A comparative engineering geological study of HATS stage 1 Tunnels C and F

Abstract

The objective of this dissertation is to undertake a comparative engineering geological study of information generated during the design and construction of HATS Stage 1 Tunnels C and F. This dissertation also touches on delays which occurred during construction and how the ground conditions affected the construction programme and cost. Furthermore, similarities and differences in the geological settings which influenced the construction of the tunnels have also been assessed.

Various published papers, reports and government documents have been collected and retained by the Drainage Services Department (DSD), which contain a wealth of engineering and geological data relating to the HATS Stage 1 tunnels. Data has been obtained from various sources, including the Geotechnical Information Unit (GIU) of the Civil Engineering Development Department (CEDD) and the Hong Kong Geological Survey (HKGS) archives and is presented in the form of maps, tables and figures extracted from available reports and records and from the author’s review of available information.

Tunnels C and F were constructed in Jurassic age granitic and Cretaceous age volcanic and granitic rocks. A comparison between predicted and actual ground conditions reveals that as-encountered zones of adverse rock mass are 6 times lower than predicted for Tunnel C and 2.5 times lower than predicted for Tunnel F. It is observed that there is an approximate relationship between the major faults (≥ 3m wide) and changes in the number of joint sets in all rock types, with exception of the Rennie’s Mill Fault in Tunnel C. The number of joint sets tend to decrease in the direction of the centre of the granite plutons, contrary to what is observed in the volcanic rocks. The older granitic rocks tend to exhibit a greater number of joint sets compared with the younger granites. In general, the volcanic rocks are more heavily fractured compared with the granites.

There is a relationship between locations and orientations of the major faults and distribution of groundwater inflows along the alignment of Tunnel F. In the latter, faults are considered to have provided a cut-off barrier to groundwater inflows over a section of tunnel 1800 m in length. In contrast, groundwater inflows of varying magnitudes are observed along the full length of Tunnel C. Moreover, groundwater patterns in Tunnel C appear to be affected by scale. At the small scale there does not appear to be a relationship between faults and groundwater, but at basin (Junk Bay) scale, faults clearly played a role in groundwater flow and affected the spatial distribution of groundwater drawdown and settlement of nearby land.

The onset of drawdown and settlement associated with construction of Tunnel C is not thought to be linked to particular geological features per se, but rather to a period of significant inactivity, during which time construction of the tunnel was halted and groundwater inflows into the tunnel were uncontrolled. In the latter stages of Tunnel C construction, settlement is thought to be linked to poorly controlled groundwater inflows into the tunnel. In the case of Tunnel F, it is also thought that poorly controlled groundwater inflows are the main reason why settlement occurred at the HIT Container Terminal. Promulgating the lessons learned from driving the HATS Stage 1 tunnels to the general public would be beneficial. Furthermore, information generated from this study could be used as a tool to help predict ground conditions for future underground projects in Hong Kong.