Improved Model Upscaling of Overburden Hydraulic Conductivity for Input into Groundwater Models

The key objective for this project was to determine an upscaling method that bridges the gap of geotechnical model fracture conductivity at longwall panel detail, to groundwater model bulk conductivity input requirements. It has so far been a challenge across the geotechnical and hydrological disciplines to upscale these detailed fracture flow conductivity results to the groundwater models, whilst resulting in pore pressure monitoring and flows that are consistent with experience.

Collaboration with groundwater modelling consultants informed this project's understanding of groundwater model inputs and limitations. Groundwater modelers need to use mining induced overburden conductivity 3-5 orders of magnitude less than the modelled, measured and observed overburden conductivity.

This project concludes that upscaling is not the issue between the detailed rock failure vertical hydraulic conductivity and the required inputs into groundwater models. It proposes that desaturation is responsible for the significant reduction in relative conductivity and a method of estimating the relative conductivity for overburden groundwater model inputs is suggested.

Investigations into upscaling methodologies included local (mathematical averaging) and non-local (large, detailed flow models) averaging techniques. These upscaling methods provided changes to mining induced hydraulic conductivity of only about one order of magnitude. Resulting in overburden hydraulic conductivities higher than the required groundwater model inputs.

Additional modelled factors that may reduce overburden conductivity were also investigated. These geotechnical and geological factors include modelled background conductivity, bedding strength and weak lithologies. These geotechnical factors again only produced about one order of magnitude reduction in overburden hydraulic conductivity.

This project proposes that desaturation is responsible for the significant reduction in relative conductivity. There is a significant amount of research that highlights the reduction in relative conductivity with desaturation.

With the understanding that unsaturated groundwater models with current computing power is not practicable for timely and cost appropriate solutions to the client, a method for determining the site specific relative conductivity to use in groundwater models, based on the limiting conductivity layer at the top of the desaturated zone is provided.

The proposed approach on an actual mine site groundwater model to prove the hypothesis will be investigated in the next stage of this project.