Water resources evaluation in the danakil basil, Ethiopia: groundwater supply for potash solution mining

English: This dissertation investigates three alluvial fans in the Northern Danakil Depression, Ethiopia for potential groundwater supply to a solution mining operation. The presence of groundwater seeps at the toe of the alluvial fans further indicated the potential of the alluvial fans for groundwater exploration and possible abstraction to meet the mine water demand of 30 cubic hectometres (30 hm3) or 30-million cubic meters per annum. Cumulative life of mine demand was determined to be 450.73 hm3. Lithological, structural and hydrostratigraphic investigations revealed that the Precambrian basement rocks of the Dogua Mountains outcrop in the west adjacent to the Bussaba and North Dogua fans, while further south, Cenozoic-Jurassic rocks of the Antalo Limestones and Adigrat sandstones are juxtaposed against the Asabuya fan. Structural mapping showed the presence of north-south trending rift related faults, including the Main Danakil Rift-boundary Fault, which transect and underlie the alluvial fans. These faults were determined to be likely pathways for deep hydrothermal fluids to vertically recharge the fans with hot, hypersaline groundwater and brines. Additionally, east-west trending fault and fracture zones identified in the Cenozoic-Jurassic rocks implied that fresh-brackish water is likely transmitted from the west into the Asabuya fan, but no such through flow is expected into the Bussaba and North Dogua fans due to the Precambrian basement rock acting as an aquitard. This was determined to cause a considerable difference in recharge sources and hydrochemical signatures between Asabuya fan in the south and Bussaba and North Dogua fans in the north. Two automatic weather stations, were established in recharge zones of the Dogua Mountains. The annual average rainfall was determined to be 90 mm/a with an average evaporation rate of 66 mm/a. Moving Loop Electromagnetic geophysical surveys showed low conductivity zones extending through each fan from south to north, confirming the presence of mapped faults transecting the alluvial fans. A maximum fan thickness of 160 m was determined to occur toward the centre of each fan. The low conductivity fault zones and active depositional channels informed the final positions for production and monitoring boreholes. Drilling results of three production boreholes and five monitoring boreholes confirmed the unconfined nature of the aquifer. Drilling depths proved a minimum thickness of 170 m. Analysis of time-drawdown data yielded high Transmissivity values in the order of 11 000-20 000 m2/d. Due to a lack of observable drawdown in the monitoring holes, storage estimations were inconclusive. Water level response during pumping increased confidence in the assumption of recharge by hydrothermal water upwelling via rift related faults. An Aquifer Stress Test (AST) showed minimal drawdown and rapid recovery proving that pump rates of up to 60 l/s were sustainable for up to two years. A total of 603 370.7 m3 was abstracted during the AST. No cross-fan interaction was observed. Hydrochemical analyses indicated boreholes nearer the wadis at the head of the fans had improved water quality compared to boreholes targeting the MDRF. Water quality in the Asabuya fan deteriorated as pumping continued, while the quality of the Bussaba and North Dogua fans both improved, implying different recharge sources to specific fans. Insights gained from works carried out were used to inform a regional conceptual and numerical model. The numerical model assessed the influence of regional boundary conditions on the groundwater flow system as well as the feasibility of three different recharge scenarios to the alluvial fan aquifers. Calibration results indicated that only scenario 3 was feasible to use for predictive simulations. 20 wells spread across the fans were simulated to pump 22 hours/day at 60 l/s for 20 years and further for 40 years. Results showed an average drawdown of 5.5 m in the alluvial fans and that the mine water demand could be met in terms of quantity. Quality considerations however, were not incorporated into the model results and outcomes. Resource estimation determined that with a Sy value of 0.2, the life of mine demand could be met for only 18 years at which time 96% of water in storage, without recharge contributions, would be depleted from the aquifer.