Optimal choice: assessing the probability of additional damage to buildings caused by water level changes of larger areas
Abstract
In the Northern parts of the Netherlands exploration of natural gas reservoirs causes subsidence over large areas. As a consequence, the water levels in canals and polders have to be adjusted over time in order to keep the groundwater levels at a constant depth relative to the surface level. In the middle of the subsidence area it is relatively easy to follow the settlements by a uniform lowering of the water level. This would however result in a relative lowering of the groundwater table at the edges of the subsidence area. Given the presence of soft compressible soils, this would result in induced settlements. For buildings in these areas this will increase the chance of damage. A major design challenge lies therefore in the optimisation of the use of compartments. The more compartments the higher the cost therefore the aim is to make compartments in the water management system that are as large as possible without causing inadmissible damage to buildings. In order to asses expected damage from different use of compartments three tools are needed. The first is a generally accepted method of damage determination, the second a method to determine the contribution to damage of a new influence, e.g. a groundwater table change. Third, and perhaps most importantly, a method is needed to evaluate effects not for single buildings but for larger areas. The first need is covered by established damage criteria like those of Burland & Wroth or Boscardin & Cording. Up until now the second and the third have been problematic. This paper presents a method which enables to assign a contribution to the probability of damage of various recognised mechanisms such as soil and foundation inhomogeneity, uneven loading, ground water level changes. Shallow subsidence due to peat oxidation and deep subsidence due to reservoir depletion can be combined. In order to address the third issue: evaluation of effects for larger areas, the method uses a probabilistic approach. Apart from a description of the method itself validation of the approach is described by applying the theory to an area in the North of The Netherlands, near a canal, where a water level change was considered. This area consists of soft soil overlying sandy deposits. It was found that the damage percentages as given by the theory are in the right order of magnitude when compared to the actual damage observed in this area. For study of a large area affected by subsidence, input parameters have been established based on a field inspection of the state of the buildings present in that area. Groundwater changes resulting from water level adaptations have been calculated based on measurements of the present situation for ten cross sections. Results of the analyses show that lowering of the water table by 0.1-0.15 m influences a relatively small zone next to the canal only. Use of the new damage assesment method shows that even within this influenced zone, which is mostly less than 10 m wide, effects on buildings are rather limited. Almost in all cases the increase of chance of building damage is negligible.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFMNH33B..04B
- Keywords:
-
- 1822 HYDROLOGY / Geomechanics;
- 1873 HYDROLOGY / Uncertainty assessment