Modelling borehole flows from Distributed Temperature Sensing data to monitor groundwater dynamics in fractured media

International audienceFractured aquifers are known to be very heterogeneous with complex flow path geometries. Their characterization and monitoring remain challenging despite the importance to better understand their behavior at all spatial and temporal scales. Heat and correspondingly temperature data have gained much interest in recent years and are often used as a tracer for characterizing groundwater flows. In the current work, a fast computer code is developed using Ramey and Hassan and Kabir analytical solutions which converts the temperature profile to the flow rate profile along the borehole. The method developed is validated through numerical simulations. A global sensitivity study recognizes the media thermal properties as the most influential parameters. For testing the method in the field, fiber-optic distributed temperature sensing (FO-DTS) data were used to monitor the dynamic behavior of fractured aquifers at the borehole scale at the Ploemeur-Guidel field site in Brittany, France. DTS data are used to infer the flow rates in the different sections of a fractured wellbore (flow profile) and calculate the contribution of each fracture to the total flow. DTS data were acquired for about three days in three different hydraulic conditions corresponding to two different ambient flow conditions and one pumping condition. Flow profiling using distributed temperature data matches satisfactorily with results from heat-pulse flow metering performed in parallel for cross-checking. Moreover, flow profiling reveals the daily variations of ambient flow in this fractured borehole. Furthermore, it shows that during ambient flowing conditions, shallow and deep fractures contribute roughly equally to the total flow while during the pumping condition, the deepest fractures contribute more to the total flow, suggesting a possible reorganization of flow and hydraulic heads depending on the hydraulic conditions. Thus, although the proposed method (DTS data and proposed framework) may be costlier and is based on indirect characterization through temperature measurements, it provides real-time monitoring of complex fracture interactions and recharge processes in fractured media. Thus, this method allows for a full analysis of the temporal behavior of the system with a simple and fast analytical model. Furthermore, thanks to its narrow width, DTS can be used and installed in boreholes for long-term monitoring while heat-pulse flow metering may lead to head losses in the borehole and may not be always possible depending on some borehole conditions. One of the limitations the approach proposed is the proper knowledge of the thermal properties of media required to infer the flow rate from the temperature. Nevertheless, surface rate measurement can be useful to constrain these properties and reduce the flow profiling uncertainty. Thus, the method proposed appears to be an interesting and complementary method for characterizing borehole flows and groundwater dynamics in fractured media such as for instance, monitoring the recharge dynamic