A new mining site situated near Kolwezi in the Democratic Republic of the Congo plans to develop a pit in phases over a period of six years. The mine requires dewatering volume estimates of the pit as well as a constant water supply to the plant. Hydrogeologic data available at the site during the scoping phase was limited to a few water level measurements and blowout yields from only five hydrological boreholes. Hydraulic properties from reports at neighbouring sites were extrapolated to the geological units at the site. The depth to water level at the site is about 20 m, with a planned final pit depth of approximately 180 m below surface.
Based on the limited data available an analytical approach to estimate the inflow into the mine was adopted. Analytical calculations proposed by Marinelli and Niccoli (2000) were used to estimate the inflow into the Pumpi mine pits. The analytical calculations consider recharge, depth of mining vertical and horizontal hydraulic conductivities. Drawdown evolution of pit dewatering are obtained by using different mining depths at different mine stages. The output results from the analytical calculations are the maximum extent of influence of the pit as well as the volume of water inflow into the pit. Limitations of the analytical equations are that they, amongst others, cannot consider complex boundaries.
Drilling and pump testing to obtain local hydraulic properties and boundary conditions are planned during the first quarter of 2013. The numerical model will be set up after the drilling and pumping tests, using the new data for calibration. The numerical model will contain as much of the physical layer definitions and potential internal boundaries as possible with model boundaries incorporated along far field fault zones and hydraulic boundaries. The numerical model should improve the reliability of estimates of pit inflow and water supply to the plant.
The results between the analytical and numerical approaches can then be compared to improve future dewatering estimates with limited data. It is expected that the reliability of the analytical predictions will reduce after year 4, where the role of boundaries are expected to influence the drawdowns and related flow towards the pit.