Abstract Pumping tests carried out in three highly tectonized Mediterranean areas, which share similar climate conditions, were used to characterize hard-rock aquifers and examine the issue of the sustainable yield of wells. On the basis of the identified drawdown trends and hydraulic parameters, pumping scenarios were simulated through different methods. The results show that the long-term trend in the drawdown and the initial thickness of the aquifer constrain the sustainable yield of a well. In the worst cases examined, namely those related to an aquifer with a barrier boundary and a delayed response of the aquifer, the sustainable well yield is 1–2 L s −1 . These well discharge values are significant when compared with those found in other regions of the world, and may be related to the dense network of the discontinuities which characterizes the sites.

Abstract The success of translating a conceptual site model for a karst site into a numerical groundwater model will depend on both the experience of the user and the capabilities and limitations of the selected computer program. Despite its numerous advantages, even MODFLOW – probably the most widely used, tested and verified modelling program currently available – has conceptual limitations that many karst hydrogeologists have to deal with on a routine basis while searching for an equivalent porous medium approach that may work. This includes assigning very high values of hydraulic conductivity to those model cells known, or suspected, to contain highly transmissive conduits, or assigning an unreasonable, very low effective porosity to the model cells with virtual ‘conduits’ to simulate high groundwater velocities. A new version of MODFLOW called MODFLOW-USG (UnStructured Grid) has been developed and released to the public domain. This new version retains full compatibility with previous versions of MODFLOW while taking advantage of unstructured grids and finite volume numerical solutions. It enables hydrogeologists to accurately translate even the most complex conceptual site models in karst into a numerical environment, thus eliminating the need for various surrogate modelling solutions based on an equivalent porous medium approach.

Heat transfer in the subsurface can vary with depth due to variations in the thermal conductivity of the geologic medium as well as variations in groundwater flow velocity. However, traditional thermal response tests (TRTs) do not allow for evaluation of the depth variability of heat transfer. We investigate the potential for using discrete-depth heat dissipation tests in open, water-filled boreholes to evaluate variations in heat exchange rate with depth. Heat dissipation tests were initiated at target depths in a test well using an electrical resistance heater. Heat dissipation was monitored by measuring borehole water temperature through time using a fiber-optic distributed temperature sensing system. Temperature data were used to compare the thermal response at different depths in the borehole. To account for both the thermal conductivity of the geologic medium and the groundwater flow velocity, we used a numerical groundwater flow model (MODFLOW) and solute transport model (MT3DMS) to simulate heat dissipation tests. Simulation results indicate the measured response to a heat dissipation test in an open borehole is strongly dependent on the measurement location within the borehole; thus, data are ambiguous when the measurement location is uncontrolled. However, modeling results also indicate that the thermal response of a heat dissipation test as measured at the center of the borehole is sensitive to variations in thermal conductivity and groundwater flow velocity, suggesting that heat dissipation tests are a potentially useful method for characterizing depth variability in thermal properties if a centralized temperature measurement method is used to monitor the tests.