Mine Voids Research Articles

Block geometry and rockmass characterization for prediction of dilution potential into sub-level cave mine voids

When sub-level caving is used as an underground mining method for the exploitation of diamondiferous kimberlite pipes, a deep, steep-sided pipe is created that is susceptible to sloughage. Pipe wall sloughage allows waste rock to mix with and dilute the caved kimberlite ore. Using the Koala Kimberlite Pipe at Ekati Diamond Mine as a case study, a method has been developed, using rockmass and rock structure characterization integrated with discrete block (DEM) simulation, for comparative assessment of dilution potential and the classification and mapping of dilution source zones around the kimberlite pipe. The assessment process involves the development of a geomechanical model to delineate the structural fabric within the host rockmass and characterize the size and shape distributions of discrete rock blocks defined by intersecting discontinuities. Combining the slope geometry with the structural geology model allows for a zone-by-zone numerical analysis of pipe wall stability under gravitational loading. A final dilution map is developed in order to target problem areas where the highest levels of slope instability are anticipated.

Two modelling approaches to water-quality simulation in a flooded iron-ore mine (Saizerais, Lorraine, France): A semi-distributed chemical reactor model and a physically based distributed reactive transport pipe network model

The flooding of abandoned mines in the Lorraine Iron Basin (LIB) over the past 25 years has degraded the quality of the groundwater tapped for drinking water. High concentrations of dissolved sulphate have made the water unsuitable for human consumption. This problematic issue has led to the development of numerical tools to support water-resource management in mining contexts. Here we examine two modelling approaches using different numerical tools that we tested on the Saizerais flooded iron-ore mine (Lorraine, France). A first approach considers the Saizerais Mine as a network of two chemical reactors (NCR). The second approach is based on a physically distributed pipe network model (PNM) built with EPANET 2 software. This approach considers the mine as a network of pipes defined by their geometric and chemical parameters. Each reactor in the NCR model includes a detailed chemical model built to simulate quality evolution in the flooded mine water. However, in order to obtain a robust PNM, we simplified the detailed chemical model into a specific sulphate dissolution–precipitation model that is included as sulphate source/sink in both a NCR model and a pipe network model. Both the NCR model and the PNM, based on different numerical techniques, give good post-calibration agreement between the simulated and measured sulphate concentrations in the drinking-water well and overflow drift. The NCR model incorporating the detailed chemical model is useful when a detailed chemical behaviour at the overflow is needed. The PNM incorporating the simplified sulphate dissolution–precipitation model provides better information of the physics controlling the effect of flow and low flow zones, and the time of solid sulphate removal whereas the NCR model will underestimate clean-up time due to the complete mixing assumption. In conclusion, the detailed NCR model will give a first assessment of chemical processes at overflow, and in a second time, the PNM model will provide more detailed information on flow and chemical behaviour (dissolved sulphate concentrations, remaining mass of solid sulphate) in the network. Nevertheless, both modelling methods require hydrological and chemical parameters (recharge flow rate, outflows, volume of mine voids, mass of solids, kinetic constants of the dissolution–precipitation reactions), which are commonly not available for a mine and therefore call for calibration data.

Passive neutralisation of acid mine drainage by fly ash and its derivatives: A column leaching study

Active treatment of acid mine drainage using fly ash (FA) generates solid residues. This work investigated the suitability of using the solid residues (SR) generated from the process for passive treatment of acid mine drainage in a backfill set-up. A column leaching study of the solid residues blended with varying amounts of fly ash (5%, 25%, 40%) and 6% ordinary portland cement (OPC) and fresh fly ash was carried out to assess the neutralisation and contaminants attenuation in AMD with time. All the solid cores acidified in a stepwise fashion, exhibiting three buffer zones (12–10, 9–7.5 and 6–4). Dissolution of CaO in fly ash and solid residues blended with fly ash impacted high pH in the initial leachates while hydrations reactions in OPC blend solid residues contributed to high initial pH. Dissolution of SiO 2(amorphous) and mullite in the solid residues contributed to sustain buffering at pH 7–9.5 for SR and FA blended solid residues. Encapsulation of the solid residue particles by the calcium silicate hydrate gels (CSH) in OPC blended solid residues reduced interaction of particles with SAMD hence the buffering at pH 7–9.5 was not observed. Eventual clean-up of the AMD for all solid cores was dependent on the increase of the pH to alkaline and circum-neutral levels and sustainability of the buffering capacity at circum-neutral pH. It is hypothesized that the duration of the effectiveness of generation of alkalinity and AMD clean-up by the solid residues will greatly depend on the quality of the AMD percolating through. The SR and SR + 6% OPC were observed to be superior in removal of Al, Fe and SO 4 2 - compared to FA and SR + 25% FA. Lowest efficiency was observed in FA and SR + 25% FA cores for all contaminants except for Mn in FA core. This study shows that the solid residues (SR) can successively generate alkalinity over an extended period of time and passively treat AMD. This has positive implications upon the application of the solid residues for backfilling of mine voids. Different solid cores exhibited varying degrees of leachability for various toxic elements. B was the most leached element and the leaching was more pronounced in the SR and SR + FA blends. The SR + OPC core was observed to be superior in retention of Mo and B. Calculation of % toxic trace elements leached based on total composition from XRF data indicated insignificant amounts are leached over the experimental period for SR core. However the pH regime generated will depend on backfill scenario which in-turn will dictate the extent of leaching of toxic trace elements. It is recommended that more studies be carried out on the leachability of the toxic trace elements on a long-term basis before suitability of the SR and SR + FA blends can be fully evaluated.

Predicting Rebound in a Deep Colliery in South Africa

The main challenge facing many of the coal mines in South Africa is the management of mine water following the closure of mining operations. The Sigma Colliery is situated in the Free State Province, adjacent to the town of Sasolburg and bordering the Vaal River, one of the country's largest rivers. The mining includes both opencast and underground operations; however, this paper will only discuss the main underground operations. There are several aquifer systems overlying the deeper mining, which was done by bord and pillar and high extraction mining. Detailed conceptual models of the interactions between several aquifer systems and the rebounding mine voids were constructed using mining and monitoring data. From this, numerical flow models were used to model the complex flow system where rebound of water levels is expected. The results have led to an accurate understanding of the complex flow system and the important controls on the final water levels in the area.

Imaging mining hazards within coalbeds using prestack wave equation migration of in-seam seismic survey data: A feasibility study with synthetic data

Gas and water accumulations in old mine workings, voids and fault zones are threats to safety in coal mining. There are a variety of measures that can be taken to extract fluids prior to, or during mining; therefore, the ability to locate geological structures and old abandoned working and voids which may not have been mapped accurately are the key to overall success. However, current techniques are of limited use because of either high cost or low resolution. We simulate and evaluate use of high-frequency seismic data acquired in an in-seam geometry through the use of synthetic two-component elastic data for 2-D models. Elastic common-source seismic data collected at the mining coalface contain body waves reflected at coalbed interfaces such as faults and old workings. Reverse-time prestack elastic migration is used to image the interfaces. Numerical tests on synthetic data indicate that this approach is expected to be effective to detect abandoned mining workings containing different fluids (such as water or gas/air) and faults. However, the use of this approach is restricted to the nearest in-seam interface from the mine face. The three items that are new in this paper are (1) the application of wave equation migration to in-seam data, (2) use of a multiplicative (data derived) mask to enhance the migrated image, and (3) the demonstrated potential to distinguish fluids in mine voids from seismic data.

Resistivity Level Runs to Detect Water-filled Mine Voids between Drill Holes

Resistivity level runs are collected by lowering a current source down one well and measuring the resulting voltage at the same depth in another well. Mine voids between the wells that contain acid water appear as conductive anomalies on the resulting apparent resistivity profiles. Resistivity level runs can be collected rapidly and without lowering expensive equipment down holes of unknown stability. The data can be interpreted on-site, and are relatively insensitive to positioning errors. The method is well suited to sites where several drill holes have failed to intersect a known mine void. We demonstrated the feasibility of resistivity level run profiling at an abandoned mine complex in central Pennsylvania, where resistivity level runs were successfully used to locate haulage ways containing mine water.

Geotechnical characterization of fly ash composites for backfilling mine voids

Backfilling of mine voids is mandatory to avoid subsequent ground stability problems in the form of subsidence. River sand and mill tailings have been widely used since a long time as backfilling materials. However, with a strict regulation banning river sand mining in India, research for developing alternative engineering materials substituting sand has gained importance. In the present study four fly ash composite materials (FCMs) was developed from the fly ash obtained from a captive thermal unit of Rourkela Steel Plant (RSP). The main constituent of the composite were fly ash, lime and gypsum. Detailed physical, and engineering properties were determined for the FCMs. Significant increases in the compressive strength were obtained after 56 days of curing time. A detailed SEM studies was undertaken to account for the increase in strength with time. The fly ash composite developed from RSP has potential to be used as substitute to sand for backfilling the mine voids.

English

Since its creation, the European Union (EU) has developed a substantial amount of environmental legislation in the form of directives, regulations and decisions which are applicable to all the Member States. However, mining policy in the EU has been driven mainly by industrial considerations: the energy extractive industry being covered by the Directorate General Transport and Energy (DG TREN) and non-energy by DG Enterprise. For historical and economic reasons mining had been specifically excluded from much of the environmental policy developed by DG Environment. In the aftermath of the Baia Mare accident, the European Commission created a Task Force that proposed a plan of action with ideas for new legislation. They included the amendment of the Seveso II Directive and the production a document on Best Available Techniques (BAT) for the environmental control of mineral processing, similar to those produced under the Integrated Pollution Prevention and Control (IPPC) Directive. The main policy development has been a new Directive on the management of waste from the extractive industry, currently under discussion and expected to be approved in 2005. This article reviews the coverage of mine water in European legislation before and after these recent legislative changes from the point of view of the Water Framework Directive. Six critical issues are identified: the need to cover energy and non-energy industries, full life-cycle approach, mine water in the BAT document, mine voids in the mine waste initiative, abandoned mines and the whole catchment approach. The conclusion is that although the new policies constitute a substantial improvement in the regulatory framework of the EU, they address only some of the key issues. In particular, the full life-cycle and catchment approaches have not been sufficiently taken into account by legislators.

English

A geochemical study was completed as part of an Environmental Assessment Report for De Beers Canada Inc. Snap Lake diamond mine located in the Canadian Northwest Territories. The deposit will be mined using underground methods and consists of a diamond bearing planar kimberlite dyke dipping at about 15 degrees, hosted in metavolcanic and granite rock. Processing will include grinding of the kimberlite ore and a gravity separation circuit on this ore which will result in processed kimberlite waste including a coarse fraction and a finer fraction which will be used to produce a cemented paste backfill material. The cemented paste composed of processed kimberlite, dilution rock, and cement is thickened to a paste-like consistency and will be used to fill the mine voids. The study discussed in this paper was completed to evaluate potential leachate chemistry of the alkaline cemented paste backfill material slated for use underground. The laboratory test program included solids testing, short-term leach testing, and longer-term kinetic testing on a variety of cemented paste blends and additives. Results show that the pH of the leachate remained alkaline for subaqueous test cells and decreased over time for sub-aerial test cells. Geochemical speciation modeling using PHREEQC that was completed to further evaluate leachate chemistry for site specific conditions indicated probable solubility constraints due to secondary mineral precipitation for parameters calcium, chromium, manganese and nickel.

Sulphate generation in South African underground and opencast collieries

Deterioration in the water quality results from coal mining, and the significant impact on the environment in South Africa has been reported (Funke in SA Water, Sci Technol 15:115–144; 1983). Pyrite in with South African coal can be oxidised to yield acidic and high salinity water. This study deals with the rate at which sulphate could be produced for collieries under different conditions. Four collieries in the Witbank coalfield were selected for this investigation. The boreholes were investigated with an in-situ multi-parameter probe, and then sampled at different times with a specific depth sampler to determine the chemical composition of the water in the collieries. Water level measurements and volumetric calculations were done to determine influx rates. Daily sulphate generation rate for underground mining was found to be in the range of 0.4–2.7 kg/ha, depending on the availability of oxygen and the residence time of the water (the time available for neutralising agents present to react) in the mining void. The exclusion of oxygen due to flooding and the presence of calcareous carbonates influences the observed sulphate generation.

Contaminant dispersion at the rehabilitated Mary Kathleen uranium mine, Australia

This study reports on the transfer of contaminants from waste rock dumps and mineralised ground into soils, sediments, waters and plants at the rehabilitated Mary Kathleen uranium mine in semi-arid northwest Queensland. Numerous waste rock dumps were partly covered with benign soil and the open pit mine was allowed to flood. The mineralised and waste calc-silicate rock in the open pit and dumps has major (>1 wt%) Ca, Fe and Mg, minor (>1,000 ppm) Ce, La, Mn, P and S, subminor (>100 ppm) Ba, Cu, Th and U, and trace (<100 ppm) As, Ni, Pb, Y and Zn values. Consequently, chemical and physical weathering processes have acted on waste rock and on rock faces within the open pit, mobilising many elements and leading to their dispersion into soils, stream sediments, pit water and several plant species. Chemical dispersion is initiated by sulfide mineral breakdown, generation of sulfuric acid and formation of several soluble, transient sulfate minerals as evaporative efflorescent precipitates. Radiation doses associated with the open pit average 5.65 mSv year−1; waste dumps commonly have lower values, especially where soil-covered. Surface pit water is slightly acid, with high sulfate values accompanied by levels of U, Cu and Ni close to or above Australian water guideline values for livestock. Dispersion of U and related elements into soils and stream sediments occurs by physical (erosional) processes and from chemical precipitation. Plants growing in the mine void, on waste dumps and contaminated soil display evidence of biological uptake of U, LREE, Cu and Th and to a lesser degree of As, Ni, Pb, Y and Zn, with values being up to 1–2 orders of magnitude above background sites for the same species. Although rehabilitation procedures have been partly successful in reducing dispersion of U and related elements into the surrounding environment, it is apparent that 20 years after rehabilitation, there is significant physical and chemical mobility, including transfer into plants.

Hydrologic and Geochemical Factors Governing Chemical Evolution of Discharges from an Abandoned, Flooded, Underground Coal Mine Network

Discharges from some underground flooded coal mines have exhibited increases in pH and reductions in contaminant loadings with time. Data from a study of mine water quality evolution in interconnected, flooded mines of the Uniontown syncline, Southwestern Pennsylvania were evaluated with the aid of modeling to elucidate the hydrologic and geochemical factors responsible for such changes. Coal barriers left in place from mining operations define three hydraulically distinct but interconnected zones: the southern, central, and northern pools. Assuming each mine pool to behave as a completely mixed tank reactor, a steady-state, tanks-in-series model was developed to describe system hydraulics. Chemical modeling components were coupled with the tank reactor hydraulic model to simulate inputs to the mine voids, acid generation from pyrite dissolution, and discharge water quality. Empirical in-mine chemical production terms were estimated for each of the mine pools based on discharge data from 1974 to 1975 and ...

Turbulent layering beneath the pycnocline in a strongly stratified pit lake

Temperature microstructure observations from a pit lake show alternate layers of high and low turbulence beneath a very strong pycnocline. The layers were clearly defined, around 0.3‐0.5 m thick, and were found both at the edge and in the center of the pit lake. In the most energetic regions, the turbulent energy dissipation rate reached «5 10 26 m 2 s 23 and the dissipation rate of thermal variance almost reached xu 5 10 24 8 C 2 s 21 . The turbulent fluid in the middle of the lake had a higher dissipation ratio, the thermal component of potential energy gain vs. kinetic energy loss, when compared with the sidewalls. This is attributed to different forcing mechanisms for the turbulence. We hypothesize that basin-scale internal waves, in conjunction with the topography, create sidewall boundary layers. In the interior regions of the pit lake, the turbulent layers are probably generated by instability due to focusing of short-wavelength internal waves. A pit lake is a water body that forms in a mining void on the cessation of mining activity. Usually a lake only forms in regions where the accumulation of groundwater inflow and rainfall exceed evaporation. Alternately, the pit lake can be actively filled as an aid to remediation or for storage. Some of the dissolved material that finds its way into a pit

English

More than two hundred homes in Hazleton, Pa. were found to have benzene vapors (a known carcinogen) from one of the largest underground gasoline storage tank spills in PA history. Remediating this site became the number one priority of USEPA Region III (EPA). A relic underground coal mine situated along a 3000 ft long plunging syncline beneath the Laurel Gardens neighborhood complicated site geology. The U.S. Army Corps of Engineers (USACE) determined if air filled mine voids existed and contained gasoline and gasoline vapors. The plan called for four study phases: Determining the mine location, Visualizing the geology and water table, Modeling mine pool effects on plume behaviors, and Field sampling to verify study results. Mine Location: GIS was used to georeference and compare a 19 th century mine map of unknown datum and coordinate system with old aerial photos, boreholes data and microgravity survey results. Extremely accurate location of the mine was critical for locating potential gas filled voids. Visualization: A three dimensional geologic model was constructed from the GIS data and available site data depicting the relationship between site geology, the mine void, the mine pool elevation and the surrounding groundwater table. This predicted air filled mine voids in the upper areas of the syncline. Modeling: A finite element groundwater model (FEMWATER) was developed to study mine pool effects on gasoline plume movements. Two different approaches were compared for modeling the mine pool. In one approach the mine void was modeled as high conductivity aquifer. In the adopted approach, it was modeled as an interior boundary with assigned heads. The model showed dissolved gasoline contaminants migrating to the mine. Field Sampling: Using the gravity data and 3 dimensional mine model, several target areas for drilling were identified. Gas vapor explosion was a concern during drilling. The drilling plan incorporating the use of on-demand nitrogen gas connected to a drill rig. A gas filled mine void was encountered and an aqueous sample confirmed gasoline contamination in the mine.

Long-Term Behavior of Fixated Flue Gas Desulfurization Material Grout in Mine Drainage Environments

In this research, we examine the long-term ( ;4 years! behavior of fixated flue gas desulfurization ~FGD! material grout following placement within the Roberts-Dawson underground coal mine. Surface water and groundwater samples were collected to examine the impact of grouting on water quality, and core samples were obtained to assess the geochemical stability of the grout material. Surface water samples collected from the main seep at the Roberts-Dawson mine indicated that 4 years after grout placement the long-term fluxes of acidity, iron, sulfur, and calcium were slightly elevated compared to pregrout conditions. The long-term discharge of these constituents was likely due to continued dissolution of grout material ~for Ca and S! as well as changes in flow paths and subsequent solubilization of metal salts accumulated within the mine voids ~for acidity, Fe, Al, and S!. Although the fluxes of these elements were elevated, no measurable deleterious impact was observed for the underlying groundwater or adjacent surface water reservoir. Groundwater samples collected from monitoring wells installed within the grout material indicated that acid mine drainage waters were neutralized by the grout material. Mineralogical analyses demonstrated minimal penetration of mine drainage water into the high strength fixated FGD material grout, and little weathering of the material was observed. These data indicate that the high strength fixated FGD material grout injected into the Roberts-Dawson mine was geochemically stable and could locally neutralize mine drainage waters. However, more complete grouting and more extensive mine flooding is likely needed in order to bring about significant improvements in seep water quality.

REMOTE SENSING, VISUALIZATION AND MODELING OF THE DIAMOND COAL MINE IN HAZLETON, PA

Abstract: More than two hundred homes in Hazleton, Pa. were found to have benzene vapors (a known carcinogen) from one of the largest underground gasoline storage tank spills in PA history. Remediating this site became the number one priority of USEPA Region III (EPA). A relic underground coal mine situated along a 3000 ft long plunging syncline beneath the Laurel Gardens neighborhood complicated site geology. The U.S. Army Corps of Engineers (USACE) determined if air filled mine voids existed and contained gasoline and gasoline vapors. The plan called for four study phases: Determining the mine location, Visualizing the geology and water table, Modeling mine pool effects on plume behaviors, and Field sampling to verify study results. Mine Location: GIS was used to georeference and compare a 19th century mine map of unknown datum and coordinate system with old aerial photos, boreholes data and microgravity survey results. Extremely accurate location of the mine was critical for locating potential gas filled voids. Visualization: A three dimensional geologic model was constructed from the GIS data and available site data depicting the relationship between site geology, the mine void, the mine pool elevation and the surrounding groundwater table. This predicted air filled mine voids in the upper areas of the syncline. Modeling: A finite element groundwater model (FEMWATER) was developed to study mine pool effects on gasoline plume movements. Two different approaches were compared for modeling the mine pool. In one approach the mine void was modeled as high conductivity aquifer. In the adopted approach, it was modeled as an interior boundary with assigned heads. The model showed dissolved gasoline contaminants migrating to the mine. Field Sampling: Using the gravity data and 3 dimensional mine model, several target areas for drilling were identified. Gas vapor explosion was a concern during drilling. The drilling plan incorporating the use of on-demand nitrogen gas connected to a drill rig. A gas filled mine void was encountered and an aqueous sample confirmed gasoline contamination in the mine.

Environmental impacts of coal mining and associated wastes: a geochemical perspective

Abstract Ever since the commencement of industrial-scale coal mining (in northeast England around 1600), substantial environmental impacts have been recorded as arising from both the mined voids and from the wastes left behind at the surface. In the early days of coal mining, complaints about such impacts were strident, as the newly established industry adversely affected long-established agricultural interests. When the coal trade had come to dominate regional economies in mining districts, its negative impacts came to be accepted as a necessary byproduct of the generation of coal-based wealth. It has only been since large-scale mine closures began to take place in the major coal-mining economies of the developed world during the last few decades that the negative impacts of coal mining have once more been deemed unacceptable. The environmental impacts arising from coal mining activities are fundamentally attributable to the exposure of reduced earth materials (especially coal, pyrite, siderite, and ankerite) to the oxidizing power of the Earth’s atmosphere. The consequences range from the spontaneous combustion of coal to the release of acidic waters from pyrite oxidation. A typology of the known impacts arising from mine voids and wastes in coal mining districts has been developed, which recognizes many subcategories of impacts under five major headings: air pollution, fire hazards, ground deformation, water pollution, and water resource depletion. A robust understanding of geochemical processes is key to understanding how these impacts arise, and to developing sustainable mitigation strategies. The application of the newly developed typology is illustrated using the case of the Shilbottle Coalfield (Northumberland, UK). Although few demonstrable impacts have arisen in the categories of air pollution, fire hazards, or ground deformation, major problems of water pollution have required both preventative and remedial interventions. For the flooded underground voids, these took the form of a pump-and-treat system, whereas emissions of leachates from surface spoil heaps have necessitated the installation of an innovative ‘hybrid’ passive treatment system, comprising a permeable reactive barrier, oxidation ponds, and a wetland. Inverse geochemical modelling has clarified the linkages between the various types of water encountered in the coalfield, providing a baseline geochemical understanding upon which future investigations of remedial system sustainability can be based.

Exemple d’impacts de l’arrêt de l’exploitation minière et de l’exhaure dans le bassin houiller de Faulquemont (Lorraine)

A rock-mass mined by underground exploitation can be assimilate as an aquifer reservoir which is not delimited by physical boundaries unlike a geological reservoir. In fact, this ground-water reservoir includes the residual mining voids and mining works (drifts, shafts, galleries), natural porosity (interstitial porosity and of the natural discontinuities) of the fractured rock-mass, including the fracturation induced by mining. Firstly, this communication present the results obtained with two different methods used to evaluate the total volume of the residual mining voids and compare these results with the volume of water flooding calculated from the curve of the underground water level rise measured in the Faulquemont coal basin (Lorraine). The second part of this communication explains the effects of water level rise on the stability of the surface. Indeed, the water level recovery, when pumping is stopped, can induce a rise of ground level. The studies concerning mining respectively in The Netherlands and in Germany highlighted this phenomenon. The observations made also in the district of Faulquemont, where the works flooding has been stabilised, give similar results. No consequence on buildings and surface infrastructures due to this phenomenon has yet been noticed.

03/02134 Integrated hydraulic-hydrogeochemical assessment of flooded deep mine voids by test pumping at the Deerplay (Lancashire) and Frances (Fife) Collieries: Nuttall, C. A. et al. Geological Society Special Publication, 2002, 198, 315–326

03/02134 Integrated hydraulic-hydrogeochemical assessment of flooded deep mine voids by test pumping at the Deerplay (Lancashire) and Frances (Fife) Collieries: Nuttall, C. A. et al. Geological Society Special Publication, 2002, 198, 315–326

Long-term changes in quality of discharge water from abandoned underground coal mines in Uniontown Syncline, Fayette County, PA, USA

Changes in water quality over 25 years have been documented for discharges from an extensive network of abandoned underground coal mines in the Uniontown Syncline, Fayette County, PA, USA. A baseline study of 136 mine discharges in the syncline was conducted in 1974–1975. In 1998–2000, follow-up water flow and quality monitoring was conducted at 21 selected discharges for 2 years to assess the degree of mine water-quality improvement since 1974–1975. The data from the two periods of time were compared, with consideration of differences in measurement methods. The degree and rate of water-quality improvement was found to be highly dependent on the amount and duration of flooding in the mine voids. Water quality of discharges from the substantially flooded mine voids improved significantly, going from acidic water with high sulfate and iron concentrations in 1974–1975 to alkaline water with substantially lower sulfate and iron concentrations in 1998–2000. In contrast, the water quality in the unflooded mines showed less improvement over the 25 years between studies. The water discharging from the unflooded mines in 1974–1975 was acidic with high sulfate concentrations and in 1998–2000 was still acidic but showed somewhat lower sulfate and iron concentrations, reflecting depletion of readily available pyrite. The data obtained provide insight into the potential and rate of natural amelioration of mine water quality in different abandoned underground coal mine systems.