On the relevance of the initial conditions in field tracer tests monitored by time lapse ERT: the Settolo (Italy) case study

A problematic aspect characterizing field tracer tests monitored by hydrogeophysical methods, such as Electrical Resisitivity Tomography (ERT), is a correct definition of the initial distribution of the solute in the injection volume. An incomplete or erroneous knowledge of the initial conditions can in fact heavily affect the interpretation of the tests, typically resulting in large errors on the estimation of the soil parameters obtained by inverse modeling procedures. Although this problem is broadly recognized, still little attempts have been made to develop methodologies suitable to avoid or minimize these impacts. In the present work we discuss the setup and realization of a field tracer test in Settolo, an alluvial phreatic aquifer in a piedmont area of North-Eastern Italy, exploited for irrigation and drinking water supply. A specific configuration has been realized by means of a 30 m deep well equipped with a filter specifically designed to make the saline tracer initial distribution as uniform as possible along the vertical of the injection well. The method used to design the filter is illustrated, along with the results of the test as collected by the ERT instrumentation and some recommendations for the design of similar experiments.