An early warning system for groundwater flooding in the Chalk

Abstract

An early warning system has been developed for groundwater flooding and trialled in the Patcham area of Brighton. It provides a fit-for-purpose approach for forecasting groundwater flood events in the Chalk and is capable of operating across longer time scales than had previously been possible. The method involves a set of nested steps or tasks. Initially, the catchment's response to recharge is determined and, using a representative hydrograph, a simple regression model that relates annual groundwater level minima and autumn and winter rainfall to subsequent annual maxima is developed. The regression model is then applied at the end of each summer recession using the observed annual minimum and estimates of winter rainfall to predict the following groundwater level maximum. Based on the results of this prediction a variety of steps may then be appropriate. Where the model predicts potentially high groundwater levels the frequency of groundwater level monitoring observations can be increased. A novel element of the method developed is the monitoring of changes in the matric potential of the unsaturated zone. Specific trigger levels to initiate either the next step of the method or promulgation of warnings of varying severity will be developed through experience of use of the system.

Process-based modelling to evaluate simulated groundwater levels and frequencies in a Chalk catchment in south-western England

“…Groundwater levels were simulated in quite a few studies (Adams et al, 2010;Ladouche et al, 2014;Jimenez-Martinez et al, 2016) but mostly relied on very simple representations of karst hydrological processes and disregarding the scale discrepancy between borehole (point scale) and modelling domain (catchment scale) at which they were applied.…”

“…This may bring some advantage concerning approaches that used transfer functions (Jimenez-Martinez et al, 2016) or regression models (Adams et al, 2010) for estimating groundwater levels if enough data for model calibration and evaluation are available.…”

Groundwater flood or groundwater-induced flood?

“…As a result, the southern region of the UK received 201% of its long-term average rainfall during September-December 2000 and there was no precedent for the intense recharge the aquifer received that winter. The likely mechanism of the surface flooding at Patcham was simply the saturation of the Chalk matrix, starting initially above poorly permeable marl and hard bands and working upwards until the entire matrix was saturated between the normal water table elevation and the ground surface (Adams et al 2010). The result was surface ponding with the flood waters characteristically blue/grey in colour, standing up to a metre deep at one location.…”

A lumped conceptual model to simulate groundwater level time-series

“…These tools have been used in the past to reconstruct and extend groundwater level records (Conrads and Roehl, 2007), forecast extreme events in the near future (Adams et al, 2008;Daliakopoulos et al, 2005) as well as aid investigations into long-term water resource sustainability under climate and land use stresses (Goderniaux et al, 2009;Jackson et al, 2011;Sun et al, 2011).…”

“…Indeed others have emulated this approach since (Anaya and Wanakule, 1993;Barrett and Charbeneau, 1997;Kazumba et al, 2008;Thiéry, 2012). Lumped models have also been used to simulate spring flows and groundwater levels in highly heterogeneous aquifers (Keating, 1982), estimate aquifer parameters (Olin, 1995), and to develop early warning systems for groundwater flooding (Adams et al, 2008). Flores W et al (1978) and Pozdniakov and Shestakov (1998) made stochastic simulations of groundwater levels using linear reservoirs with parameters drawn from a probability distribution.…”