A probabilistic screening tool for long-term pit lake water balance estimation

Authors: Navarro-Valdivia, L; Burgess, R; Weber, P Download Paper Open access courtesy of:

DOI https://doi.org/10.36487/ACG_repo/2615_111

Cite As:
Navarro-Valdivia, L, Burgess, R & Weber, P 2026, 'A probabilistic screening tool for long-term pit lake water balance estimation', in AB Fourie, G Boggs, J Heyes & M Tibbett (eds), Mine Closure 2026: Proceedings of the 19th International Conference on Mine Closure, Australian Centre for Geomechanics, Perth, pp. 1-8, https://doi.org/10.36487/ACG_repo/2615_111



Abstract:
Anticipating the filling behaviour of pit lakes is essential for effective mine closure planning and environmental management. Although sophisticated deterministic and probabilistic models are available (e.g. GoldSim), the industry also requires practical, high-level screening tools for the rapid assessment of long-term risks associated with pit lakes. To address this gap, a probabilistic water balance tool was developed to facilitate efficient preliminary pit lake assessments. This approach allows environmental and closure professionals to conduct robust preliminary assessments independently, thereby reducing the reliance on external resources. The underlying computational engine calculates net volume changes by integrating direct precipitation, evaporation, catchment run-off, groundwater exchange and active pumping. The model employs a fourth-order Runge-Kutta (RK4) numerical method to handle stage-volume-area relationships. Optimised for rapid screening rather than granular simulation, the tool operates on a monthly timestep and assumes a repeating annual climate cycle, thereby optimising computational demands and simplifying user inputs. To address the uncertainties inherent in long-term forecasting, a Monte Carlo simulation framework is integrated into the engine. By assigning statistical variances to key hydrological-hydraulic parameters, the model executes numerous iterations to produce probabilistic filling trajectories. This methodology captures statistical confidence intervals from the 5th to the 95th percentiles for both lake elevations and individual flux volumes. While the tool is designed for future cloud-based deployment via a web interface to provide cross-platform accessibility, it is currently implemented as an open-source repository for local execution. This approach ensures immediate usability for environmental professionals while providing a foundation for automated, web-hosted reporting.

Keywords: mine closure, pit lake, water balance, Monte Carlo simulation, Python

References: