It is well documented that groundwater levels in a number of cities in this country and abroad have been rising over the past few years. Parts of central London are experiencing rises of one metre per year, while in Paris, previously dry basements are now affected by severe flooding. The extraction of good quality water from city wells for a variety of industrial and domestic purposes led to a general lowering of groundwater levels from the early 1800s. During the period of depressed groundwater levels development within restricted city centre sites required higher buildings with deeper basements and foundations. Communication networks also moved underground to ease congestion at the surface. A large percentage of this work was carried out during a period when the groundwater was tens of metres below ground level; a situation accepted as being stable at the time. Consequently the effect of a higher water table was often not considered in the geotechnical design of structures. Within central London the water table is still very deep and may not cause problems for years to come. In Birmingham the situation is more critical and the rise in groundwater level is already causing problems. Computer model predictions play a leading role in predicting tomorrow’s groundwater levels. Evidence suggests that groundwater levels in the near future will attain those of natural conditions. The effect that these levels will have on current structures requires assessment. Simplified calculations for the London area show that the bearing capacity of deep basement foundations and piles could be reduced by 25’30% if the rise in groundwater level continues. High levels of sulphates have been recorded during recharge experiments and these will produce adverse effects on unprotected foundations. Problems with high artesian pressure during construction are also likely if the aquifer is confined. These is an urgent need to assess the geotechnical risk to engineering structures with respect to the rising groundwater levels. This should be carried out now so that problems can be properly anticipated and solutions developed in good time.
No takes yet. Share an insight, caveat, or question.
Hurst et al. (1986) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: