as the link’s technical design provided a physical symbol of European unity within the with the objective of optimising the link—technically, Scandinavian context, which was of vital importance to economically and environmentally—in order to reduce any the geopolitical ambitions of the European Union (EU). harmful effects on the environment. Furthermore, this Being part of the Trans European Network (TEN) approach made it possible to implement strategies that Programme the Øresund Fixed Link was subsidised by enabled a trustworthy control of the environmental the EU. impact during the construction stages and to supply

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 134 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

information continuously on the actual environmental on communication technology. Although conditions. exceedingly rudimentary and limited compared with With regard to the subject of the present paper, the the potentialities of currently ongoing developments role of hydrodynamics in the completion of the Øresund based on the new telecommunications technologies, Fixed Link, hydroinformatics certainly took an important the EAGLE system can be seen as a precursor of place, right from the beginning of the feasibility phase presently planned and future decision support through to its completion. It was fully integrated into the systems that can be distributed across the general planning and design, as well as the execution and control public (see for example Abbott and Jonoski 2001). phases of the project. Its significance, which may prove 4. These developments not only served the interests of relevant for similar or other kinds of future construction the project and the general public but also provided projects having water and an environmental side to them, substantial advantages to the contractors and others may be summarised as follows. who were concerned with the construction. For example, the dredging works were able to proceed 1. The environmental monitoring and modelling with a high level of efficiency, while the widespread programme that emerged from the concerns of many use of hydroinformatics and GPS systems organisations and the public generally was substantially reduced the costs of the placing of rock accordingly directed to the purpose of dissemination armour, the tunnel elements, bridge piers and other of environmental information, as well as to the structural elements. There were thus important needs of the construction process itself, and business advantages, much less negotiation and very institutional arrangements were made little litigation as compared with similar large correspondingly. This emergence of a complex of infrastructural projects carried out elsewhere in aims and methods, of public aspirations and Europe. government legislation, of citizen groups and 5. An important consequence of this situation was that institutional arrangements, of criteria and controls, the link was built to specification, its construction of instrument networks and modelling tools, and, and operation satisfied all environmental conditions, most generally, of people and technical equipment, it was constructed within its original budget and it was prepared by and remains the ongoing subject of was completed six months ahead of schedule. This hydroinformatics. was a compliment not only to all the parties 2. It seems reasonably established that, without the concerned, but also to the achievement of promises held out by the development of hydroinformatics as a new discipline in this field of hydroinformatics already in the early 1990s, the endeavour. transport link between Denmark and Sweden could only have been realised with great difficulties, despite its support by powerful political interests. Of course, without the political support it would never ACKNOWLEDGEMENTS have succeeded either, regardless of any amount of development in hydroinformatics. The success of the The authors wish to thank Professor Mike Abbott, project was the result of a synergy between these Consultant to DHI Water and Environment, and Environ- two streams of developments. It was a consequence mental Consultant Bo Mogensen, DHI Water and of a synergy between a new kind of political will and Environment, for their help in the preparation of this new kinds of sociotechnical means. paper. The authors also wish to compliment the following 3. From the point of view of hydroinformatics, the directors, project managers, engineers and biologists from Øresund project marked the transition from an the former DHI and VKI as well as SMHI: Jacob Steen emphasis on information technology to an emphasis Møller, Anders Jensen, Erland Rasmussen, Kurt Jensen,

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021 135 M. Thorkilsen and C. Dynesen | An owner’s view of hydroinformatics Journal of Hydroinformatics | 03.2 | 2001

Jens-Erik Lyngby, Anders Højgård and Jonny Svensson. Dynesen, C. & Dietrich, J. 1999 How essential were the Their excellent work in the application of hydro- environmental requirements—alternatives? Proceedings of the Øresund Link Dredging & Reclamation Conference, informatics has been vital to the successful design of the Copenhagen, Denmark, Ed. Iversen and Mogensen, link on its hydraulic and environmental sides both during Øresundsbro Konsortiet, Øresund Tunnel Contractors and and after the construction and has assisted in changing International Tunnelling Association. Jansen, E. 1999 Introduction to spill, spill monitoring and spill the public attitude towards the project from a very management, Øresund Marine Joint Venture. Proceedings of sceptical one in the beginning to a very positive one the Øresund Link Dredging & Reclamation Conference, towards its completion. Copenhagen, Denmark, Ed. Iversen and Mogensen, Øresundsbro Konsortiet, Øresund Marine Joint Venture and Central Dredging Association (CEDA) pp. 175–184. Jensen, A. & Lyngby, J. E. 1999 Environmental management and monitoring at the Øresund Fixed Link. Terra et Aqua, 74, REFERENCES March 1999 (http: www.iadc-dredging.com/terra-et-aqua/ 1999/74content.htm) A selection of computer animations from the Øresund Fixed Link Lundhus, P. 2000 The owner’s organisation and management Project can be found at http://www.dhi.dk/dhiproj/country/ principles. Proceedings of the Øresund Link Immersed Tunnel denmark/oresund/index.htm Conference, Copenhagen, Denmark, Ed. Øresundsbro Abbott, M. B. 1991 Hydroinformatics: Information Technology and Konsortiet, Øresund Tunnel Contractors and International the Aquatic Environment, Avebury, Aldershot, UK. Tunnelling Association. Abbott, M. B. 1997 Range of tidal hydraulics. J.Hydraul.Engng. Matthiessen, C. W. 1999 Potential integration of Copenhagen and ASCE 123, 4, 257–275. Malmo¨ /Lund—a cross border project on the European Abbott, M. B. & Jonoski, A. 2001 The democratisation of decision Metropolitan Level. Proceedings of the Øresund Link Dredging making in the water sector II. J.Hydroinformatics 3(1), 35–48. & Reclamation Conference, Copenhagen, Denmark, Ed. Babovic, V. et al. 2000 Real-time current forecasting for tunnel Iversen and Mogensen, Øresundsbro Konsortiet, Øresund element towing in Øresund, Denmark. Proceedings of the Marine Joint Venture and Central Dredging Association Øresund Link Immersed Tunnel Conference, Copenhagen, (CEDA) pp 1–10. Denmark, Ed. Øresundsbro Konsortiet, Øresund Tunnel Moeller, J. S. 2000 Environmental management for the Contractors and International Tunnelling Association. Øresundsbron—integrated monitoring and modelling. Presented Dynesen, C. 1999 Environmental management and the impact on at Shoreline 2000—Information society, 26 May 2000, Nice. execution. Proceedings of the Øresund Link Dredging & University of Nice and IAHR European Division, France. Reclamation Conference, Copenhagen, Denmark, Ed. Iversen Øresundsbro Konsortiet 2000 Environmental impact of the and Mogensen, Øresundsbro Konsortiet, Øresund Marine construction of the Øresund Fixed Link. Øresundsbro Joint Venture and Central Dredging Association (CEDA), Konsortiet (http://www.oeresundsbron.com/infobank/ pp. 51–59. us/publications/regionpubl/rep12.htm) Dynesen, C. 2000 The original project constraints and Thorkilsen, M. 1999 Feedback monitoring—implication on the pre-conditions on the project—what was imposed as dredging works. Proceedings of the Øresund Link Dredging & obligations. Proceedings of the Øresund Link Immersed Tunnel Reclamation Conference, Copenhagen, Denmark, Ed. Iversen Conference, Copenhagen, Denmark, Ed. Øresundsbro and Mogensen, Øresundsbro Konsortiet, Øresund Marine Konsortiet, Øresund Marine Joint Venture and Central Joint Venture and Central Dredging Association (CEDA) Dredging Association (CEDA), pp. 213–220. pp 193–203.

Downloaded from http://iwaponline.com/jh/article-pdf/3/2/105/392226/105.pdf by guest on 30 September 2021