THREE DIMENSIONAL GEOLOGICAL MAPPING AND GROUNDWATER MODEL CALIBRATION FOR MANAGED AQUIFER RECHARGE (MAR) TECHNIQUES IN ARID AND SEMI-ARID ENVIRONMENTS

Intrinsic ecological fragility of drylands and increasing anthropic pressures, which often relate to the unsustainable management of groundwater resources, are leading factors in desertification processes. Natural processes alone could not be enough to regenerate those fragile environments from degradation, and it is necessary the corrective interference of local and international communities. In this perspective, the WADIS-MAR Project (www.wadismar.eu) has been funded by the European Community in the frame of the SWIM Programme (www.swim-sm.eu) and started implementation. The aim of the Project was to enable local communities to manage groundwater resources through the application of sustainable agricultural practices and the implementation of water harvesting techniques and MAR systems. The Biskra and Batna region (Algeria) and the Jeffara of Medenine (Tunisia) are the two intervention areas selected in the frame of the Project because characterised by overexploitation of groundwater resources and are highly exposed to desertification processes. The main objective of the present research was the development of a methodology for the hydrogeological characterisation of the Biskra and Batna region finalised to the study of sustainable systems for the management of groundwater resources that could be reply in similar context. The first phase of the research has focused on the collection of literature data, freely released satellite datasets, and field measures carried out by local partners. Collected data were verified, organised within a geodatabase, preprocessed and then imported into a geo-modeler software for the realisation of two 3-D hydrogeological models. The regional scale model extends over the South-western slope of the Aures mountains. 14 and 6 geological cross sections at 1:200.000 and 1:50.000 scale respectively, were realised perpendicular to the main geological structures and integrated with hydrogeological information to obtain the hydrogeological conceptual model of the Biskra and Batna region. From its analysis emerged that: 1) interference between two main fold systems generates dome and basins geometries constituting large storage areas; 2) groundwater recharge can occur in the Aures mountains where the hydrogeological units crop out; 3) a complex faults system influences the groundwater circulation allowing the communication among various aquifers and acting as preferred pathway for groundwater rises towards the surface. The local scale model was developed to define the geometries of the Infèro-Flux phreatic aquifer, near the city of Biskra. 19 geological cross sections were realised at 1:10.000 scale and perpendicular to the groundwater flow direction by integrating geological information into a geomorphological interpretation of the river system. Results show that the aquifer morphology is characterised by an asymmetric geometry of the confining bed and maximum deposit depth of 50 metres. The volume of the alluvium was also estimated and corresponds to 83 Mm3. The conceptual model of the Inféro-Flux aquifer was developed by integrating the hydrogeological and hydrological data into the local scale 3-D model and used to develop a groundwater flow numerical model to estimate the hydraulic conductivity parameter and to obtain quantitative information about the system dynamics. A first hydrogeological dataset was used for the automated calibration of the Kx parameter through the PEST. Resulting values range between 0.00019 and 0.00869 m/s and are coherent with braided stream deposits. A second hydrogeological dataset was used to validate the calibration results. Groundwater budget was compared to the hydrological balance. Results suggest that infiltration alone is not sufficient to allow the simulated inflow from the northern head boundary. Therefore, it is reasonable to consider a further source of alimentation for the Inféro-Flux aquifer. Low residual values confirm the consistency of the numerical model