The use of elevation models to predict areas at risk of groundwater flooding

Abstract Groundwater flooding, which occurs when the groundwater table rises in response to exceptional recharge rates either to the ground surface or to a point where subsurface infrastructure is affected, has been recognized as a significant issue with real economic impacts. A methodology has been developed to produce maps of groundwater flooding susceptibility, using geological and hydrogeological data. While good geological map data are available in digital form for England and Wales, there are much less data on water levels. These levels are usually measured during the construction of water boreholes, and while there is a national groundwater level monitoring network for regulatory purposes, at a national level data are sparse. To assist in developing a comprehensive map of water levels, the British Geological Survey (BGS) has adopted a number of strategies for data interpolation for areas with limited water level data and a surface has been derived from a terrain model by considering interactions between groundwater and surface water in rivers and lakes. When comparing the calculated levels against the available field measurements, a high correlation was found to exist. However, it was considered that in areas where bedrock aquifers dominate, this interpolated surface was probably inaccurate, and so refinements were developed to improve the modelled water levels surfaces. The resulting groundwater levels have been used to develop maps of areas where shallow groundwater may pose a risk. With potential changes in groundwater recharge postulated as a result of global climate change, identifying areas prone to flooding from groundwater, or areas where groundwater is likely to increase the impact of surface water flooding, is increasingly important.