Por: Jorge Zafra, Principal Hydrologist Minera Las Bambas; Luis Suescún, Senior Hidrogeologist Modeler, y Robin Dufour, director general de DHI Perú.

Abstract

Open-pit mining is an important economic activity in several countries around the world, including Peru. However, this activity can have a significant impact on the environment and on groundwater quality, especially if appropriate local hydrogeological management and monitoring measures are not implemented. 

For this reason, hydrogeological studies are essential to understand the operational phase of open-pit mining and the potential impacts of its execution on surface and groundwater resources, as well as to design mitigation, optimization, and operational strategies for mining projects.

In this context, various methodologies and tools have been developed to model the hydrogeology of open-pit mining operations and to improve the accuracy and efficiency of the numerical models used to estimate these impacts and operational progress. 

Among these methodologies is inverse modeling, a key tool in hydrogeology that addresses uncertainty in numerical and conceptual groundwater models. However, modelers often perform and analyze steady-state and transient calibrations separately due to operational limitations in numerical codes and a lack of awareness of efficient work methodologies in mining projects. 

For this reason, new ways have been explored to optimize pre- and post-processing, as well as to integrate the execution of steady-state and transient models during the calibration phase, with the aim of saving modeling time and allowing for more time to focus on result evaluation, uncertainty analysis, and other critical tasks.

To address this challenge, a major open-pit copper mine at the Ferrobamba pit of the Las Bambas MU in Peru was analyzed, using a geological model in Leapfrog and a 3D numerical model in Feflow to represent the complex geology and hydrogeology of the Andean region in its pre-mining and current conditions. Based on the conceptual hydrogeological model, fieldwork and hydraulic testing made it possible to identify zones of reduced hydraulic conductivity at depth, as well as fractured and karstic zones, which resulted in the parameterization of more than 1,500 zonations. 

IFM programming methods in Python were used to minimize model run times for mining facilities, along with factorization techniques, pilot points, and multipliers, in a coupled steady-state and transient calibration process using PEST, aimed at reducing computational times and supporting the environmental impact assessment (EIA) requirements.