During operation of tailings impoundments, water is lost through decanting, evaporation and seepage. Techniques have been developed to regulate or control each of these components. Collectively implemented, these techniques form the basis of tailings impoundment management. Regulation of the rate and level of decanting influences the size of a free water pond, and, thereby, the amount of evaporation and seepage, which are both area-dependent. Conversely, the encouragement of seepage and/or evaporation could also be used to reduce decanting requirements. Management of one of the components of a tailings impoundment’s water balance is reflected in changes in the other components
The mine and mill in question have facilities for processing zinc, smelting lead, and using the acid by-product for producing fertilizer. The waste impoundment receives effluent from all these operations, as well as mine drainage.
The waste impoundment covers an area of about 60 ha, and is divided into two main cells to receive tailings, mine drainage, and mill process water in the eastern two-thirds and gypsum in the western third. Further subdivision of the tailings area has been completed to provide storage for hydroxide sludge from water treatment processes.
The floodplain on which the impoundment has been constructed consists of two coarse gravel layers 6 m to 15 m thick, separated by a zone of fine sand, silt and clay also 6 m to 15 m thick. The prevailing water table varies from 1 m to 4 m below the floodplain surface.
The conditions for acid-generation and leaching exist in both the tailings and impoundment foundation soils in the study area.
The estimated seepage losses would be affected by small changes in decant level for the month. After assessing all data and evaluating material permeabilities it was apparent that the estimated seepage for the tailings pond was low, probably reflecting a small pond area and little impact from seepage.through dikes. For the gypsum pond, embankment seepage was not a serious concern.
To assess groundwater flow, parameters describing initial moisture content, porosity and permeability of the various materials in the impoundment areas were required. A laboratory testing programme was undertaken to determine the required parameters for the stratification of materials identified in borings completed during the field programme.
Both manual and computer methods were utilized to analyse seepage flow from the impoundment area. Recent developments in analytical techniques have permitted the modelling of transient flow conditions as produced by fluctuating water levels and increasing thickness of the transmitting media.
The manual calculations were based on McWhorter and Nelson’s model of a layered sequence of lower permeability soils overlying an initially unsaturated zone of higher permeability – a drain or sink. The model considers various stages of flow through the zone beneath the tailings with water being supplied from a source at the surface of the tailings. The stages of flow include initial advance of a wetting front through the unsaturated zone to ultimate saturation of the zone. Seepage quantities can be computed for each stage of flow based on pressure head and a composite permeability defined as the average permeability of the tailings zone.
Pollution Control Alternatives
A slurry trench or grout curtain would have to totally enclose the impoundment, to prevent seepage bypassing around the ends of the barrier, and penetrate to the clayey silt aquitard to be effective. Subsequent analyses showed that positive cutoffs would result in an increase in pressure gradient and greater seepage into the lower gravel aquifer, partially offsetting the benefits of the cutoff. In addition to not providing zero pollution seepage out of the system, this procedure would have been extremely expensive.
Wells would be an effective seep¬age barrier if closely spaced. However, in addition to contaminated seepage, the pumped wells would also intercept flow from the river and regional groundwater flow. Recycling the anticipated volume of pumped water to the tailings pond for subsequent treatment would cause further complications, as well as result in costs comparable to the positive cutoff. This system would require continuous and long-term power and maintenance costs.
As a result of the drilling and piezometric monitoring programme, and a theoretical analysis, it was apparent that seepage through the tailings backfill was almost vertical with little or no phreatic surface near the perimeter embankments, provided the pond was kept away from the coarse dike fill. Similarly, the quantity of total seepage from the ponds was almost directly proportional to the area of the pond inundated by free water, when dike seepage near the decant was discounted.
For the embankment materials in which seepage velocities are one to three orders of magnitude greater than the tailings and gypsum, the effect of seepage control would be discernible in a much shorter time. When it is considered that between 50 and 80 per cent of the past seepage losses from the impoundment areas was attributed to flow through the embankments, it is probable that seepage control by beach development and embankment liner programmes would be responsible for observable decreases in dissolved solids concentrations in the observation wells. However, because of the time lag associated with basal seepage through the ponds, the monitoring period would have to be continued for several more years to verify and quantify the effectiveness of the blanket and liner placement, and pond surface area control.