Castendyk, D, Evans, E, Banton, D, Nutini, J & Young, A 2023, 'Closure modelling of the Eagle Ni-Cu mine, Michigan: Part 1. Hydrogeology and water quality of the underground mine pool ', in B Abbasi, J Parshley, A Fourie & M Tibbett (eds), Mine Closure 2023: Proceedings of the 16th International Conference on Mine Closure, Australian Centre for Geomechanics, Perth, https://doi.org/10.36487/ACG_repo/2315_057
(https://papers.acg.uwa.edu.au/p/2315_057_Castendyk/)
Abstract: The Eagle underground Ni-Cu mine on the Upper Peninsula of Michigan, owned by Eagle Mine LLC (Eagle), a subsidiary of Lundin Mining, began operations in 2014. Ore is trucked 80 miles to the Humboldt Mill (see Part 2) whereas waste rock is stockpiled at the surface and returned underground as cemented and uncemented rockfill. After mining concludes, one possible reclamation plan involves flooding the underground workings with deep, saline groundwater plus fresh water pumped from the overlying Quaternary aquifer and reclaiming the surface footprint as a greenfield property. The mine pool must not deteriorate water quality in the Quaternary aquifer or the nearest downgradient surface water receptor, the Salmon-Trout River, located 1.5 miles north of the mine. To assess closure costs, timeframe, and risk associated with this reclamation plan, Eagle developed a four-part closure study in 2021 that calculated the time to flood the mine pool, estimated the vertical components of the hydraulic gradient post-flooding, predicted the water quality of the mine pool, estimated the travel time to the Salmon-Trout River, and predicted the water quality in both the downgradient Quaternary aquifer and river. 
The multi-disciplinary modelling approach presented herein provides several insights for predicting the closure of underground mines: (1) Where steady-state vertical components of the hydraulic gradient are downward, groundwater is likely to flow downward from the shallow aquifer into the mine pool, but not from the mine pool into the shallow aquifer; the mine pool can be expected to have little to no influence on water quality in the overlying aquifer. (2) For mines with low bedrock hydraulic conductivity and downward hydraulic gradients, the addition of bulkheads between different mine workings at different depths may direct flow from the flooded mine workings towards the surface. (3) Underground mines with low bedrock hydraulic conductivity and low steady-state flowthrough rates (Q) may generate low mass loading (M) to downgradient aquifers, regardless of mine pool concentrations (C), where M=Q×C; a small load addition from a mine pool to a large aquifer may generate a steady-state water quality in the aquifer that is indiscernible from background water quality. (4) The representation of backfill pore space water in both water balance and geochemical models can influence geochemical predictions, especially where cemented backfill and weathered, uncemented waste rock are used. (5) The reaction time specified for the release of mass from wall rocks and backfill to mine pool water influences the predicted water chemistry of the mine pool. 

Keywords: FEFLOW, GoldSim, PHREEQC, bulkheads, case study