Hydrogeological Assessment For A Land Reclamation Dewatering Operation

The Kingston Foreshore Redevelopment was an urban renewal project in the suburb of Kingston on the southern shore of Lake Burley Griffin in Canberra. It involved reconfiguration of the foreshore by the excavation of part of the existing foreshore and the reclamation from the lakebed to form a harbour. Conventionally, reclamation process for “wet conditions” involves end dumping of granular fill, with subsequent vibro-compaction or dynamic compaction stabilisation. However, this method would not be possible since local sources of granular fill were not economically viable. Moreover, while the excavated materials from the existing foreshore were expected to be firm to stiff clays, the dumping of these excavated “lumpy” clays into the ponded reclamation areas without compaction, and the reliance upon subsequent preloading for stabilisation, would be difficult and problematic. The more feasible option in the geotechnical context was the “dry reclamation”, where the reclamation areas was dewatered and fill placed in the dry. This approach required that the reclamation areas to be enclosed and water be pumped out and discharged into the lake. The selection of a dewatering system would in turn be a function of the hydrogeological model and the level of drawdown required.

This paper focuses on the hydrogeological model and properties of the site that were considered to be critical for the assessment and design of the construction options. The assessed hydrogeological model in the foreshore areas comprises an upper fine-grained alluvium, which acts as an aquitard to restrict the flow of the water from the lake to the underlying gravel/sand aquifer. The gravel/sand aquifer is therefore likely to be semi-confined. Laboratory tests and full scale pumping/recovery tests were undertaken to estimate the permeability and storativity of the gravel/sand aquifer. Back-analysis of the pumping test results indicated that the permeability values derived from various analytical methods such as the distance drawdown analysis and the recovery analysis compared reasonably well with those estimated using the more simplified Hazen’s (1911) empirical method, which was related to the particle sizes of the gravel/sand materials. No in-situ permeability test was conducted in the upper alluvial aquitard. The hydrological conductivity of this layer was instead inferred from the dissipation test results obtained from CPTU soundings. It had been shown that the inferred permeability of the upper alluvial aquitard compared reasonably well with published correlations.

A number of dewatering schemes were assessed by undertaking 2D Finite Element Analysis (FEA) to check the viability of various options and sensitivity to changes to soil permeability properties. A summary of the various FEA results is presented.