Comparing Groundwater Drawdown with Estimated Influence Radius – A Case Study of Infrastructural Projects in Sweden

Abstract: Infrastructural projects may sometimes require excavation of soils and the groundwater table to be lowered, temporarily or permanently. As there are risks in connection to groundwater lowering, the extent of the affected area is of interest. The distance from the source of the drawdown to the point of unaffected groundwater table is known as the influence radius, and can be analytically estimated by a number of formulas, using input based on aquifer properties. Using data from two infrastructural projects in Sweden, the formulas could be evaluated in respect to groundwater level measurements and actual influence radius. The aim of this study was to compare different the results of the formulas to observed drawdown, as well as to evaluate the sensitivity of the formulas, i.e. how changes in input yielded changes in influence radius. Data from the two projects were used as input to the formulas, where the output could be compared to time series of groundwater table measurements in order to evaluate the accuracy of each specific formulas in each case. The sensitivity analysis was carried out by changing the values of the input data, one parameter at a time within a range of typical values, and evaluating the change that occurred. The evaluation was made by comparisons between the original calculation of influence radius and the new set of influence radii, calculated by changes in input parameters. A large change in the value of influence radius indicate a high sensitivity and vice versa. The calculated influence radii varied largely between the two cases, and no clear result as to the accuracy of the formulas could be seen. What this implies is that the choice of formula when estimating influence radius matters greatly, and that a few different formulas should be used if the input data is available. For greater knowledge of the suitability of the formulas, a greater number of case should be investigated, but overall the theoretically derived formulas, with a greater number of input parameters, seem to be more reliable. The sensitivity analysis showed that a certain formula could have different sensitivities, depending on the magnitude of the input – a small change on the low part of the input range could have a greater change on the influence radius than a large change at the high part of the influence range. When making estimations, it’s thus a good practice to use a range of input values, i.e. minimum and maximum values, for a better estimation of influence radius. Hydraulic conductivity is a particularly important parameter when calculating the influence radius, and is oftentimes hard to determine exactly. Using a safety margin for the input when using the formulas is a good method for a better understanding of the extent of the influence radius. While hard to determine a single parameter associated with a high uncertainty, the empirical formulas did exhibit a larger sensitivity than the theoretical, further promoting the use of theoretical formulas in general when possible, and Theims (confined) well equation in particular, regardless of the aquifer type.