Introduction
Amulsar Gold Mine is located at the border of Vayots Dzor and Syunik Marzes of the Republic of Armenia, and to the north of the Zangezur Ridge. In 2014, the Government of Armenia granted Lydian Armenia CJSC (hereinafter Lydian) the right to exploit the Amulsar mine and the permit for construction.

In 2015, Lydian released the Amulsar Environmental and Social Impact Assessment, and its updated version in 2016. On April 29, 2016, the Amulsar Project Environmental Impact Assessment (EIA) was approved and signed by the Ministry of Nature Protection.

The extraction of gold from the rocks is planned to be done by heap leach system, which will be used for the first time in Armenia. According to the program presented by Lydian, it is planned to extract about $2.6 billion worth of gold during the 11 years of mine operation. About $426 million is planned to be invested in the first two years. Total revenues through taxes and royalties paid to the state budget during the entire life of the mine would total approximately $635 million. Approximately 1,300 people would be employed during the construction phase and 770 in the operational phase.

It is projected that the cumulative impact of mine operation on GDP will average about $185 million or 1.4% of GDP. The average annual export from the Amulsar Gold Mine would make about 6.5% of Armenia’s total export volume.

Project Disagreements
Although Lydian has committed to build and operate the Amulsar mine in line with the best mining practices under which it has designed and formulated an unprecedented metal mining project in Armenia, nonetheless, professionals and a large portion of the public (including environmentalists) are not convinced this mining operation would be without major environmental risks.

Concerns are about the uncontrollable risks associated with the mine development. The problem is compounded by the geographical location of Amulsar where it is the source of a major water resources formation. It requires scientifically accurate, independent, and objective analysis by reputable international experts to determine the extent of the problems.

Although in accordance with the legislation of the Republic of Armenia it is not required, but Lydian has initiated the environmental impact assessment with the involvement of international experts. However, results of the various studies do not develop sufficient confidence in the possibility of safe operation of the mine. This is due in part to the fact that the international experts that have been selected to perform the environmental impact study and evaluation are paid for by Lydian. On the other hand, the Ministry of Nature Protection of Armenia did not have (and still does not have) the mandate by the law and financial means to hire independent international experts to review complex environmental documents. As a result, some local and international experts do not object to Lydian’s assurances that the mine can be operated safely.

Environmentalists got a broader community involved and organized the blockade of the roads leading to Amulsar mine, which added the political component to the environmental and economic components of the issue.

Building on the latter and forming high public expectations, the government prolonged the resolution of the matter and initiated a criminal case on the grounds of possible abuses. As part of the investigation, government hired an internationally recognized company (Elard and TRC censorism) for $400,000 to conduct an Independent evaluation of some of the Lydian’s documents. However, results of this study were not convincing to the professional community that it is safe to operate the Amulsar mine. Furthermore, the general public confused with the ever-changing views of the government on this issue and this flip flop of the government is hurting the country’s international standing.

This approach cannot be continued forever. Political saber raddling component must be replaced by a well-reasoned and professionally justified approach which would be acceptable to the public. It is crucial to ensure that this process is followed by clarifying professional approaches, by integrating economically, environmentally as well as socially balanced perspectives, and ultimately by finalizing it by organizing a national referendum on Amulsar.

Project related Technical Concerns

Some of the questions raised by locally and internationally recognized professionals have generally been reflected in the audit report prepared by ELARD.

The ARF Environmental Committee, after studying and analyzing the report prepared by ELARD, notes that it addresses four areas, Water Resources, Geology, Biodiversity, and Air Quality. Report indicates that Lydian has presented issues related to Biodiversity sufficiently but they could have improved their presentation if they had included some additional information such as maps. However, the biodiversity management plan is missing the operational section and map detailing the measures, their location, how to implement and who is responsible for implementation, and mostly how will those measures reduce the impact of each receptor.

ELARD has criticized using Raduga model to assess air quality. They conclude that this old model is not adequate for a project like Amulsar. ELARD has also identified some short comings in evaluating impacts from fine dust particles and certain gases that are typically generated in gold mine operations. ELARD concludes that the monitoring plan presented by Lydian for air quality is relevant and acceptable but shall be augmented to include seven additional items.

ELARD is extremely critical about water resources and geology impact analysis.

Cross-sections from the 3-D geologic model prepared by Lydian only show the geologic relationships in the vicinity of the pits area. According to the ELARD report, this deficiency translates to poor understanding of the connection between sources and receptors of groundwater contaminants. For such an environmentally-sensitive area, the omission of illustrations of the structural and stratigraphic relationships across the Project Area is a serious shortcoming in the model. The geologic model is the basis for models that numerically represent groundwater flow and contaminant transport.

ELARD report continues to say that the description of the local geology in the Lydian environmental document (ShMAG) is disorganized, incomplete, and incomprehensible. The text gives the impression of poorly understood structural and stratigraphic relationships, distribution and causes of alteration types, and the sequence of events in the genesis and occurrence of the various rock types.

There are several geological faults in the project area and these faults may be barriers and/or conduits of groundwater flow. It is the finding of ELARD that faults within the project impact area are not considered in the modeling of the geological effects by Lydian. Furthermore, ELARD reports that volcanic rocks are brittle, with widespread fracture permeability. These characteristics influence groundwater flow and transport rates. Rocks cannot be assumed to be a homogeneous clay zone, void of fractures as it is done by Lydian. Correct numerical model representations of groundwater flow and solute transport from the pits and project facilities to receptors (rivers and springs) are dependent on the structure and characteristics of the rock throughout the flow and transport paths. ELARD report concludes that it is incorrect to state that the mine cannot impact regions (including fresh water springs) next to the areas adjacent to the project sites.

ELARD mentions that it is noteworthy to state that the Modified Sobek method which was used for the Project Acid-Base Accounting (ABA), which determines potential acidity based only on sulfide sulfur. According to ELARD, this approach is clearly incorrect for the Project because nearly all samples from the pit areas have acidic pH values that is indicative of acidic sulfate salts. ELARD suggests that the analyzed percentage of sulfate should have been included in the potential acidity calculation.

ShMAG states that the Upper Volcanics rock layer is not acid generating. However, ELARD mentions that data in ShMAG appendixes show that there are many samples with significant pyritic sulfur. The higher sulfide percentages provide more evidence that the Upper Volcanics layer is not homogeneous and should be sub-divided into distinct rock sub-types.

According to ELARD, Lydian’s assessment of Acid Rock Drainage (ARD) reactions that would occur in the Amulsar rocks is misleading because the analysis fails to recognize the importance of ferrous iron oxidation. Using this approach, underestimates the potential for ARD generation and the associated water quality, environmental impacts, as well as water treatment requirements.

No pumping tests were undertaken by Lydian, which per ELARD it is a serious omission in the characterization of hydraulic properties of the Project vicinity. Pumping tests are a standard procedure for hydraulic characterization and are indispensable for fractured rock. Given the environmentally-sensitive setting, the limited distribution of hydraulic tests and the lack of pumping tests are inadequate for characterizing the hydraulic properties across the groundwater study area. ELARD concludes that areas from the mine pits and facilities to the rivers and reservoirs are uncharacterized and cannot be assumed to have the same hydraulic properties of the tested areas.

Transport of contaminated groundwater is several orders of magnitude greater in fractured media over porous media in accordance to ELARD. The extent of connectivity of fractures determines how rapid flow and transport of underground water occurs. Only through conducting pumping tests and groundwater tracer tests can fracture connectivity be assessed.

ELARD has reviewed water pit seepage study conducted by Lydian. Even though Elard mentions that the right models were utilized but results of the models are incorrect because an unrealistic high amount of evaporation from exposed rock and backfill is assumed for the modeling. In another word, Lydian claims that a large percentage of rain and snow waters will be evaporated and not seeped into the ground, where as it is known that most of the precipitation and snowmelt infiltrates rapidly through the voids in the ground.

ELARD suspects that Lydian uses incorrect runoff coefficients for the pit walls, which means they assumed less water will go into the ground. Elard assumes that much less evaporation will occur in simulations.

It is the determination of ELARD that the water fluxes from the pit seepage modeling are incorrect. Use of these incorrect fluxes in the regional groundwater flow model results in incorrect assessments of impacts to groundwater levels and springs. Furthermore, solute transport simulations are severely underestimating potential impacts to the groundwater and springs from acid mine drainage. Due to overestimating the evaporation and underestimating the runoff, ELARD determined that water and mass fluxes reporting to the passive treatment system (PTS) are underestimated and the amount of makeup water is overestimated.
Elard has determined that the calculated volumetric fluxes that report to the base of the barren rock storage facility (BRSF) are greatly underestimated due to the incorrect model parameterization. The underestimated water fluxes translate to underestimated mass fluxes related to contact of the seepage with the acid-generating rocks.

The geology in the pit areas is poorly represented per ELARD. The simplistic numerical representation of the subsurface in the existing model is inadequate for making quantitative predictions. Notwithstanding this inadequacy, numerous quantitative predictions, which are unreliable, were made. ELARD continues by saying that transient simulations were performed to estimate pit inflows, but the model is also a poor predictor because a transient calibration was not performed. Based on their review, ELARD finds that model is a poor indicator of the effects of mining activities on groundwater levels and groundwater flows.

ShMAG states that hydrographs for wells in the Project site indicate that groundwater levels rise rapidly in the Spring in response to snowmelt and decline rapidly following snowmelt. ELARD concludes from this data that this behavior is indicative of fractures in the rocks provide an effective connectivity; therefore, Lydian’s predicted changes in groundwater levels are unrealistic as contaminated water would be traveling underground much faster. Keep in mind that the Amulsar rocks have essentially no pH buffering capacity, which means the acidity of the water cannot be reduced as water travels underground.

Surface water and groundwater are interconnected. According to ELARD, Lydian’s assumption of fixed amount of water percolates through the ground uniformly across large areas of the site and recharges groundwater at a rate that is constant with time is an overly simplistic approach that leads to poor model representation of subsurface properties and unreliable conclusions about surface water and groundwater.

The water runoff, infiltration through backfill, seepage to groundwater from the pit seepage modeling of the Tigranes-Artavasdes-Arshak pit are significantly underestimated. The seepage rate to groundwater from the Erato pit is also underestimated. ELARD concludes that underestimated pit seepage underestimates the potential impacts to groundwater and surface water. The screening level spreadsheet (analytical) model approach used for solute transport is inappropriate for a project of this scope, extent, and complexity in such an environmentally-sensitive area.

ELARD mentions that ammonia in the wastewater will most likely be present at concentrations well in excess of the discharge criterion, but the treatment process for the ammonia is not discussed by Lydian except in brief comments. Nitrate treatment is discussed in a little more detail, but is not given the focus that it should have and the nitrate concentrations appear to have been underestimated. Nitrate and ammonia are likely to be major contaminants that require treatment. The system as designed may not be able to achieve the treatment criteria for either nitrate or ammonia. In accordance to ELARD, a passive system recommended by Lydian is most appropriate for closure and post-closure phases, while an active system is more appropriate for operational phases.

Results of ELARD review showed that Lydian analysis does not include iron and aluminum concentrations typical of ARD, furthermore, iron concentrations projected at detention pond 8 are underestimated by a factor of 106. The discrepancy and high uncertainty in iron concentrations does not give confidence in the modeled water quality and raises concerns about the certainty and reliability of other parameters.

According to ELARD the water quality modeling has significant discrepancies and uncertainties that raise significant concerns about the reliability of water quality projections and ultimately the feasibility and effectiveness of the proposed Passive Treatment System.

ELARD was not able to determine that at the end of the heap leaching operation what will happen to the Heap Leach Solution used in the operation which will contain large amount of cyanide. The reports are not clear on how the Heap Leach Solution will be treated immediately after mining operations cease.
The Amulsar Mine closure cost basis and estimates that are provided in ShMAG was reviewed by ELARD and determined that some cost items are questionable and the overall mine closing cost appears to be underestimated by 50%.

ELARD’s conclusion is that the baseline characterization of the geology of the Project study area is deficient, as well as the interpretation and conceptualization of the geology across the area is too simplistic. Furthermore, ELARD concludes that the water flow from the pit seepage sub-model is incorrect. The major issue is assumption of too much evaporation by Lydian. Use of these flaws in the regional groundwater flow model result in incorrect assessments of future impacts to groundwater levels and springs.

Vorotan Tunnel

The most significant structure at the Groundwater Study Area below depth of groundwater is the Vorotan tunnel, which is a concrete-lined structure traversing from the Spandaryan Reservoir to the Kechut Reservoir. The entrance to the tunnel at Spandaryan Reservoir is currently closed but water flowing out at the discharge of the tunnel at the Kechut Reservoir indicates that groundwater is entering the tunnel. Lake Sevan and the Kechut Reservoir are linked by the Arpa-Sevan tunnel that directs water to Lake Sevan.

Samples of the Vorotan outfall water show similar composition to the Amulsar ridge spring and groundwater. Lydian report states that the Vorotan tunnel intersects the water table along its length. Presently surface of mountains are oxidized and only a little of the sulfides and other materials find their way to the tunnel by water from rain or snow melt. But once pits are dug, sulfides and other heavy metals will get exposed and they will penetrate deeper into the ground with water from rain and snow melt. In addition, the bottom of these pits will be touching the underground water table. Therefore, sulfides and other heavy metals will easily get into the underground water table and from there the contaminated water would find its way to the Vorotan tunnel which is also located in the same groundwater table.

Lydian claims that it will take 130 years for the contaminated water to reach the tunnel. However, this claim is not substantiated and there is no basis for it. Furthermore, results of tests indicate that after heavy rain, composition of some springs at the base of the mountain change just after few days. In another word, with the undisturbed ground surface, it takes water to travel only few days and when pit is dug, contaminations will get into the ground table much faster and their effects would be seen much faster.

ELARD report confirms that contaminated waters will find their way to the Kechut Reservoir and from there to Lake Sevan. However, they claim that these contaminations would probably not generate significant in concentrations of Lake Sevan water due to the dilutions in reservoir and the size of Lake Sevan. The report mentions that there would local concentrations of harmful chemicals at the discharge location of the Arpa-Sevan tunnel. Lake Sevan is protected by Armenian law which permits no activity that may negatively impact the lake and its ecosystem.

Summary and Recommendations
The Amulsar project has not provided the necessary assurance to a broad range of professionals and the public that the implementation of the project will not create non-mitigatable impacts.

The operation of the gold mine in the Amulsar area contains risks that are uncontrollable and its operation will lead to irreversible consequences. There is no economic progress, social inclusion, and community development when there are overwhelming evidences that there will be negative impacts on various environmental fields, such as air, water, biodiversity, soil, and landscape as well as ultimately affecting human health.

Based on the risks identified by local and international experts as well as conclusions reached by ELARD in its review, the project cannot continue at this stage.
The chemical processes currently under way in the Amulsar mine need to be properly evaluated and controlled. Therefore, it is necessary that Lydian and the state authorized body to carry out continuous monitoring to ascertain the present situation and extent of possible damages to the environment as of today, to identify the magnitude of the damage, and to present the findings to the public.

Preparation for mine operation cannot continue until comprehensive answers to the above questions are provided, as there are obvious uncontrollable risks.
Although the issue of Amulsar as a mine has gone beyond the purely professional domain, it is necessary to re-establish a professional and public dialogue as well as discussions based on the outcome indicators of the processes described above. It is recommended that this issue shall be decided by a referendum.

The following schematic shows the typical mine pit, its relationship to the underground water table, and location of Vorotan tunnel as well as how the pollutants can reach the tunnel.The follwing map shows relationship of Sevan, two reservoirs, two tunnels, and Amulsar mine.