Investigations of the impact of effluent from the Ok Tedi copper mine on the fisheries resource in the Fly River,papua new guinea

The Ok Tedi copper mine currently discharges overburden and partially treated ore residues into the Ok Tedi, a major tributary of the Fly River. At peak production these discharges will result in an increase in the suspended solids levels in the Fly River from a background of 76 mg L^-1 to approximately 800 mg L^-1, and an increase in median particulate copper levels from 90 g g^-1 to 1220 g g^-1. The dissolved copper levels are not expected to exceed 10 g L^-1. The Strickland River, a tributary of the Fly River, has natural suspended solids levels in excess of 500 mg L^-1 and provided a natural bioassay for the impacts of elevated suspended solids levels. The fish communities of the Strickland River were dominated by the sediment-tolerant Ariidae catfish, but the levels of biomass caught per standard sampling effort were comparable to those in the Fly River. Laboratory bioassays were performed to assess the impact of particulate associated copper. The fish fauna of the Fly River does not include any standard test species, and it is unlikely that standard species would be unaffected by the high suspended solids levels of the Fly River. Test species were selected on the basis of ecological importance. Two freshwater prawns,Macrobrachium rosenbergii andMacrobrachium sp., a catfish,Neosilurus ater, and a cladoceran,Ceriodaphnia dubia, were tested for acute toxicity and bioaccumulation of copper from particulates. There was no evidence of acute toxicity to prawns or fish, nor for bioaccumulation by prawns. Acute toxicity toC. dubia and bioaccumulation byN. ater were probably due to dissolved copper in the test environments. The test results indicated that provided dissolved copper levels did not exceed the predicted levels, there should not be toxic affects of particulate-associated copper in the Fly River.     

Environmental contaminants in freshwater fish and their risk to piscivorous wildlife based on a national monitoring program

Organochlorine chemical residues and elemental concentrations were measured in piscivorous and benthivorous fish at 111 sites from large U.S. river basins. Potential contaminant sources such as urban and agricultural runoff, industrial discharges, mine drainage, and irrigation varied among the sampling sites. Our objectives were to provide summary statistics for chemical contaminants and to determine if contaminant concentrations in the fish were a risk to wildlife that forage at these sites. Concentrations of dieldrin, total DDT, total PCBs, toxaphene, TCDD-EQ, cadmium, chromium, mercury, lead, selenium, and zinc exceeded toxicity thresholds to protect fish and piscivorous wildlife in samples from at least one site; most exceedences were for total PCBs, mercury, and zinc. Chemical concentrations in fish from the Mississippi River Basin exceeded the greatest number of toxicity thresholds. Screening level wildlife risk analysis models were developed for bald eagle and mink using no adverse effect levels (NOAELs), which were derived from adult dietary exposure or tissue concentration studies and based primarily on reproductive endpoints. No effect hazard concentrations (NEHC) were calculated by comparing the NOAEL to the food ingestion rate (dietary-based NOAEL) or biomagnification factor (tissue-based NOAEL) of each receptor. Piscivorous wildlife may be at risk from a contaminant if the measured concentration in fish exceeds the NEHC. Concentrations of most organochlorine residues and elemental contaminants represented no to low risk to bald eagle and mink at most sites. The risk associated with pentachloroanisole, aldrin, Dacthal, methoxychlor, mirex, and toxaphene was unknown because NOAELs for these contaminants were not available for bald eagle or mink. Risk differed among modeled species and sites. Our screening level analysis indicates that the greatest risk to piscivorous wildlife was from total DDT, total PCBs, TCDD-EQ, mercury, and selenium. Bald eagles were at greater risk to total DDT and total PCBs than mink, whereas risks of TCDD-EQ, mercury, and selenium were greater to mink than bald eagle.     

Trace metal levels in Prochilodus lineatus collected from the La Plata River, Argentina

Most of the industrial, urban and sewage discharges are released into the La Plata River, Argentina without any previous treatment. However, few works have investigated the extent of metal contamination. The aim of this study was to assess the levels of cadmium, copper, lead and zinc in liver and gills of adults Prochilodus lineatus collected from three sampling stations along the coast of the La Plata River: Berazategui, Berisso and Atalaya (from north to south). Samplings were performed during 2002 and 2004. Berazategui and Berisso were located nearby the main ducts that discharge the urban and domestic waste disposal from the cities of Buenos Aires and La Plata, respectively. The third station, Atalaya, was free of sewage discharges. Levels of cadmium and copper in liver were always higher than those found in gills. Instead, for lead and zinc, high levels were observed either in liver or gills, depending on the sampling station and the sampling period. In both tissues, the concentrations of metals did not differ significantly between male and female fish. In liver samples, the concentrations of cadmium, copper and zinc tended to increase from north to south. Instead, the levels of lead followed an opposite pattern. No clear tendencies were observed in gill samples. The data may be useful as reference levels of metal contaminants in P. lineatus, the most important fish species in the La Plata River system.     

Evaluation of modeling as a tool to determine the potential impacts related to drilling wastes in the Brazilian offshore

Numerical models that predict the fate of drilling discharges at sea constitute a valuable tool for both the oil industry and regulatory agencies. In order to provide reliable estimates, models must be validated through the comparison of predictions with field or laboratory observations. In this paper, we used the Offshore Operators Committee Model to simulate the discharges from two wells drilled at Campos Basin, offshore SE Brazil, and compared the results with field observations obtained 3 months after drilling. The comparison showed that the model provided reasonable predictions, considering that data about currents were reconstructed and theoretical data were used to characterize the classes of solids. The model proved to be a valuable tool to determine the degree of potential impact associated to drilling activities. However, since the accuracy of the model is directly dependent on the quality of input data, different possible scenarios should be considered when used for forecast modeling.     

Validation of two air quality models for Indian mining conditions

All major mining activity particularly opencast mining contributes to the problem of suspended particulate matter (SPM)directly or indirectly. Therefore, assessment and prediction are required to prevent and minimize the deterioration of SPM due tovarious opencast mining operations. Determination of emission rate of SPM for these activities and validation of air quality models are the first and foremost concern. In view of the above, the study was taken up for determination of emission rate for SPMto calculate emission rate of various opencast mining activitiesand validation of commonly used two air quality models for Indianmining conditions. To achieve the objectives, eight coal and three iron ore mining sites were selected to generate site specific emission data by considering type of mining, method of working, geographical location, accessibility and above all resource availability. The study covers various mining activitiesand locations including drilling, overburden loading and unloading, coal/mineral loading and unloading, coal handling orscreening plant, exposed overburden dump, stock yard, workshop, exposed pit surface, transport road and haul road. Validation of the study was carried out through Fugitive Dust Model (FDM) and Point, Area and Line sources model (PAL2) by assigning the measured emission rate for each mining activity, meteorologicaldata and other details of the respective mine as an input to the models. Both the models were run separately for the same set ofinput data for each mine to get the predicted SPM concentrationat three receptor locations for each mine. The receptor locationswere selected such a way that at the same places the actual filedmeasurement were carried out for SPM concentration. Statisticalanalysis was carried out to assess the performance of the modelsbased on a set measured and predicted SPM concentration data. The value of coefficient of correlation for PAL2 and FDM was calculated to be 0.990-0.994 and 0.966-0.997, respectively, which shows a fairly good agreement between measured and predicted values of SPM concentration. The average index of agreement values for PAL2 and FDM was found to be 0.665 and0.752, respectively, which represents that the prediction by PAL2 and FDM models are accurate by 66.5 and 75.2%, respectively. These indicate that FDM model is more suited for Indian mining conditions.     

The value of long-term environmental monitoring programs: an Ohio River case study

As a subset of environmental monitoring, fish sampling programs have been an important part of assessing the potential impacts of water withdrawals and effluent discharges on fish populations for many years. New environmental regulations often require that adverse environmental impacts to fish populations be minimized. Without long-term field data, population evaluations may incorrectly indicate adverse impacts where none exist or no impact where one is likely to occur. Several electric utility companies have funded the Ohio River Ecological Research Program, which has been in existence for over 40 years and consists of fish, habitat, and water quality studies at multiple power plant sites on the mainstem Ohio River. Sampling includes seasonal night-time electrofishing and daytime beach seining at three upstream and three downstream locations near each plant. The long-term nature of the program allows for the establishment of aquatic community indices to support evaluations of technology performance, the collaborative development of compliance metrics, and the assessment of fish population trends. Studies have concluded that the Ohio River fish community has improved in response to better water quality and that power plant fish entrainment and impingement and thermal discharges have had little or no measureable impact. Through collaboration and the use of long-term data, $6.3 million in monitoring costs have been saved during recent fish impingement studies. The ability to access a multiyear fish abundance database, with its associated data on age, growth, and fecundity, improves the quality of such evaluations and reduces the need for extensive field sampling at individual locations.     

Decision support methods for the environmental assessment of contamination at mining sites

Polluting mine accidents and widespread environmental contamination associated with historic mining in Europe and elsewhere has triggered the improvement of related environmental legislation and of the environmental assessment and management methods for the mining industry. Mining has some unique features such as natural background pollution associated with natural mineral deposits, industrial activities and contamination located in the three-dimensional sub-surface space, the problem of long-term remediation after mine closure, problem of secondary contaminated areas around mine sites and abandoned mines in historic regions like Europe. These mining-specific problems require special tools to address the complexity of the environmental problems of mining-related contamination. The objective of this paper is to review and evaluate some of the decision support methods that have been developed and applied to mining contamination. In this paper, only those methods that are both efficient decision support tools and provide a ‘holistic’ approach to the complex problem as well are considered. These tools are (1) landscape ecology, (2) industrial ecology, (3) landscape geochemistry, (4) geo-environmental models, (5) environmental impact assessment, (6) environmental risk assessment, (7) material flow analysis and (8) life cycle assessment. This unique inter-disciplinary study should enable both the researcher and the practitioner to obtain broad view on the state-of-the-art of decision support methods for the environmental assessment of contamination at mine sites. Documented examples and abundant references are also provided.     

Relationships between water mass characteristics and estimates of fish population abundance from trawl surveys

The Canadian Department of Fisheries and Oceans conducts annual bottom trawl surveys to monitor changes in the abundance of the major commercially important groundfish populations. Some of these surveys have been in operation for almost 20 yr. The estimates from these surveys often indicate rapid changes in abundance over time beyond that expected from the population dynamics of the fish. Much of this interannual change has been interpreted as variation, the magnitude of which has often made it difficult to measure anything but the most severe effects of fishing, pollution or any other intervention on the population. Recent studies have shown that some of this variation may be attributed to changes in catchability of fish due to the effects of environmental variables on fish distribution. Annual changes in abundance as estimated from such field surveys may be confounded by changes in catchability due to annual changes in environmental conditions. In this study, trawl catches of age 4 Atlantic cod (Gadus morhua) from surveys conducted during March 1979–1988 were compared with concurrent measurements of bottom salinity, temperature and depth. Large catches of age 4 cod are more likely to occur in water characterized as the intermediate cold layer defined by salinities of 32–33.5 and temperatures<5°C. This relationship also appears to be modified by depth. We further show that internnual changes in the estimated abundance from the surveys were, in a number of cases, coincident with changes in the proportion of the bottom water composed of the intermediate cold water layer. The implications that these patterns may have on interpreting trends in the estimates of abundance from trawl surveys are discussed.     

Microplankton bloom in a brackish water lagoon of Terengganu

Increased primary plankton productivity was observed in a brackish water lagoon of Terengganu during the study period between January 1988 to December 1988. The lagoon is also the site for the fish cage culture activities of sea bass during the study period. An examination of water quality at the sampling stations during the study period indicated that both the organic and inorganic nutrients were high during the pre-monsoon period. The source of the nutrient in the lagoon was believed to be derived from the agro-based industrial effluents, fertilisers from paddy fields as well as untreated human and animal wastes. This coincided with the peak production of plankton in the surface waters of the brackish water lagoon. During this period both cultured and indigenous fish species were seen to suffer from oxygen asphyxiation (suffocation due to lack of oxygen). The primary productivity values ranged from 9 to 22 g/L/h during the peak period while the microplankton species were composed of diatom, flagellates and dinoflagellates. Reduction in the primary productivity values were obtained with reduction in sallinity, specially during the peak monsoon months (November to March) corresponding to the Northeast monsoon period.     

Metal accumulation in the tissues of grass carps (Ctenopharyngodon idellus) from fresh water around a copper mine in Southeast China

Mining effluents are the main source of metals in the surrounding aquatic environment. The mining district of Purple Mountain has a history of copper mining for more than 30 years, but there is limited investigation of metal bioaccumulation in the aquatic creatures from the Tingjiang river catchment affected by the mining activities. In this study, we collected grass carps (Ctenopharyngodon idellus) from four sites, and analyzed the accumulation of chromium (Cr), nickel (Ni), manganese (Mn), copper (Cu), zinc (Zn), arsenic (As), cadmium (Cd), mercury (Hg), and lead (Pb) in ten tissues (scale, skin, muscle, gill, liver, kidney, fish maw, heart, stomach, and intestine) of the fish samples. Among all tissue samples, the highest concentrations (micrograms per gram wet weight) of Ni (0.263), Cu (69.2), Zn (84.0), As (0.259), Cd (0.640), Hg (0.051), and Pb (0.534) were noted in the liver, gill, and kidney tissues, whereas the highest concentrations of Cr (0.356) and Mn (62.7) were detected in the skin and intestine, respectively. These results gave a better understanding of the variability of metals distribution in different fish tissues. In comparison with the sample sites, metals (especially Mn, Cu, Zn, Ni, and Pb) in liver, gill, kidney, stomach, and intestine showed more inter-site differences than other tissues. The inter-site differences also revealed that site 1 and 2 increased fish uptake of Cu, Zn, Ni, and Pb, which may indicate that the copper mine and urban effluents contributed to high levels of these metals in aquatic environments in site 1 and 2. A potential food safety issue may emerge depending on the mining activities in this region because some metals in a few tissue samples exceeded the guideline values for human consumption of fish.     

Dispersion and bioaccumulation of elements from an open-pit olivine mine in Southwest Greenland assessed using lichens, seaweeds, mussels and fish

This study investigated dispersion and bioaccumulation of mining-related elements from an open-pit olivine mine at Seqi in Southwest Greenland (64°?N) using lichens (Flavocetraria nivalis), seaweeds (Fucus vesiculosus), mussels (Mytilus edulis) and fish (Myoxocephalus scorpius). The mine operated between 2005 and 2009, and samples were taken every year within a monitoring area 0–17 km from the mine during the period 2004–2011. A total of 46 elements were analysed in the samples. After mining began, highly elevated metal concentrations, especially nickel (Ni), chromium (Cr), iron (Fe) and cobalt (Co), were observed in lichens relative to pre-mining levels (up to a factor of 130) caused by dust dispersion from the mining activity. Elevated metal concentrations could be measured in lichens in distances up to ~5 km from the mine/ore treatment facility. Moderately elevated concentrations of Ni and Cr (up to a factor of 7) were also observed in seaweeds and mussels but only in close vicinity (<1 km) to the mine. Analyses of fish showed no significant changes in element composition. After mine closure, the elevated metal concentrations in lichens, seaweeds and mussels decreased markedly, and in 2011, significantly elevated metal concentrations could only be measured in lichens and only within a distance of 1 km from the mine.     

Assessment of the potential of the rock gunnel (Pholis gunnellus) along the Atlantic coast of Canada as a species for monitoring the reproductive impacts of contaminant exposures

Evaluating the impacts of point source discharges on fish species in estuarine environments can be challenging because of a paucity of resident species. We evaluated the biology of rock gunnel (Pholis gunnellus) at three relatively uncontaminated sites in the Bay of Fundy, along the Atlantic coast of Canada. Rock gunnel are seasonally resident (April to November) in tide pools, but little was known about their life history in Atlantic Canada or their potential for use for monitoring environmental quality. Fish were collected between April and November, and ranged from 2.46 g--15.2~g in weight and 97 mm-170 mm in length, with a maximum age of 7 years. Both males and females were similar in size, and both reached sexual maturity at a size of 5.5 g. Organ weights and condition indices of fish were stable from spring when they returned from offshore (April to May) until late summer (August to September), but fall fish (October to November) had slightly larger gonads, livers and condition indices. Rock gunnel may be a useful indicator to provide insight into local impacts of point sources over a short time period. However, they do not provide adequate information on reproductive development and performance since they are not exposed to onshore contaminants during the periods of gonadal development that have most commonly found to be sensitive to anthropogenic stressors.     

Mercury trends in fish from rivers and lakes in the United States, 1969�C2005

A national dataset on concentrations of mercury in fish, compiled mainly from state and federal monitoring programs, was used to evaluate trends in mercury (Hg) in fish from US rivers and lakes. Trends were analyzed on data aggregated by site and by state, using samples of the same fish species and tissue type, and using fish of similar lengths. Site-based trends were evaluated from 1969 to 2005, but focused on a subset of the data from 1969 to 1987. Data aggregated by state were used to evaluate trends in fish Hg concentrations from 1988 to 2005. In addition, the most recent Hg fish data (1996?C2005) were compared to wet Hg deposition data from the Mercury Deposition Network (MDN) over the same period. Downward trends in Hg concentrations in fish from data collected during 1969?C1987 exceeded upward trends by a ratio of 6 to 1. Declining Hg accumulation rates in sediment and peat cores reported by many studies during the 1970s and 1980s correspond with the period when the most downward trends in fish Hg concentrations occurred. Downward Hg trends in both sediment cores and fish were also consistent with the implementation of stricter regulatory controls of direct releases of Hg to the atmosphere and surface waters during the same period. The southeastern USA had more upward Hg trends in fish than other regions for both site and state aggregated data. Upward Hg trends in fish from the southeastern USA were associated with increases in wet deposition in the region and may be attributed to a greater influence of global atmospheric Hg emissions in the southeastern USA. No significant trends were found in 62% of the fish species from six states from 1996 to 2005. A lack of Hg trends in fish in the more recent data was consistent with the lack of trends in wet Hg deposition at MDN sites and with relatively constant global emissions during the same time period. Although few significant trends were observed in the more recent Hg concentrations in fish, it is anticipated that Hg concentrations in fish will respond to changes in atmospheric Hg deposition, however, the magnitude and timing of the response is uncertain.     

Association of spinal deformities with heavy metal bioaccumulation in natural populations of grass goby, Zosterisessor ophiocephalus Pallas, 1811 from the Gulf of Gabès (Tunisia)

The present study illustrates an analysis of spinal deformities associated with metal accumulation in natural populations of Zosterisessor ophiocephalus derived from polluted (S1) and unpolluted (S2) areas in the Gulf of Gabès in Tunisia. Three basic types of spinal deformities were detected: kyphosis, scoliosis and lordosis. These basic deformities frequently co-occur. Spinal deformities were observed in 10.72% of the total examined fish (n = 494). Deformed fish were 3.85 times more frequent in S1 than in S2. In both sexes, the highest occurrence of deformities was observed in the 111-120 mm class decreasing thereafter with fish length. Hepatic concentrations of cadmium (Cd), copper (Cu) and zinc (Zn) were significantly higher in Z. ophiocephalus from S1 than those from S2. Comparisons between fish in each site showed that liver concentrations of Cd and Zn were significantly higher in deformed fish than in normal fish. The relationship between metals accumulation and observed spinal deformities as well as the suitability of this kind of studies for environmental monitoring are discussed.     

Prediction of stream fish assemblages from land use characteristics: implications for cost-effective design of monitoring programmes

Increasing human impact on stream ecosystems has resulted in a growing need for tools helping managers to develop conservations strategies, and environmental monitoring is crucial for this development. This paper describes the development of models predicting the presence of fish assemblages in lowland streams using solely cost-effective GIS-derived land use variables. Three hundred thirty-five stream sites were separated into two groups based on size. Within each group, fish abundance data and cluster analysis were used to determine the composition of fish assemblages. The occurrence of assemblages was predicted using a dataset containing land use variables at three spatial scales (50 m riparian corridor, 500 m riparian corridor and the entire catchment) supplemented by a dataset on in-stream variables. The overall classification success varied between 66.1–81.1% and was only marginally better when using in-stream variables than when applying only GIS variables. Also, the prediction power of a model combining GIS and in-stream variables was only slightly better than prediction based solely on GIS variables. The possibility of obtaining precise predictions without using costly in-stream variables offers great potential in the design of monitoring programmes as the distribution of monitoring sites along a gradient in ecological quality can be done at a low cost.     

Distribution of suspended particulate matter with trace element composition and apportionment of possible sources in the Raniganj coalfield,India

Ambient air monitoring for suspended particulate matter was carried over a period of one year in some coal mining areas of the Raniganj coalfield. Concentrations of seven elements in suspended particulate matter were determined. The set of data obtained was analysed to determine the sources of trace elements by factor analysis. The data could be interpreted on the basis of five factors. These factors are attributed to various sources of particulate matter by noting the dependence of factors on the elements.     

Development of GP and GEP models to estimate an environmental issue induced by blasting operation

Air overpressure (AOp) is one of the most adverse effects induced by blasting in the surface mines and civil projects. So, proper evaluation and estimation of the AOp is important for minimizing the environmental problems resulting from blasting. The main aim of this study is to estimate AOp produced by blasting operation in Miduk copper mine, Iran, developing two artificial intelligence models, i.e., genetic programming (GP) and gene expression programming (GEP). Then, the accuracy of the GP and GEP models has been compared to multiple linear regression (MLR) and three empirical models. For this purpose, 92 blasting events were investigated, and subsequently, the AOp values were carefully measured. Moreover, in each operation, the values of maximum charge per delay and distance from blast points, as two effective parameters on the AOp, were measured. After predicting by the predictive models, their performance prediction was checked in terms of variance account for (VAF), coefficient of determination (CoD), and root mean square error (RMSE). Finally, it was found that the GEP with VAF of 94.12%, CoD of 0.941, and RMSE of 0.06 is a more precise model than other predictive models for the AOp prediction in the Miduk copper mine, and it can be introduced as a new powerful tool for estimating the AOp resulting from blasting.     

Development of the concept of spatial–temporal mask for testing effects of discharge from well-drilling activities on biological communities

This paper describes the use of Bayesian spatial models to develop the concept of a spatial–temporal mask for the purpose of identifying regions in which before and after drilling effects are most clearly defined and from which the consequences of exposure of macrofauna and meiofauna to the release of drilling discharges can be evaluated over time. To determine the effects of drilling fluids and drill-cuttings on the marine benthic community, it is essential to know not only where discharged materials ended up within the possible impact area, but also the chemical concentrations to which biota were exposed during and after drilling. Barium and light hydrocarbons were used as chemical tracers for water-based and non-aqueous-based fluids in a shallow water site in the Campos Basin, off the coast of Brazil. Since the site showed evidence of exposure to waste material from earlier drilling, the analysis needed to take into account the background concentrations of these compounds. Using the Bayesian models, concentrations at unsampled sites were predicted and regions altered and previously contaminated were identified.     

Radiation exposure to Marine biota around the Fukushima Daiichi NPP

The dose rates for six marine organisms, pelagic fish, benthic fish, mollusks, crustaceans, macroalgae, and polychaete worms, representative in marine ecosystems, have been predicted by the equilibrium model with the measured seawater activity concentrations at three locations around the Fukushima Daiich nuclear power plant after the accident on March 11, 2011. Model prediction showed that total dose rates for the biota in the costal sea reached 4.8E4 μGy/d for pelagic fish, 3.6E6 μGy/d for crustaceans, 3.8E6 μGy/d for benthic fish, 5.2E6 μGy/d for macroalgae, 6.6E6 μGy/d for mollusks, and 8.0E6 μGy/d for polychaete worms. The predicted total dose rates remained above the UNSCEAR’s (United Nations Scientific Committee on the Effect of Atomic Radiation) benchmark level (1.0E4 μGy/d for an individual aquatic organism), for only the initial short period, which seems to be insufficiently long to bring about any detrimental effect on the marine biota at the population level. Furthermore, the total dose rates for benthic fish and crustaceans approximated using the measured activity concentration of the biota and bottom sediment was well below the benchmark level. From these results, it may be concluded that the impact of the ionizing radiation on the marine biota around the Fukushima NPP as a consequence of the accident would be insignificant.     

Predicting copper concentrations in acid mine drainage: a comparative analysis of five machine learning techniques

Acid mine drainage (AMD) is a global problem that may have serious human health and environmental implications. Laboratory and field tests are commonly used for predicting AMD, however, this is challenging since its formation varies from site-to-site for a number of reasons. Furthermore, these tests are often conducted at small-scale over a short period of time. Subsequently, extrapolation of these results into large-scale setting of mine sites introduce huge uncertainties for decision-makers. This study presents machine learning techniques to develop models to predict AMD quality using historical monitoring data of a mine site. The machine learning techniques explored in this study include artificial neural networks (ANN), support vector machine with polynomial (SVM-Poly) and radial base function (SVM-RBF) kernels, model tree (M5P), and K-nearest neighbors (K-NN). Input variables (physico-chemical parameters) that influence drainage dynamics are identified and used to develop models to predict copper concentrations. For these selected techniques, the predictive accuracy and uncertainty were evaluated based on different statistical measures. The results showed that SVM-Poly performed best, followed by the SVM-RBF, ANN, M5P, and KNN techniques. Overall, this study demonstrates that the machine learning techniques are promising tools for predicting AMD quality.