La série de conférences en hydrogéologie

ABSTRACT

Karst develops through the dissolution of carbonate rock. Karst groundwater in Europe is a major source of fresh water contributing up to half of the total drinking water supply in some countries. Climate model projections suggest that in the next 100 years, European karst regions will experience a strong increase in temperature and a serious decrease of precipitation especially in the Mediterranean region. To be prepared, policy-makers need quantitative and reliable estimates of potential changes to karst water resources. This study presents an attempt to quantify karst water resources over the whole Mediterranean. This is done by quantifying largescale karst recharge with a newly developed hydrologic model that considers the strong heterogeneities of recharge processes that evolve from carbonate rock dissolution. The model is driven by long-term gridded data from the European Climate Assessment & Dataset project (ECA&D). Soil moisture and evapotranspiration measurements that are available across Europe’s carbonate rock regions are used for model evaluation. To assess the climatic sensitivity of recharge, we define the recharge elasticity as ratio of normalized changes of annual recharge to precipitation between years. Using large-scale simulations of the recharge model we can explore the spatial and temporal variability of the recharge elasticity among the Mediterranean’s karst regions and understand the impact climatic change.

ABSTRACT

As the world's largest accessible source of freshwater, groundwater plays a vital role in satisfying the basic needs of human society. It serves as a primary source of drinking water and supplies water for agricultural and industrial activities. During times of drought, groundwater storage provides a large natural buffer against water shortage and sustains flows to rivers and wetlands, supporting ecosystem habitats and biodiversity. Yet, the current generation of global scale hydrological models (GHMs) do not include a groundwater flow component, although it is a crucial part of the hydrological cycle. Thus, a realistic physical representation of the groundwater system that allows for the simulation of groundwater head dynamics and lateral flows is essential for GHMs that increasingly run at finer resolution. In this study we present a global groundwater model with a resolution of 5 arc-minutes using MODFLOW. With this global groundwater model we eventually intend to simulate the changes in the groundwater system over time that result from variations in recharge and abstraction. Aquifer schematization and properties of this groundwater model were developed from available global lithological maps and datasets, combined with our estimate of aquifer thickness for sedimentary basins.

We forced the groundwater model with the output from the global hydrological model PCR-GLOBWB, specifically the net groundwater recharge and average surface water levels derived from routed channel discharge. For the parameterization, we relied entirely on available global datasets and did not calibrate the model so that it can equally be expanded to data poor environments. Based on our sensitivity analysis, in which we run the model with various hydrogeological parameter settings, we observed that most variance in groundwater depth is explained by variation in saturated conductivity, and, for the sediment basins, also by variation in recharge. The method is suitable to build a global groundwater model using best available global information, and estimated water table depths are within acceptable accuracy in many parts of the world.

ABSTRACT

Spatial variations in hydraulic conductivity (K) provide critical controls on solute transport in the subsurface. Direct-push tools have been developed for high-resolution characterization of K variations in unconsolidated settings. The direct-push injection

logger (DPIL) provides an indicator (flux/pressure, Q/P) of relative variations in K at a very high vertical resolution. The more recently developed HRK (High-Resolution K) tool combines the DPIL with a direct-push permeameter (DPP), providing a means to transform the DPIL Q/P profiles into high-resolution K profiles. These tools were applied to obtain 58 K profiles with a vertical sample spacing of 1.5 cm from the heavily studied macrodispersion experiment (MADE) site. We have compared the data from these 58 profiles with those from the 67 flowmeter profiles that have served as the primary basis for characterizing the heterogeneous aquifer at the site. Overall, the patterns of variation displayed by the two data sets are quite similar, in terms of both large-scale structure and autocorrelation characteristics, although the two datasets exhibit distinct differences in geometric mean K and lnK variance.

The DPIL calibration approach used in the foregoing work has some drawbacks. There is an upper limit on accurate Q/P values due to a lower limit on measureable DPIL pressure responses in high-K zones. Our previous DPIL calibration approach assumed a linear relationship between lnK and ln(Q/P); consequently, the Q/P limit was translated into an upper threshold on accurate DPIL-based K estimates. Current work aims to improve the calibration by incorporating an adaptive, nonlinear ln(Q/P) vs. lnK

relationship into the inversion process, allowing the estimated K profiles to more accurately reflect the very high K’s indicated by some of the DPP tests.