A framework for Risk Assessment of Groundwater Drawdown Induced Subsidence
Paper in proceedings, 2015
Sub-surface constructions generally involve drainage of groundwater, which can induce land subsidence in compressible soil deposits and cause extensive damage costs in urban areas. A probabilistic framework, in accordance with the risk management framework outlined by the International Standard Organization (ISO), for assessing risks of groundwater drawdown induced subsidence is presented here. The framework consists of five modules: (1) A stratified geostatistical (Kriging) procedure for probabilistic spatial analysis of soil layers. This module is necessary for a detailed understanding of the soil stratification, drainage paths, and their potential spatial variations; (2) A stochastic hydrogeological model capable of representing possible groundwater drawdowns for a specific sub-surface construction; (3) A stochastic subsidence model; (4) A model for estimating the economic consequences and calculating the risk, i. e. the expected cost, of groundwater induced subsidence; and (5) A module for evaluating the need for additional information to reduce the risk of erroneous decisions with respect to risk acceptance criteria based on economic Value of Information Analysis (VOIA), i. e. a cost-benefit analysis (CBA) of additional information collection alternatives for suggested strategies to reduce or control subsidence. The modelled land-area is represented by a grid with calculation points. When the three first modules are linked together in a Monte Carlo-simulation, it is possible to estimate the spatial distribution of probability of subsidence and evaluate the sensitivity to different model and parameter assumptions. An estimation of the risk of subsidence is performed by combining the probability of land subsidence with the locations and expected damage costs of existing buildings across the modeled area (module 4). With sensitivity analysis, significant weaknesses can be identified and robust safety measures at locations with significant risks for subsidence can be planned for. Uncertainties can be communicated by mapping and comparing different outcomes of the model, e. g. the expected value and the 95th percentile of the risk. Together with affected stakeholders the assumptions and the outcomes of the model should be discussed - both how well the model describes the system dynamics and how safety measures should be implemented.
value of information analysis