This study presents the structural response of a Submerged Floating Tunnel (SFT) subjected to earthquake loading and other relevant static loading types. The SFT is a competitive solution for crossings of deep seas, canals and fjords, as it is submerged in water and floats at a fixed submersion depth. It is supported by its buoyancy and tethers, connected to the seabed to maintain the structure at an even submersion level. Influence on the environment is low compared to other competitive structures, as the structure does not occupy any surface of the seabed, except for the tether foundations. Moreover, the environment has little influence on the SFT’s operability, as it is insensitive to harsh weather conditions. However, in areas of high seismicity, the SFT’s dynamic characteristics must be carefully tuned to the expected type of earthquake to optimize its performance. Previous research showed that special care should be taken of the transition between the SFT and land tunnels, as this joint has proven itself to be the main challenge.

The purpose of this research is to alter the transition structure between the SFT and adjacent land tunnels, such that an optimal dynamic response of a SFT is found to seismic events (Chapter 1). Initially, its performance is measured by comparing the seismic serviceability limit state (SLS) response to the static ultimate limit state (ULS) response. Subsequently, for more severe earthquakes, it is measured by comparing the SLS seismic stresses with maximum allowed levels of concrete prestress.

To analyse the SFT behaviour, a global model has been created of the SFT+land-tunnel system in Chapter 3, which is built by means of a linear-elastic finite element method in Python. It accounts for dynamic loads through the soil and tethers, as well as static loads by the structure's weight, traffic and buoyancy. The influence of stagnant water is accounted for by the Morison equation. Soil-Structure-Interaction (SSI) is incorporated using a Substructuring method. The properties of the soil and tethers are found in local sub-models, after which the total structural response of the global model is solved in the frequency domain. Later, its response is transformed to the time domain to obtain a time-series of displacements and forces. Its dynamic characteristics are studied by comparing Fourier spectra with SFT natural frequencies, which gives insight in the influence of design choices on the time-domain response.

The global model is validated with a replica model in the finite element software DIANA FEA (Chapter 4). A realistic case study is defined based on a previous TEC project in Chapter 5 and later applied in a parameter study in Chapter 6. Here, the effects of various design choices with respect to the transition structure are monitored using 3 earthquake signals. 5 standard end-joints and 2 special end-joints with seismic base isolation and viscous dampers are tested. The effects of Multi-Support-Excitation is studied by letting earthquakes horizontally approach the SFT with various wave speeds and angles of attack. Finally, the effect of SSI is compared with a non-SSI case, to see its contribution to the research.

In Chapter 7, the results are discussed and in Chapter 8 a conclusion is provided together with recommendations for both TEC and future research.