Settlements above fractured crystalline rock masses are rarely observed and in the past, geotechnical engineers would not expect substantial subsidence to occur in association with a deep tunnelling project. However, recent high precision levelling measurements along the Gotthard pass road in central Switzerland, have revealed up to 12 cm of subsidence along sections that pass several hundred metres above the Gotthard highway tunnel. Large-scale consolidation resulting from tunnel drainage and pore pressure changes in the rock mass are believed to be the contributing mechanism. This paper presents results from an extensive numerical modelling study focussing on the processes responsible for this subsidence. Results derived from 2-D discontinuum (i.e. distinct-element) and continuum (i.e. finite-element) modelling show that both discrete fracture deformation (i.e. fracture closure and shear) and poro-elastic consolidation of the intact rock matrix equally contribute to the observed magnitudes and shape of the surface subsidence trough.
Des tassements aux dessus de massifs cristallins sont rarement observes et par le passe les geotechniciens ne prevoyait pas de subsidence significative associee avec des projets de tunnel profond. Cependant, de recentes campagnes de nivellement haute precision le long de la route du Col du Gotthard (Suisse centrale), ont mis en evidence des subsidences atteignant 12 cm sur des sections passant plusieurs centaines de mètres au-dessus du tunnel routier du Gotthard. Le mecanisme invoque pour expliquer ce phenomène est la consolidation à grande echelle due au drainage du tunnel et aux variations de pression de pore dans le massif rocheux. Cet article presente les resultats d'une modelisation numerique intensive mettant l'accent sur les processus responsables pour la subsidence. Ces resultats obtenus grace à des modèles 2-D discontinus (p. ex. elements distincts) et continus (p. ex. elements finis) montrent que la deformation sur des fractures discrètes (fermeture de fractures, cisaillement) ainsi que la consolidation poro-elastique de la roche intacte contribuent egalement aux magnitudes et aux geometries de subsidence observee en surface.
Beachtliche Oberflachensenkungen in einem kristallinen Gebirge, deren Ursache im Vortrieb eines mehrere hundert Meter tiefgelegenen Tunnels liegt, wurden bisher nicht erwartet und beobachtet. Dennoch wurden nach dem Bau des Gotthardstrassentunnels, entlang der Gotthardpassstrasse Oberflachensenkungen mit einem Maximalbetrag von 12 cm bei der Auswertung von Prazisions-Nivellements entdeckt. Dieser Maximalbetrag wurde in einem granitischen Gneis gemessen. Die örtliche Übereinstimmung zwischen dem maximalem Senkungsbetrag und dem maximalem Wasserzufluss zum Tunnel, und die zeitliche Abfolge von Tunnelbau und auftretenden Oberflachensenkungen lassen hydromechanisch gekoppelte Prozesse als Ursache dieser Gebirgsverformungen vermuten. In dieser Arbeit werden die Ergebnisse ausgedehnter prozessorientierter numerischer Simulationen prasentiert. Resultate aus den 2-D Diskontinuum (d.h. distinct-element) und Kontinuum (d.h. finite element) Simulationen zeigen, dass beide Verformungsmechanismen, d.h. an diskreten Bruechen und der intakten Gesteinsmatrix, gleichbedeutend fuer die beobachteten Senkungsbetrage und der Form des Setzungstroges verantwortlich sind.
In 1998, the Swiss Federal Office of Topography completed a routine high-precision levelling survey over the Gotthard Pass road in central Switzerland. Comparison with the results from a survey over the same route in 1970 revealed that up to 12 cm of subsidence had occurred over a 10 km section that passed several hundred metres above the Gotthard highway tunnel. Earlier surveys between 1918 and 1970 showed only a steady alpine uplift rate of approximately 1 mm/year. Temporal and spatial relationships between the measured settlements and construction of the Gotthard highway tunnel (started in 1970 and completed in 1977), pointed to causality between water drainage into the tunnel and surface deformation. Geodetic triangulation measurements supplemented with Global Positioning System data have since confirmed the existence of the settlement trough 1.
An extensive field, laboratory and numerical modelling investigation was initiated to explore and explain the processes and mechanisms underlying the subsidence 2. As explanations relating to localized surface processes (e.g. a deep, creeping landslide) could be excluded given the absence of local indicators and the 10 km extent over which the settlements were measured, the working hypothesis pointed to surface subsidence associated with deep drainage.
