UNITED NATIONS OFFICE FOR THE COORDINATION OF HUMANITARIAN AFFAIRS - OCHA-Online
On 30 January 2000, following a breach in the tailing dam of the Aurul SA Baia Mare Company, a major spill of cyanide-rich tailings waste was released into the river system near Baia Mare in north west Romania. The contaminant travelled via tributaries into the river Somes, Tisza and finally into the Danube before reaching the Black Sea.
Following requests from the Governments of Hungary, Romania and the Federal Republic of Yugoslavia (FRY), and consultations with European Environment Commissioner Margot Wallström and the UN Office for the Co-ordination of Humanitarian Affairs (OCHA), Klaus Toepfer, Executive Director of the United Nations Environment Programme (UNEP), announced on 18 February 2000, that a team of international experts would be sent to the affected area to carry out a scientific analysis of the environmental damage caused by the spill.
The mission was a joint venture of UNEP and OCHA, organised by the Joint UNEP/OCHA Environment Unit, and headed by the Director of UNEP’s Regional Office for Europe. Its terms of reference included an independent, scientific description of the spill, the situation and events causing it, the collection and review of data related to the spill and its environmental implications, and the preparation of recommendations for future action and prevention.
The mission represented a useful model for inter-agency cooperation and multi-disciplinary rapid assessment work.
The mission was limited in size, scope and time, and consequently not intended to provide a full overview of the emergency and its implications. It mainly represented environmental input to a process of international investigation and reviews by, inter alia, the Baia Mare Task Force set up following the visit of European Union (EU) Environment Commissioner Wallström to the area. The results of the assessment mission should be seen in this context, as a starting point rather than the final conclusion. The data and conclusions will have to be refined as further study continues.
The mission, which lasted from 23 February – 6 March 2000, combined sampling, analysis, interviews with relevant national and local experts, discussions with national authorities, affected populations and local Non Governmental Organisations (NGOs).
Sixteen experts from seven countries (Austria, Czech Republic, Finland, Germany, Norway, Sweden, Switzerland) were selected at very short notice to travel to the affected areas. The composition of the expert group is given in appendix 11.2. The range of expertise included in the team covered chemistry, ecotoxicology, biology, process engineering and dam engineering. In addition to the expert group, a four-person UN Disaster Assessment and Coordination (UNDAC) team from the Disaster Relief Branch of OCHA was dispatched to provide essential logistic and coordination support for the mission. Apart from the mission leader, UNEP’s Regional Office for Europe provided a Press Officer and a Scientific Coordinator. The mission also included representatives of the World Health Organization, the UN Economic Commission for Europe, and the European Commission delegations in Romania and Hungary.
The mission had at its disposition three mobile/portable laboratories, provided by the Governments of Germany, Switzerland and the Czech Republic.
Backstopping was provided by the Joint UNEP/OCHA Environment Unit, as well as by the Field Coordination Support Unit (FCSU) and the Military and Civil Defence Unit (MCDU) of OCHA’s Disaster Response Branch in Geneva with technical advice being provided by mining specialists in the UNEP Division of Technology, Industry and Economics in Paris. Considerable logistical and other support was also received from the UN Development Programme (UNDP) Office in Bucharest, the UN Liaison Office (UNLO) in Croatia and the OCHA Office in Belgrade.
During the entire mission, contacts were maintained and consultations held, in writing and by phone, with representatives of the International Commission for the Protection of the Danube River (ICPDR), the Regional Environmental Center (REC), the Worldwilde Fund for Nature (WWF), donor countries and others. With respect to its work in the Federal Republic of Yugoslavia, the mission received valuable advice and support from the UNEP/United Nations Centre for Human Settlements (UNCHS) Balkans Task Force.
The team assembled in Bucharest in Romania, then traveled to the breach site in Baia Mare before crossing the border into Hungary and followed the river system down to the FRY border. Finally, sampling was undertaken along the Danube in the FRY. Through a specific sampling at the mouth of the Danube the team managed to capture evidence of the spill. Maps of the area indicating the routing and sampling work of the mission can be found in appendix 11.1.
The mission divided into seven key areas of investigation :
This report has been produced to provide an overview of the key findings of the mission. Subsequent specialist information detailing analytical methods and findings will be available on a special website, already established by UNEP, to allow the public authorities, industry and the public at large to follow the evolution of events and to gain a better understanding of the technical issues involved in the use of mining, emergency response procedures, etc. The site can be found on http : //www.natural-resources.org/environment/Baiamare.
The mission is grateful for the full and open support of the national and local authorities in each of the countries visited. Without such access to good offices, facilities, experts, information and logistic support, the mission would not have been possible. Furthermore, the support provided by the UNDP office in Bucharest and OCHA in FRY also proved essential, as well as the assistance of the Water Research Institute in the Czech Republic, THW in Germany, the Swiss Disaster Relief together with the Swiss Agency for the Environment and AC-Laboratorium Spiez, the Swedish Rescue Services Agency together with the Royal Institute of Technology and the Swedish Environment Research Institute, the Austrian Ministry of Interior, the Finnish Environment Institute, and the Norwegian Institute for Water Research.UNEP/Global Resource Information Database (GRID)-Geneva provided valuable maps of the area and the course and particulars of the cyanide spill.
The mission is grateful for the support it received, in cash and in kind, from Austria, Czech Republic, Finland, Germany, Norway, Sweden, Switzerland and the United Kingdom.
AURUL is a stock company, jointly owned by Esmeralda, Exploration Limited, Australia, and the Romanian Compania Nationala a Metalelor Pretiosasi si Neferoase (REMIN), established in 1992. The company processes solid wastes from earlier mining activity to recover precious metals, especially gold and silver. In 1993 the company obtained an environmental permit from the Ministry of Waters, Forests and Environmental Protection. In 1997, after receiving the Site Construction Permit from the Maramures County Council, construction of the recovering plant commenced. In 1999 the operational permit, based on documentation contained in an environmental impact assessment (EIA), was obtained. The company started operation in May 1999 by processing an existing 30 year-old tailing dam (Meda dam) located near Baia Mare city, to the west, close to the residential area.
The source of the raw materials utilized by the Aurul company is of mining residues accumulated in the Meda tailing dam. These solid wastes resulted from former gold and silver extraction. The technology introduced by Aurul utilizes high concentrations of free cyanide in the process waters for the extraction of the precious metals. The whole process is designed to operate in closed circuit with the cyanide containing waters being re-used, after solids sedimentation in the Aurul pond.
Aurul provided new jobs (150 Romanians directly or indirectly employed and 200 jobs being created during the construction phase) and investments into the Baia Mare mining area, which is experiencing high unemployment. The carbon-in-pulp (Clearing in Place) technology was used for the first time in Romania for precious metals recovery. This technology is capable of recovering gold and silver from tailings containing low contents of precious metals originating from previous production processes of ores. The major tailing dams planned to be mined were:
The precious metal treatment plant of Aurul was designed with a throughput capacity of 2.5 million tons per year. The whole project would produce approximately 1.6 tons of gold and 9.0 tons of silver per year.
The project was to have an effective life span of 10-12 years although this may increase with the addition of resources resulting from the recent exploration-joint-ventures concluded with the Romanian companies Remin and Cuart.
Meteorological conditions: on 30 January 2000, there was reportedly: 60-70 cm accumulated snow in the pond; 30-liters/square meter precipitation (solid and liquid); and a temperature which rose above 0 °C .
On 30 January 2000 at 11 p.m., the company informed EPA Baia Mare about "incidents" at the technical installations; a field observation showed that because of the high level of the waters long in the Aurul pond, the dam had overflowed and washed away a stretch around 25 meters and 2.5 meters deep. About 100,000 cubic meters of tailings water containing free cyanide and cyanide complexes were released and reached the surrounding areas and the Lapus river; the company shut down the activity and started to close and seal the breach.
On 31 January 2000, the company treated the spillage, which had decreased to 50 L/s with sodium hypochloride in order to neutralize the cyanide. The National Mining Company REMIN started the intake of the remaining water into its active tailing dam located close to the Aurul tailing dam.
On 31 January 2000, EPA Baia Mare established the County Commission for the Defense against Disasters under the coordination of the Prefect of the Maramures County; the water authorities started frequent monitoring of the water quality in the Lapus and Somes rivers. The authorities in Hungary and other downstream countries were informed.
On 1 February 2000, experts from the National Commission for the Safety of Dams arrived to diagnose the causes and suggest possible technical solutions.
On 2 February 2000, at 1:30 a.m., the spillage from the Aurul tailing dam was stopped, and the decontamination of the affected area, around 14 ha, starts; EPA Baia Mare explored legal action against Aurul Co. while in Satu Mare a first report on dead fish was recorded.
On 8 February 2000, experts from Romania and Hungary met in the field and after that in Satu Mare City for a technical evaluation of the impact; eight individual wells contaminated with cyanides were found in Bozanta Mare village.
On 10 February 2000, a meeting between the Hungarian Environment Minister Pal Pepo and Romanian Secretary of State for the Environment, Anton Vlad, takes place.
On 17 February 2000, Mrs. Margot Wallström, EU Commissioner for Environment, the Romanian Environment Minister Romica Tomescu, Mr. Fotion Fotiadis, Head of the European Commission delegation in Romania, and Hungarian Environment Minister Pal Pepo met and visited the site of the accident.
On 25 February 2000, the UNEP/OCHA mission started its assessment work in accordance with formal requests from the three affected countries.
The Maramures County, situated at the northern border of Romania with Ukraine, consists of the old "lands" of Maramures-Chioarul, Lapus and Baia Mare Depressions. The area is rich in gold, silver, lead, copper and salt and there is a long history of mining in the region.
There are seven key mining sites in the county, which are potential point sources of pollution. The facilities produce gold (Au), silver (Ag), lead (Pb), zinc (Zn) and manganese (Mn). Waste waters and materials from these facilities are stored in flotation ponds and tailing dams, a standard international practice. However, the quality and protection available to such dams can vary depending on the location and the technology applied. For example in Maramures County there are nineteen flotation ponds identified by the local Environmental Protection Agency (EPA). Twelve of these are abandoned and seven are active sites. One of these is the Novat settling pond of the Mining Company Baia Borsa where a dam broke on 10 March, 2000. Overall, the EPA lists 215 dams from mining operations on the territory of Maramures alone.
Other sources of potential pollution include metallurgical plants. A site of a lead smelter at Baia Mare has been used for 150 years and is reported as being heavily contaminated. Air and water contamination are also reported. Other industrial plants in the area include a copper smelter and sulfuric acid plant. Decades of such industrial activity with insufficient waste treatment, only now being introduced, have resulted in a high level of chronic pollution of the ground, water and air of the region.
Representatives of the local population expressed concern about nuisances several times. These included especially complaints about high dust loads. The cause of this nuisance has been identified as the erosion of particles from the above-mentioned tailings dams - including the Meda and Remin dams - and the dust and sulfur emissions to the air from the two metallurgical plants. Erosion could occur, because the dams have not been covered by vegetation. EPA Baia Mare confirmed that the dust concentration in the Baia Mare area exceeds the limits during the dryer periods and when the wind is blowing. No significant differences before and after Aurul started its operations have been recorded.
As regards environmental monitoring, the water and environmental authorities do not analyze sediments for contamination. Observation wells have been installed and are monitored for cyanide and heavy metals, on a monthly basis. The presence of cyanide and heavy metals can be detected but none of the compounds have shown increased concentrations since the Aurul operation started.
It should be remembered that Aurul is not the only company extracting precious metals with cyanide. Next to the Aurul plant tailing dam there is the National Mining Company REMIN that operates a process similar to the one used by Aurul. According to the Baia Mare EPA authorities, the Remin dams were not built in proper conditions and do not have preventive measures installed such as impermeabilisation or installations of water spreading against the dust blowing, etc. As a consequence, the soil and groundwater in this area have been polluted before.
The NGOs in Baia Mare reported that leakages have occurred in the pipeline system of the Aurul Company. EPA Baia Mare has confirmed two such accidents in 1999. A negative effect on the quality of the surface waters was not found.
A WHO report "Concern for Europe’s Tomorrow" identified Baia Mare as a "hot spot". The survey reported on a survey that found that lead (Pb) levels in the blood of adults living near the lead smelter averaged 0.523 mg/L compared with the WHO recommended limit of 0.2 mg/L. Children living near the plant had mean blood lead levels of 0.633 mg/L compared with the threshold of 0.10-0.15 mg/L now thought likely to be associated with detectable impairment of cognitive ability. The report concluded that the exposure of the population in Baia Mare to lead proved to be among the highest ever recorded.
It can be concluded that this area has a high level of potential and actual chronic contamination from a range of pollutants. The accident and any subsequent contamination that may have arisen from the Aurul plant needs to be seen in the broad context of long-term chronic pollution.
Baia Mare has a growing population and urban development with expansion restricted in some areas by the presence of old tailing ponds. In the early 1990s it was agreed that three such ponds should be cleaned and recovered to allow development to proceed. Residents are living within 50 m of highly toxic, potentially chronically leaking, waste sites that cause concern, especially in the dry months. There would be a clear environmental improvement from removing such waste sites.
Aurul was set up as a joint venture with Australian and Romanian partners following an international tender to clean up the sites. The design concept was that the waste would be transported away from the city (at least for two dams) where the remaining gold and silver in the tailings waste could be recovered using efficient modern technology that was not available when the original ponds were established.
The company described the process as follows:
"after removal of the precious metals, the tailings will be redeposited in a plastic lined dam which will provide a totally closed water circuit with zero discharge to the surrounding environment".
As per intention, the dam was therefore the most modern in the region, designed to be a major environmental improvement to the existing chronic polluting ponds. It was intended to be a safe and efficient method of meeting the requirements of the Romanian authorities and the Australian investors.
The local authorities required a feasibility study (produced by Lyco Resources, Australia in 1992), technical documentation for the Environmental Agreement designer (Mining Research Institute in Baia Mare in 1992) and an Environmental Impact Assessment (produced by the Engineering and Research Institute for Environment, Bucharest in 1993), from the company.
The company received fifteen permits before obtaining an Environmental Agreement to proceed. Permits were provided by a range of national and local authorities including the Public Service for Waters Baia Mare (permission number 3035/1992), State Inspection for Mineral Resources Baia Mare (no 290/1992), and the Prefecture of Maramures County (559/1993).
Further permits were received for an Environmental Operating Permit. These covered Water Management System Permit (96/1999), Agreement of the Local Council of Recea Village (1593/1999), Operational Protection Permit (813/1998) and four Public Health permits . Thus, the plant obtained all necessary permits under Romanian legislation. These permits took seven years to obtain before anything could be processed on the site. However, despite the Environmental Impact Assessment, Governmental permits and operating licenses, at 22:00 local time on 30 January 2000, a breach opened in the dam, allowing some 100,000m³ of tailings water to escape into the environment.
The tailings in the old Meda pond in Baia Mare were mixed with water to form a pumpable slurry. The pond contained some 4.43 million tons of flotation solid wastes. A pipeline carried the slurry to the processing plant where cyanide was added to the mixture. High concentrations of cyanide are used in the process to extract the gold and silver. The tailings fluid, after gold extraction, was pumped through a pipeline to the new pond some 6,5 km away from the city. (Cyanide concentration as derived from the Process Design Criteria for the gold extraction was approximately 700 mg/L; total concentration of cyanide pumped to the new pond was appr. 400 mg/L, containing free cyanide of about 120 mg/L.
The new pond, covering some 93 hectares and nearly 20 m high in its future final stage, was constructed on gently sloping terrain by forming a surrounding dam and a decant well in the pond centre to allow ponded liquid to be re-circulated to the plant. The entire pond was lined with a plastic membrane to prevent loss to the ground and drains were fitted in the dam wall to collect any seepage which was included in the decant water fed back to the processing plant. Thus, the system would be completely contained, with no loss to the surrounding environment as per the permit requirements. Observation wells were installed along the perimeter to monitor the success of the operation with regard to groundwater pollution.
For the wall of the tailings dam, the concept of "construction by operation" was chosen. The dam wall around the pond is built up progressively over a low "starter dam" using the coarse solids in the tailings fluid pumped to the site from the processing plant. The coarse solid fractions are sorted out (separated) by means of a number of hydrocyclones operated along the dam axis. The remaining water with the fine solids (slimes) is fed into the pond formed by the gradually growing dam. This method of tailings dam construction, which works well under certain circumstances, is very economical, as it avoids the use of borrowed material to form the embankment.
Safe dam formation using this technique requires control of the pond water level by an appropriate outlet structure, so that the slimes fed into the pond can form a "beach" inside the dam wall, in order to limit water percolation to an acceptable extent. The regular dam reinforcement and beach formation in line with the control of the pond’s water level are the most important elements in the safety of a tailings dam. The decant well is usually designed to regulate the pond water level to ensure that a large dry beach remains to provide the strength and integrity of the dam.
At the Aurul scheme, decant water from the pond was re-used in the remining of the old Meda deposits. This return water still had high cyanide levels, thus reducing the need for fresh cyanide addition at the processing plant.
In order to cope with the problem of storm run-off, the storage capacity of the pond (between operation water level and maximum allowable water level) was estimated and found sufficient for any event of extreme rain upto 118 mm.
There were some critical points in this approach and problems were encountered in the operation of the system:
Under Romanian governmental regulations on construction standards, under the Ministry of Public Works, the plant was self-assessed by Aurul as Category "C" (out of a … from "A" – structure of exceptional importance to "D" – structure of low importance). Therefore, the plant and pond, categorised under regulation as of "regular" importance, did not require any mandatory special surveillance and monitoring. It seems that as far as the local authorities were concerned, they expected to achieve an improvement in a long-standing environmental problem; the plant had received all the permits that were necessary, and all should have been going according to plans.
Although the above analysis of the spill at the Aurul plant had to be based on limited sources of documents and information, the interpretation of the circumstances leading to the dam failure seems clear. Important conclusions regarding design, approval, operation and surveillance aspects can be drawn from this analysis, with implications for all of the parties in a mining project. This is all the more true as this was a newly engineered pond system that failed under the circumstances, which in principle, could have been foreseeable.
The accident at the S.A. Aurul company was reported to the site management by a site worker at 22.00 on 30 January 2000. The Aurul plant stopped operations and the central and local authorities were notified. Sediments from a nearby tailing deposit were used to seal the breach, which the local authorities record as being partially closed by 01.30 a.m. on 31 January 2000. A controlled discharge of 40-50 L/s continued to leak from the dam, this discharge was neutralized with sodium hypochloride until the breach could be completely sealed. The south west side of the dam was reinforced using the same material.
The spill initially entered the Sasar river near Baia Mare, then flowed into the Lapus river before joining the Somes river which crosses the border with Hungary at Csenger. The Somes joins the Tisza river which flows through Hungary and into the FRY near Tiszasziget. Travelling at 2.1 km/h–2.4 km/h, the pollution took 14 days to reach the FRY, some 800 km away. The Tisza is a tributary of the Danube and the pollution flowed into the Danube upstream of Belgrade and continued for a further 1200 km at 2.4 km/h–2.9 km/h before entering the Black Sea. In total, some 2,000 km of the Danube catchment area wer affected by this spill.
On the basis of the characteristics of cyanide, the hydrological information of the river systems involved and the measurements of plume progress along the length of the river, it has been estimated that between 50-100 tons of cyanide (CN) was released into the river environment.
The 1992Convention on the Protection and use of Transboundary Watercourses and International Lakes prescribes that Riparian Parties shall inform each other without delay about any critical situation that may have transboundary impact. For the catchment area of the Danube river this convention has been concretized by the Convention on Cooperation for the Protection and Sustainable use of the Danube river (Danube River Protection Convention). According to this convention, as soon as a sudden increase of hazardous substances in the Danube river or in waters within its catchment area is identified, the Principal International Alert Centers in case of Accidental Pollution on the Danube River (PIACs, see Figure 5) should be informed immediately.
The cyanide spill took place on 30 January 2000 at about 22:00; at 22.30 p.m., the S.C Aurul S.A. informed the chief inspector of the local Environment Protection Agency Baia Mare (EPA); at 23.00, the Aurul company stopped all operations.
On 31 January, 2000, at 8.30 a.m., the EPA informed the Romanian Waters Authority (Water Resources Management System Baia Mare). After that, they warned the Tisza Branch of the Romanian Waters Authority Somes (Water Resources Management System Satu Mare); at 15.00 p.m., the Somes-Tisza Branch warned (by telephone) the Hungarian Water Authority and Environment Agency in Nyiregyhaza; at 17.15 p.m. the Somes-Tisza Branch transmitted an official bulletin by fax to the Hungarian Water Authority and Environment Agency in Nyiregyhaza, to the National Company of Romanian Waters in Bucharest and to the Department of Waters and Forests and Environment Protection in Bucharest.
The first warning from PIAC-08 in the Department of Waters and Forests and Environment Protection Bucharest to PIAC-05 in Budapest was sent on 31 January 2000 at 20:54. The warning was sent by fax and not by satellite transmission as planned, because the Romanian satellite transmission system did not work.
It must be noted that much time (about ten hours) was lost between the time the local Environment Protection Agency of Baia Mare received the notification of the spill from Aurul S.A. and the time the local Romanian Waters Authority (Water Resources Management Baia Mare) was informed. The reasons for this delay are not known. Thanks to the distance between Baia Mare and the Hungarian border, the Hungarian authorities received the warning early enough to take the measures necessary to ensure, for example, adequate drinking water supply to the population. However, the population in the vicinity of the plant could not be informed at the earliest desirable stage. The establishment of a good operational and prompt early warning system is essential.
In order to evaluate the information provided to the mission, the UN team visited local laboratories and discussed approaches and methods with local scientists to ensure that data could be compared and that quality control would be appropriate in each of the countries. The visits also proved very useful in obtaining background data on the regular situation in the different countries.
Cyanide is determined according to Romanian (STAS 7685/79) and Hungarian (MSZ 260-30) standard methods. The Yugoslavian standard method could not be made available to the UN team within the time of its visit. All three methods are geared to determining total cyanide, which includes free cyanide, easily accessible and complexed cyanide. Based on the method descriptions and the data shown to the UN team – also on quality control parameters - in Romania and Hungary, and the discussions with the Yugoslavian chemist, it can be concluded that with the methods applied comparable data according to international standards can be generated.
In all three countries the analysis of heavy metals is performed on filtered water samples by atomic absorption spectrometry (AAS) after digestion with nitric acid. Both free and complexed metals are included in the analysis. The team found the results of the analysis from the different countries to be comparable within the uncertainties based on different instrumentation and quality of the AAS instruments used in the laboratories.
The Romanian laboratories are not accredited, but ambitious quality management ensures the reliability of reported results. The Hungarian laboratories are accredited (with annual renewal), well equipped and are comparable to laboratories in other countries. The laboratory in FRY could not be visited during the mission.
In the past, intercalibration studies of the analytical laboratories applying their methods have been performed between Romania and Hungary and between Hungary and Yugoslavia at the sampling stations on the borders. Comparable and good results were obtained.
The UN team carried out sampling along the river system following the route of the spill. The team also included sampling sites upstream of the plant in an attempt to identify baseline contaminations in the area, which had not been affected by the spill. Thus, water samples from rivers and lakes (= surface waters), sediment and soil samples were taken. In addition, drinking water from deep well systems was also tested for potential impact and with the view to obtain baseline data. The samples were analyzed for physical parameters such as conductivity, turbidity, temperature, color, etc.; for chemical parameters with special emphasis on the determination of cyanide and heavy metals (such as lead, cadmium, zinc, manganese, copper, etc.), for possible toxic effects with in vitro tests in the laboratories, and to determine the presence of aquatic organisms.
Twenty-six sampling sites for water and sediments were visited. There were a total of 85 water samples, 48 sediment samples, and 12 soil samples taken from 25 stations and analyzed. Sites that were not affected by the spill, included a station at Lapus river upstream from the Aurul dam and six other stations at tributaries.
To study the potential impact on the drinking water supply in the three countries, samples were taken from 13 private wells, from 2 taps of private wells, from 2 public wells, from 7 waterworks wells and from 1 water treatment plant and analyzed for physical, biological and chemical parameters.
The sampling sites are shown in maps in Appendix 10.1.
The quality of the river system varies considerably with respect to the concentrations of heavy metals. There is historic chronic pollution, which inevitably affects the quality along stretches of the river. Inputs include point source industrial effluents, sewage effluents and agricultural runoff. The background data for heavy metals are summarized in Table 1.
Table 1: Historical water chemical data of the River Sasar, Lapus, Somes and Tisza in September 1992
|
River |
Sampling site |
rkm |
As |
Zn |
Cu |
Pb |
Cd |
Fe |
Mn |
|
mg/L |
mg/L |
mg/L |
mg/L |
mg/L |
mg/L |
mg/L |
|||
|
Sasar |
Baia Mare |
0.40 |
2.8 |
1.7 |
0.92 |
0.02 |
14.8 |
3.7 |
|
|
Lapus |
Busag |
0.42 |
3.4 |
2.2 |
0.38 |
0.02 |
14.0 |
2.1 |
|
|
Somes |
Ulmeni |
163 |
0.00 |
0.04 |
0.02 |
0.02 |
0.00 |
1.10 |
0.27 |
|
Somes |
Cicarlau |
123 |
0.36 |
1.5 |
1.6 |
0.32 |
0.01 |
7.3 |
1.2 |
|
Somes |
Satu Mare |
68 |
0.13 |
1.0 |
0.56 |
0.12 |
0.00 |
6.03 |
2.0 |
|
Somes |
Csenger |
48 |
0.01 |
0.25 |
0.06 |
0.01 |
0.00 |
1.6 |
0.96 |
|
Somes |
Tunyogmatoles |
3 |
0.01 |
0.26 |
0.01? |
0.02 |
0.00 |
2.3 |
0.87 |
|
Tisza |
Borzhava |
725 |
0.00 |
0.05 |
0.00 |
0.00 |
0.00 |
0.38 |
0.06 |
|
Tisza |
Aranyosapati |
669 |
0.01 |
0.28 |
0.06 |
0.02 |
0.00 |
2.7 |
0.51 |
These numbers can be compared with water quality criteria for the river Rhine, established by the European Commission. The limit concentrations are as shown in Table 2:
Table 2: Water quality criteria for the River Rhine as established by the European Commission
|
Substance |
Concentration |
Substance |
Concentration |
|
Cyanide (CN) |
0.025 mg/L |
Chromium (Cr ) |
0.025 mg/L |
|
Arsenic (As) |
0.005 mg/L |
Nickel (Ni) |
0.01 mg/L |
|
Lead (Pb) |
0.005 mg/L |
Mercury (Hg) |
0.0005 mg/L |
|
Cadmium (Cd) |
0.003 mg/L |
||
NB: In its last revision, the limit values for copper and zinc have been removed; there are also no limit values for manganese and iron.
From comparing the values in Table 1 with the quality goals in Table 2 it can be concluded that historically (reference year = 1992), at all locations in the rivers Sasar, Lapus, Somes and Tisza, the limit values are exceeded for arsenic (As; range: 0.01-0.42 mg/L) and for lead (Pb, range: 0.01-2.2 mg/L). The reported concentrations for these two heavy metals are 100- to 1,000-fold, respectively, above the quality goal concentrations of the river Rhine. For cadmium (Cd, range 0.00-0.02 mg/L), in the Lapus and Sasar rivers, the concentrations measured in 1992 are almost 10-fold above the Rhine quality criteria level.
The data in Table 1 also show that in the Baia Mare mining area, the concentrations for all heavy metals measured are higher than further downstream the river system. Nevertheless, overall the water data show that these rivers have been highly impacted with heavy metals for many years.
Authorities in Romania, Hungary, and Yugoslavia attempted to catch the plume and therefore intensified their sampling during the times when the plume was expected. As a consequence, the frequency of sampling was increased and the spectrum of chemicals monitored expanded.
EPA Baia Mare, Romania, provided the results for cyanide and copper measured in the water after the spill. The data are shown in Figure 6 for cyanide and in Figure 7 for heavy metals.
Figure 6: Graphical sketch of the maximum cyanide concentrations in the river basin (compiled by the EPA in Baia Mare)
As can be seen in Figure 6, the maximum cyanide concentrations measured by the Romanian authorities in the Somes river waters are lower than the concentrations determined by the Hungarian authorities further down the rivers by a factor of two or more. For example: on February 2 at 11:00 the Romanian measurements for cyanide gave 7.8 mg/L in the Somes river at Satu Mare, whereas the Hungarian measured 32.6 mg/L "one hour later" and 20 km downstream at Csenger.
The numbers in Figure 6 imply that, according to the Romanians, the cyanide content of the pollution reduced from 19.4 mg/L at the site of the plant to 7.8 mg/L at Satu Mare, 10 km from the border, while the information provided by Hungary to Romania reveals that at Csenger, on the Hungarian side of the border the peak concentration of cyanide in the plume reached 32.6 mg/L. The figures also show that the cyanide concentration at the Hungarian-Yugoslavian border had been reduced to 1.5 mg/L.
Figure 7: Concentrations of heavy metals in the river Somes at Cicarlau
As for the heavy metals, figure 7 shows that, according to the Romanians, the peak concentrations at Cicarlau, 50 km of the border, were 10.5 mg/L for copper (Cu), 0.419 mg/l for zinc (Zn), 0.16 mg/l for lead (Pb), 0.964 mg/l for iron (Fe) and 0.437 mg/l for manganese (Mn). It should be noted that the peak concentrations of the various heavy metals were recorded at different dates, over an interval of three weeks.
Figure 8: Cyanide pollution across Hungary giving mean concentrations (Hungarian Ministry of the Environment)
From Hungary, an abundance of data was generated. Data covering the cyanide pollution in Hungary are displayed in Figure 8. As can be seen from the graph, the highest CN average value measured was at the Hungarian-Romanian border in Csenger and amounted to almost 18 mg/L.
NB: It should be noted that the figure of 32.6 mg/L contained in Figure 6 has also appeared in a preliminary evaluation report of the Hungarian Ministry for Environment and was described as "concentration of the total dissolved cyanide".
Figure 9: Concentrations of copper in Hungarian rivers
Figure 10: Concentrations of zinc in Hungarian Rivers
Information concerning heavy metals is contained in Figure 9 (copper) and Figure 10 (zinc). No information was received on other heavy metals. The results show that the peak concentration of dissolved copper was 18 mg/L at the Hungarian-Romanian border and had been reduced to 2.5 mg/L at Csongrad, 59 km from the Hungarian-Yugoslavian border. For zinc the figures are 0.962 mg/L at Csenger down to around 0.05 mg/L.
NB: It should be noted that the peak concentration figures for copper and zinc provided by the Romanian side at the Romanian-Hungarian border were much lower.
The UN team has not been able to resolve the discrepancy between the measurement results provided by Romania and those provided by Hungary. As mentioned in section 6.1, the laboratories in both countries were found to use comparable methods and produce reliable results. Reasons for the differences in the cyanide concentrations may be due to the sampling location, the fact that there was ice on the river and that, due to longer time intervals for sample taking, the Romanian sampling missed the peak concentration.
According to information received from the Federal Republic of Yugoslavia, the peak cyanide concentration 8 km upstream of the entry of the Tisza river in the Danube amounted to 2.28 mg/L. Once in the Danube river, the cyanide plume followed the left bank of the river, and 5 km downstream of Iron Gate I, the maximum cyanide concentration had been reduced to 0.07 mg/L.
With respect to heavy metals, the Yugoslavian report indicates that in all samples taken in relation to the passing of the plume, concentrations of copper and iron were 13-20 times higher than maximum permissible levels (MPL). The concentration of manganese also exceeded MPL.
Water samples from surface water were taken by three teams from the 21 sampling stations11. Typically, the water sampling locations were identical with those for the sediment sampling. The UN sampling took place about three weeks after the plume had passed and thus, could not validate ony of the results obtained by the Romanian, Hungarian or Yugoslavian experts. The concentrations of free cyanide and easily accessible cyanide were analysed. Also, the concentrations of thirteen heavy metals were analysed in four laboratories in the Czech Republic, Finland, Germany and Switzerland, although all these metals were not analysed in every sample.
The UN analyses of water samples from the pond approximately four weeks after the spill, gave high concentrations of cyanide, copper, iron, manganese and zinc. Rather low concentrations were detected for lead, cadmium, and mercury.
In the pond of the Aurul plant, concentrations of free cyanide were very high with concentrations between 66 mg/L and 81 mg/L. Very high concentrations of heavy metals were detected as well for copper (total concentration 412.3 mg/L), iron (total concentration 31.3 mg/L), manganese (total concentration 18.0 mg/L) and zinc (total concentration 14.5 mg/L).
In the Lapus River, upstream the Sasar River confluence, surface water samples taken by the UN team on 26-27 February near Bozanta Mare gave concentrations of free cyanide of 0.010 mg/L. Downstream the confluence, the cyanide concentration was 0.88 mg/L. These results confirm the statement by the Baia Mare EPA that the Sasar River is known as "Dead River". Also the results from the Somes River upstream of the Lapus River confluence (0.004 mg free cyanide/L) and downstream (0.035 mg free cyanide/L) indicate a continuous cyanide pollution of the river systems. According to the Romanian Standard (STAS 4706-88), the permissible concentration of total cyanide in surface water is 0.01 mg/L.
The cyanide was biologically degraded and diluted as it moved along the length of the river.
In the surface water samples in Hungary, free cyanide concentrations of 0.014 mg/L for the Somes river at Csenger and 0.003 mg/L for Tisza river at Solnok were determined by the UN team. In the well water samples, no free cyanidie could be detected.
The UNEP/UNCHS (Habitat) Balkans Task Force took water samples from the river Danube in FRY on 15-17 February, 2000 from a stretch between Pancevo to the Iron Gate (Danube river kilometers 1154 to 943). The samples had cyanide concentrations from 0.008 mg/L to 0.117 mg/L and copper concentrations from 0.020 to 0.096 mg/L. The UN team took a sample from the Tisza river near Becej. The concentration of free cyanide was below the detection limit of 0.01 mg/L.
The UN Team was able to record the plume of cyanide contamination in the Danube delta (at a location named CHEATAL-IZMAIL by local authorities in Romania). Measurements taken between 26 February and 28 February gave a maximum concentration of 0.058 mg/L.
For visual comparison, all the results for copper, lead and zinc are presented in figure 11 through figure 13 for the sampling stations visited by the UN team.
For all heavy metals, the figures show a dramatic increase in the water concentrations at sampling station 6, which is the closest to the site of the accident. At site 5, which is located upstream the Aurul plant, the contamination with heavy metals is much lower. With increasing distance from the plant, the concentrations of the heavy metals (and the cyanides as shown above) decrease rapidly. It can also be seen that some of the tributaries, especially the Maros river in Hungary add to the load of lead and to a lesser extent of copper and zinc in the Tisza and Danube rivers. The concentrations measured by the UN team compare relatively well with the historic data as shown in Table 1 (on page 21) and the elevated levels found may be caused by other industrial point sources and diffuse emissions along the rivers and their tributaries.
The water concentrations measured in some of these spot check samples do in some cases exceed recommended limit values (see below). The present data cannot distinguish if the heavy metal concentrations measured in river waters originate from direct discharges into the water are due to remobilisation from sediments (for sediment data, see next section).
Figure 11: Results of the water analyses for copper
Figure 12: Results of the water analyses for lead
Figure 13: Results of water analyses
As was done for the background contamination, the data generated by the UN team will be compared with the European Union water quality criteria:
The concentrations measured by the UN team in the Danube delta at Cheatal-Izmail show that before and after the plume, the lead (Pb) concentrations between 0.004 and 0.007 mg/L are above these quality criteria (0.005 mg/L). The cyanide concentration of 0.058 mg/L in the plume at the delta was clearly above the quality criteria value of 0.025 mg/L. The concentrations of the other heavy metals were comparable to those reported for the river Rhine in the year 1990 (with Cu and Mn being above the mean concentrations from the Rhine).
In FRY, the UN team found 0.006 mg/L of lead (Pb) in the Tisza river above the confluence with the Danube. This concentration is above the water quality value for the Rhine (= 0.005 mg/L) and also higher than the concentrations measured in the Danube where the concentrations ranged between 0.001-0.004 mg/L. All other data showed concentrations comparable to those in the Rhine in the year 1990. The concentrations of manganese (Mn) and iron (Fe) in the river Tisza were found to be slightly higher than those in the Rhine, while the concentration of zinc (Zn) in the unpolluted Danube was also slightly higher than the averages from the Rhine.
In Hungary, for lead (Pb) the Rhine quality criteria were found to be slightly exceeded in the Tisza (0.006-0.008 mg/L) whereas in the Maros river, which was not affected by the spill, the concentration of 0.022 mg/L was found to be more than 4 times above the criteria level. Compared to the data from the Rhine, the concentrations for Cu, Mn, and Fe were found to be much higher in the rivers Tisza and Maros.
During the spill, the Romanian authorities investigated the ecosystem. The results indicated a total loss of the phyto- and zooplankton communities in the Somes river during the plume. Fortunately, recovery was relatively quick due to the inflow of unaffected water from upstream (as an example, see Figure 14).
Figure 14: Evolution of phytoplankton and zooplancton in the Satu Mare region (Romania)
The Hungarian laboratories (Upper Tisza Regional Environmental Inspectorate Laboratory and Lower Tisza Regional Environmental Inspectorate Laboratory) tested the river Tisza water during the cyanide plume with biological tests (Daphnia and fish test). The water was acutely toxic to fish and Daphnia magna during the plume (100 % mortality during the maximum cyanide concentrations). Tests from 1999 performed routinely show a mortality of 0-30 % in the Daphnia test. This result indicates that under normal conditions, the water is also toxic to Daphnia.
Hungarian authorities also provided data on the phyto- and zooplankton communities in the river Tisza water during the cyanide plume. The effect of the plume on the communities was similar to that reported by the Romanian authorities. At the time of the plume, the number of species and specimens in the water samples, if any, was very low, but recovery was quick (within a few days) due to the cleaner water from the upstream river and tributaries.
In summary, the results of the UN mission sampling and its discussions with local experts in the three countries reveal that the ecological state of the benthic organisms in the middle and lower Tisza region in Hungary and Yugoslavia were not destroyed by the cyanide spill in a catastrophic manner. The situation in the upper Tisza is more complex. It should be noted that the Tisza does not have a totally natural ecosystem and has been influenced by pollution or embankment and dam constructions which all have impacts on water level, bottom character, chemical parameters, etc. The influence has resulted in changes of bottom invertebrates and fish assemblages. Deterioration of river ecosystems as shown in the Tisza river is a problem of many European rivers and in the Tisza, the ecosystem had been deteriorated already before the cyanide spill.
Sediment samples were taken at 21 stations starting from the Aurul plant, where the accident happened, to the Danube at Vinca in the Federal Republic of Yugoslavia. While some sampling stations were visited by one team only, others were visited by two or all three teams. Since two of the teams also took parallel samples from some stations, up to five results can be reported from one sampling location. To complement the sediment sampling, three additional samples were taken from the tailings in the Aurul dam. Two of these were from the solids on the bottom of the ditch between the dam and Lapus river, and 14 from the soil on the field along the ditch. The latter area was flooded during the spill and may thus have been impacted with heavy metal-laden sediments from the pond.
The analyses showed that the Lapus river had high concentrations of heavy metals upstream and downstream of the Aurul plant. For example, the concentrations in sediments of copper, lead and zinc for all sampling stations visited are illustrated in Figure 15, Figure 16 and Figure 17, respectively. The sampling stations that are located outside the influence of the spill (= unaffected stations) are shown in the filled squares in the three figures. It can be noted that for all three heavy metals, the contamination rises dramatically downstream of the Aurul dam (Sampling station 6). This is a clear indicator that contamination from the pond has been deposited in the nearby sediment of the Lapus river. It can also be noted that the contamination in the sediments with heavy metals drops rapidly with increased distance from the source. At sampling station 7, in the Somes river at Cicarlau, the sediment concentrations were much lower and in the same range as for all other sampling locations.
Figure 15: Results of the sediment analyses for copper
Figure 16: Results of the sediment analyses for lead
Figure 17: Results of the sediment analyses for zinc
The data also clearly show that tributaries, which have not been affected by the spill, have high concentrations of heavy metals in their sediments. Thus, there are some local "hot spots" downstream in the Tisza. These elevated concentrations are indicative of anthropogenic inputs from industrial point sources or diffuse emissions, e.g. agriculture and domestic sewage, which discharge into the Lapus ® Tisza ® Danube river system over a long period of time .
The UN data also show variability. It can be seen that for some sampling stations the results differ by a factor of two or more. Such difference is normal for field samples even if standardized methods are applied. There are always differences between laboratories (e.g. using different digestions and equipment), but the variability within the sample itself (e.g. sampling location and depth, grain size distribution, organic carbon content, etc.) may be much greater than the variability among laboratories and analytical methods.
Although there are no background data available on heavy metal concentration in sediments for Romania, the UN team data strongly suggest that the spill caused a rather local contamination of the sediments restricted to the immediate environment of the broken dam. A relatively high existing contamination with heavy metals can be assumed due to past industrial activities.
To put the data into perspective, the results of the UN team can be compared with the limit values for heavy metals laid down in the Canadian sediment quality guidelines. The Canadian guidelines defined so-called probable effect levels (PELs) for some heavy metals as follows:
|
Cu |
197 mg/kg |
Cd |
3.5 mg/kg |
||
|
Pb |
91.3 mg/kg |
Zn |
315 mg/kg |
||
|
Hg |
0.486 mg/kg |
As |
17 mg/kg |
According to these guideline values, the concentrations of heavy metals in the river Lapus are high or extremely high. At the site of the spill, the PELs are exceeded manyfold for all heavy metals. The concentrations for lead, zinc, and cadmium (not shown here) measured upstream and downstream of Baia Mare exceed the PELs and thus are at a level where toxic effects in benthic organisms are likely to occur. In the river Somes downstream Baia Mare, the UN team measured concentrations are in the range of the PELs. For zinc and arsenic, the concentrations were above PELs even in the sediments of the river Tisza and thus, toxic effects on the aquatic ecosystem cannot be excluded.
Toxicity tests of river sediments were performed extracting samples with sodium chloride (NaCl) or methanol and testing the extract with bacterial tests. In addition, whole sediment toxicity testing procedures using the same microbes were applied. In summary, all tests with extracts showed high toxicity in the samples from the Aurul reservoir and from the Lapus river after confluence with the Sasar river. Medium toxicity was measured in sediment from the Somes river upstream of the confluence with the Lapus river and in the "fine sediment" fraction from the Tisza river at Aranyosapati. All other samples showed low or no toxicity.
As expected, the toxicity assays performed on whole sediments gave more patchy results: some indicated high toxicity whereas others showed no toxicity. An increased toxicity due to the spill can be shown in the sediments only in areas relatively close to the release.
Samples of aquatic sediments collected in the framework of the UN mission showed high toxicity in the Aurul reservoir and its surrounding. The results indicate that the pollution was rather limited in extent and thus, toxic effects on the aquatic ecosystem due to the presence of contaminated sediments from the spill at the Aurul plant, may not have moved far downstream.
A fast decline of the toxicity shows more limited adverse impact on the ecosystem from the "Aurul pollution" in comparison to surface water. However, an additional sediment toxicity survey with other bioassays should be made to verify the present condition of the soft sediments in different areas.
Sediment surveys for heavy metal analyses should be included in future monitoring programs in all three countries as the data have shown that the sediment quality is already at a stage where adverse effects on the aquatic ecosystem may occur. In addition, it is essential that background data be generated for the Baia Mare area to assess the impact from the mining and other industries.
The toxicity tests performed by the UN team could only give a snapshot of the current situation with a very limited number of samples. There is a need to better characterize the potency of the sediments to cause toxic effects. This also means that standardization and harmonization of bioassays and test methods used are necessary.
A summary of the heavy metal analyses of the soil samples taken from the partly flooded area between the Aurul dam and the Lapus river are presented in Figure 18. It can be noted that the contamination with lead is high in the flooded area but also very high in the four reference samples. These samples were taken from the field, which was not affected by the spill. These data confirmed the findings by the Baia Mare EPA that there was a high background contamination in the Baia Mare area and in the neighborhood of the Aurul dam.
Figure 18: Results for heavy metals in the sediments at the Aurul tailing dam
The evaluation of the drinking water situation goes beyond an interpretation of the drinking water analyses and thus includes some river water analyses as well as an analysis of the water at the pond where the accident happened.
On 26 February, 2000, cyanide concentrations in the well water was below the detection limit but the concentrations of cadmium, copper, manganese and iron were higher than the admissible values based on the Romanian legislation.
These private wells are shallow, in hydraulic connection with the river and highly vulnerable. The Aurul pond is in the catchment area of the wells. The wells were affected by the spill as was shown through analyses of 10February, 2000, when 0.785 g/L were measured for cyanide. Further, high concentrations of nitrates, ammonia and ortho-phosphate indicate an impact from human activities on the groundwater.
There is usually no water monitoring of these private wells.
The long-term effects of the mining activities on the public health in Bozanta Mare (chronic exposure to cyanide and heavy metals) could be very important.
Along the Somes river downstream of the Aurul operations, no cyanide could be detected in the UN samples. The concentrations of copper, manganese, iron and zinc (as indicators for the pond) were also very low.
Except for Satu Mare and probably Dara, these wells are shallow and vulnerable to surface pollution. However, a hydraulic connection with the Somes river could not be confirmed.
According to the local experts there is no groundwater monitoring except in Satu Mare.
The long-term effects of the mining activities for humans depending on these wells for drinking water should be small.
According to the analyses at the team’s disposal, the Hungarian public water supply systems were not endangered by the cyanide pollution. Neither cyanide nor heavy metals were found in the water of deep wells. High concentrations of ammonia, manganese, iron and a lack in oxygen are characteristic of deep aquifers but not the result of surface water pollution.
The surface water treatment plant in Szolnok was stopped during the plume of the pollution. The analyses of treated water during the accident showed that the cyanide concentrations were below the Hungarian standards (0.1 mg/L).
In general, it can be concluded that in Hungary, the deep wells are well protected against surface pollution. There is probably no hydraulic connection between the river Tisza and the deep groundwater.
In the surface water treatment plant of Szolnok, a stringent monitoring program of the raw water is in place for the protection of consumers.
The monitoring of the water quality in the visited water companies conforms with the Hungarian legislation.
According to the team's information, there is no long-term effect of the mining accident on consumers health through drinking water. The team cannot conclude on the situation for private wells along the Tisza river as these were not visited during the mission.
According to the analyses at the disposal of the UN team, the Becej public water supply system and probably the two assessed private wells were not affected by the cyanide pollution. Cyanide and heavy metals could not be detected in the water samples. High contents of ammonium, iron and a lack in oxygen are characteristic of deep aquifers but not the result of surface water pollution.
In Becej, the vulnerability of the deep wells is very low. There is probably no hydraulic connection between the river Tisza and the deep groundwater.
The monitoring of the water quality at the Becej water company complies with Yugoslavian legislation. The two assessed private wells are normally not monitored for any contamination.
According to the UN team’s information, there is no long-term effect from the mining accident on consumers health from Becej drinking water. It should be noted that other water companies and private wells along the Tisza river were not visited during the UN mission.
According to the Romanian authorities, their regional environment and water authorities, after having been informed of the dam break at the Baia Mare S.C. Aural S.A. on 31 January 2000, immediately undertook to:
The same day the Romanian Environment and Water Authorities informed all downstream countries (Hungary, Yugoslavia, Bulgaria, Moldova, Ukraine), using the Principal International Alert Centre in case of Accidental Pollution on the Danube River (PIAC 08), as well as the Secretariat of the International Commission for the Protection of the Danube River.
On 8 February 2000 a special meeting of the Water Quality Sub-Commission of the Romanian-Hungarian Hydrotechnica Joint Commission was organized, to inspect the accident site and exchange technical information. On 10 February a Ministerial level meeting between the two countries took place where an Aide Memoire was signed on co-operation and measures to be taken. On 14 February 2000 a joint visit to the site of the accident was organized.
The Hungarian Authorities confirmed that "the Romanian Environmental and Water Authorities continuously informed the Hungarian Authorities about the event and the degree of pollution".
The early information allowed the Hungarian authorities to inform and alert all regional and local Environment and Water Management Authorities in a timely manner and, through them, take the necessary measures and achieve minimization of the direct threat and impact of the spill which entered Hungarian territory late 1 February 2000.
The Hungarian authorities stated that measures taken by the Hungarian side included warnings to the public, operations at dams and ponds to protect aquifers, temporary closure of the Kiskore dam to increase the water level and thus the dilution of the cyanide spill, and temporary closure of the water intake from the Tisza river to the town of Szolnok. A sampling and assessment programme was started immediately in order to measure the concentration and follow the transport of pollutants down the Tisza river.
The Federal Institute of Hydrometeorology of Yugoslavia received the first information on the accident unofficially in the afternoon on 2 February and succeeded in establishing contact with the Hungarians on 3 February (with the Water Research Centre VITUKI in Budapest). On 3 February official information arrived from Hungary, including forecast of the arrival time and the characteristics and concentration of the spill.
The Serbian Republic Institute of Hydrometeorology was informed by the Federal Institute on 3 February and initiated the monitoring of the initial situation and the spill from 10 February onwards (every 2-3 hours) and arranged for regular communications with the Hungarian side. The Serbian Republic Institute informed the responsible Yugoslavian authorities (health, agriculture) on the imminent acute situation. Co-operation with Hungary was good during the spill event, within the framework of the bilateral Subcommission for waters. Co-operation was intensified through daily contacts and three joint samplings taken at the standard sampling site on the border.
On 10 February, the Serbian Republican Water Management Company sent an official announcement about the pollution plume to all water management companies along the Tisza with the request to inform all Tisza water consumers and users to stop the operation of all water supply facilities. Hydraulic constructions (gates) were operated in order to prevent the enlargement of the effect of the spill to the side branches and canals along the river. Collection of dead fish was carried out by a large number of people, dumping the fish into sanitary holes in the ground following the advice provided by the Ministry of Agriculture and Fisheries. An announcement on prohibition of fishing and fish trading was made by the Ministry of Agriculture, Forestry and Water Management of the Republic of Serbia.
The Ministry of Health of the Republic of Serbia undertook a number of preventive measures to protect public health in the affected areas, including the closing of the Belgrade water intake.
The UN assessment team had many organised and informal contacts and exchanges with representatives of NGOs (in particular WWF, ECO-Carpatia, Tisza Platform, and UNEP National Committees in Hungary and Romania), local interest groups and other representatives of the population. Organized meetings took place in Budapest, Bucharest, Baia Mara, Satu Mare, Csenger, Kiskore, and Szeged. The team also met with representatives of the population of Bozanta Mare at the site of the accident and with representatives of the Regional Environmental Centre (REC) in Belgrade.
Representatives of the local community in the Baia Mare district expressed concern over the location of the mining and related industries. Soil and groundwater had been polluted before and pipes transporting tailings had broken on several occasions, with water containing cyanides flowing outside the industrial sites. Questions were also raised about the dust problems in summer and the potential delayed damage from the accident itself: wells which were now clean could become polluted later.
NGOs from Baia Mare district worried about the longer term implications for tourist development in the region and pointed to other economic consequences of this and other forms of pollutions whereby farmers would not be able to sell their products (eggs, milk, crops) due to fear of contamination. They had repeatedly complained about the lack of sufficient information about the pollution, from the media and even the local authorities. Representatives of the local authorities stressed the need for improved contacts and communication between the local and central administrations.
The call for more extensive, objective and reliable information about the spill, its composition and potential impact, including the characteristics of the dangerous substances involved, was repeated in almost every meeting and exchange with NGOs and other representatives of the local population downstream along the Tisza river course. Representatives from agricultural and fisheries branches stressed the fact that there were direct as well as indirect effects of the spill: in addition to the direct toxicological effects, the image of products from the region had suffered and thereby complicated marketing and export. Correct and constructive information could help to mitigate the latter effects.
A representative of a Cooperative of fishermen along 100 km reach of the Tisza river said the Cooperative had 61 members (professional fishermen) with an additional 15 employees. These people suffered from direct and indirect effects of the spill, including psychological effects. In addition to the professional fishermen there were approximately 30,000 recreational fishermen in the region. A society of such people had 12,000 members. Most of them had already paid their annual licence and the compensation to them had to be clarified.
An organization for teachers pointed out that Tisza is very sensitive to effects caused by human activities and therefore much attention should be given to education and information of the people to prevent ecological and other damage from such activities. A number of specialized organizations and NGOs offered and expressed the wish to be more closely involved in assessing and monitoring the effects of this and other forms of pollution on flora and fauna, pointing to their networks and expertise.
Representatives of NGOs pointed to the interrelationship between the social, economic and ecological concerns and perspectives in the Tisza/Danube river basin and the prevailing constraints (insufficient resources, co-operation, communication) in providing solutions to the individual problems posed. They stressed the need for increased international co-operation and support for a holistic river basin management approach.
According to official Romanian sources the cyanide spill caused an interruption in the water supply in 24 localities, inconvenience to citizens, and supplementary costs in the sanitary field and in industry by interruption of the production process. The amount of dead fish reported was very small and the phytoplankton and zooplankton in the Somes river "was regenerated in 16 days to the proportion of 60 %".
The Hungarian Authorities completed an echo sounder survey along the Tisza river to detect the remaining quantity of consumable fish and to estimate the quantity of dead fish. The result gives an estimated amount of dead consumable fish of 1240 tons.
According to the Yugoslavian Authorities a large amount of dead fish appeared in the Yugoslavian part of the Tisza river. No major fish kills were reported from the Danube.
The estimates of the degree of biological damage and recovery, given by the authorities and experts are very variable, both in Hungary and Yugoslavia. The assessment work is ongoing and the monitoring programmes have been or are being put in place.
- tailings dams at operating mines are under continuous construction, as solid material and effluent (plus natural inflow due to precipitation) are added. Besides safe control of pond water volume under storm runoff conditions, the safety of the dam is mainly due to a sound balance between dam height and pond water level. In the case of the new Aurul pond at Baia Mare, the flows of solids and waters were out of balance with the increase of the storage capacity of the pond, as the process of dam construction could not keep up with the rise in the reservoir water level. The climatic conditions of the winter season aggravated the situation and led to an uncontrolled rise of pond level resulting in an overflow of the dam.
- The company responded by repairing the breach using borrow material from nearby, and by adding sodium hypochlorite to the overflow (and to the area flooded by the spill). A large volume of heavily contaminated effluent nevertheless escaped before the breach could be closed.
- There were no provisions for coping with situations of a rise of pond water level due to uncontrollable input into the reservoir system.
1. A re-assessment should be made of the relationship between environmental "benefits and risks" of the mining scheme of the Aurul S.A. company. In particular, a risk assessment study should be carried out of the entire system of remining the old tailings. This study should complement the Environmental Impact Assessment and the data contained therein.
2. In this connection, special emphasis should be given to the following aspects:
3. An inventory and risk assessment study should be made of all mining and related industries in the Maramures region, including abandoned sites, as a basis for better accident prevention and improved emergency preparedness and response measures.
4. In order to ensure prompt early warning and response, the existing on and off-site contingency plans should be revised with the relevant partners in line with Article 8 and Annex VII of the UN/ECE Convention on the Transboundary Effects of Industrial Accidents. Special attention should be paid to the possibility of a dam failure. The organizational roles and responsibilities off-site for dealing with a dam breach and the ensuing water pollution should be clarified. The plans should be practical, targeted to the site and fully accessible by workers and local stakeholders. The APELL process (Awareness and Preparedness for Emergencies at Local Level) as developed by UNEP can be a useful model on which to base such a review.
5. Romania should acceed to the UN/ECE Convention on Transboundary Effects of Industrial Accidents.
- further analysis of the composition of the sediments in the new pond at Aurul to determine the amount and types of cyanide present;
7. In the light of a number of earlier accidents with tailing dams, it is advisable to review construction concepts and operation procedures related to enterprises using such dams, including concepts of secondary security or retention of spills at dams containing toxic effluents or other liquids. Also, more attention should be paid to better integrating the construction and operational aspects of the design.
8. With respect to enterprises using cyanide, special attention is needed for emergency preparedness, emergency response and public communication measures (the APELL process), as well as special monitoring and inspection by the authorities.
9. Process water ponds should, wherever possible, be reduced in quantity and to sizes which can be handled in emergencies. They should have retention systems for overflow or for accidents resulting from a break of the dam.
10. In the Maramures area, consideration should be given to changing drinking water supply systems for private households to public / collective systems.
11. Urgent immediate action, as a basis for long-term drinking water improvements, should include:
12. Both in the case of acute emergency and with respect to longer term impact, much can and should be done to raise awareness and inform the local population along the Somes, Tisza, Danube rivers and in the catchment area as a whole, concerning the characteristics and potential risks involved at the mining and other industrial activities upstream. Unnecessary concerns and potential economic losses can be avoided with well informed local communities. Communications channels between the respective central government, the district and local authorities, should be optimized and NGOs and other interest groups, especially at the local level, should be mobilized and assist in informing the population and in providing replies to their concerns. The APELL process (Awareness and Preparedness for Emergencies at Local Level) would be an appropriate process on which to base such a programme.
13. There is a strong need for a broad, longer term environmental management plan and sustainable development strategy for both the Maramures region in Romania and the entire catchment area of the Tisza river; a strategy which would address, inter alia, the mining and related industries, other economic activities (such as tourism and fishing), biological diversity requirements, and social needs and imperatives.
14. The UN mission did not address the question of liability and compensation related to the spill and its consequences. The issue of liability and compensation would be easier settled if there were an international regime. Support should be given to the proposal to develop a protocol on liability and compensation on accidents with transboundary impact, to the UN/ECE Convention on the Protection and Use of Transboundary Watercourses and International Lakes and the UN/ECE Convention on the Transboundary Effects of Industrial Accidents.
15. UNEP and other relevant international organizations should pay special attention to promoting:
16. UNEP and its partners should also continue the dialogue with the mining industry to review design and operation codes, and promote a review and consultations on governmental approval permits and inspection procedures related to mining operations.
17. The Disaster Response Branch of OCHA and its Joint UNEP/OCHA Environment Unit should take appropriate steps to further develop the application of the concept of the UN Disaster Assessment and Coordination (UNDAC) to various environmental emergencies, including large-scale spills of mining tailings. The establishment of a small team of associated environmental experts should also be considered.
MEMBERS OF UNEP/OCHA ASSESSMENT MISSION: ROMANIA-HUNGARY-YUGOSLAVIA
UNEP
Mr. Frits Schlingemann, UNEP Mission Team Leader
Director Regional Office for Europe, Geneva
Dr. Heidelore Fiedler, UNEP Scientific Coordinator Chemicals
Mr. Anders Renlund, Spokesperson/Press Officer
OCHA/UNDAC
Mr. Sergio Piazzi
UNDAC Team Leader
Head European/NIS Desk
Disaster Response Branch
Geneva
Mr. Gerard le Claire
State of Jersey Planning and
Environmental Department
Mr. Joe Bishop
Emergency Management
Consultant
Mr. Martin Perez
Secretary, Disaster Response Branch
UN/ECE
Dr. Martin Schiess, UN/ECE Representative
Dr. Josef Klinger, DVGW Technologiezentrum Wasser, Germany
EUROPEAN COMMISSION
Mr. Cesar Niculescu, Task Manager, EC Delegation in Romania
Ms. Krisztina Palla, Environmental Project Manager, EC Delegation in Hungary
EXPERTS
DISASTER MANAGEMENT/RISK ASSESSMENT
Dr. Martin Schiess, Deputy Head Section, Agency for Environment, Switzerland
Dr. Nyström Magnus, Finnish Environment Institute, Chemicals Division, Finland
ECOTOXICOLOGY
Dr. Premysl Soldan, Water Research Institute, Czech Republic
CHEMISTRY
Dr. Jussi Kukkonen, University of Joensuu, Finland
Dr. Peter A. Solyom, Royal Institute of Technology, Sweden
Dr. Christiane Bettin, THW, Germany
Dr. Basil Al Naqib, THW, Germany
Dr. Armin Spühler, AC-Laboratorium Spiez, Switzerland
Dr. Michal Pavonic, Water Research Institute, Czech Republic
Dr. Michael Ratzenhofer, Austria
DAM SAFETY
Professor Dr. Ing. Josef Brauns, Universität Karlsruhe, Tailings Dams, Germany
Dr. Jan Boucek, Water Research Institute, Czech Republic
Dr. Lars Jonas Henriksson Fejes, Swedish Environment Institute, Sweden
Dr. Karl Wachter, Danube Expert, Austria
Dr. Jiri Kokes, Water Research Institute, Czech Republic
HYDROLOGY/HYDROGEOLOGY
Dr. Juha Sarkkula, Finnish Environment Institute, Finland
Dr. Patrik Adatte, Swiss Disaster Relief, Switzerland
Dr. Svein Stene-Johansen, Norwegian Institute for Water Research, Norway
Pollutants that have been released through the accident at the Aurul S.A. gold and silver operations included cyanide and heavy metals. These two classes of compounds exhibit very different chemical, environmental, and toxicological behaviour and thus, the short-term and long-term effects and impacts are not the same. The major difference between these two classes is that cyanide is acutely toxic but does not stay for a long time in the environment and does not accumulate in sediments or organisms (including humans) whereas the heavy metals do not break down and are bio-accumulative; their most serious toxic effects are due to long-term and chronic exposures.
Cyanide
Cyanide compounds are widely used by the mining industry for the extraction of precious and non-precious metals from rock. The cyanide attaches to particles of gold to form a water-soluble, gold-cyanide compound from which the gold can later be recovered. Cyanide is used in a similar manner to extract silver from ores. In the extraction of non-precious metals, such as copper, nickel, cobalt, and molybdenum, cyanide is used in the milling and concentration processes to separate the desirable metals from the wastes. Consequently, cyanide and related compounds often are contained in discarded mine wastes.
The general term "cyanide" refers to various compounds having the chemical group CN, that is, one single atom of carbon (C) and one single atom of nitrogen (N). Several plants, some soil bacteria, and several species of invertebrate organisms produce natural cyanide and related compounds. Nevertheless, cyanide compounds are seldom present in uncontaminated waters in measurable concentrations.
Cyanide readily combines with most major and trace metals, a property that makes it useful in extracting metals from ores. Cyanide also tends to react readily with many other chemical elements, producing a wide variety of toxic, cyanide-related compounds. And because cyanide is carbon based—an organic compound—it reacts readily with other carbon-based matter, including living organisms.
Fish are approximately one thousand times more sensitive to cyanide than are humans. Dose levels as low as 0.03 mg/L HCN can be ultimately fatal to sensitive species, while 0.2 mg/L is lethal to most species. In each case, levels less than lethal do provoke physiological and pathological responses that reduce swimming ability, interfere with reproductive capacity and can lead to seriously deformed offspring, and also leave fish more vulnerable to predators (see Table A 1).
Table A 1: Acute, chronic and sublethal toxicity of cyanides to fish (Ingles, 1982).
|
LETHAL EFFECTS |
SUBLETHAL EFFECTS |
||||
|
ACUTE mg/L CN |
CHRONIC (Juniors/Adults) mg/L CN |
Activity or Organ Affected |
Nature of Effect at mg/L CN |
||
|
|
|
Spawning Egg Production Egg Viability Spermatogenesis Abnormal embryonic development Hatching Swimming |
completely inhibited reduced by 42% eggs infertile permanent reduction severe deformities up to 40% failure reduced 90% at 6 °C |
0.005 0.01 0.065 0.02 0.07 0.01-0.1 0.015 |
|
In addition to the total cyanide level in the water, a number of other factors associated with water chemistry exert a modifying effect on acute toxicity; these include: dissolved oxygen concentration, temperature, pH, salinity, and other dissolved constituents (e.g., zinc, ammonia).
Cyanide toxicity in fish increases 3-fold with a 12 °C decrease in temperature. Furthermore, seventeen parts per thousand chloride ion (full strength sea water) decreases the survival time (Hynes et al. 1999).
Cyanide is not persistent in the environment and under normal conditions will not permanently destroy a fish habitat. Due to this acute sensitivity, fish make excellent biological markers for the presence of cyanide in water. If fish are living after exposure, then no other form of life will be harmed.
In Hungary, cyanide limit values have been established for a number of uses. Surface water quality is classified in a five-category evaluation system. With respect to cyanide concentration, any water containing cyanide with a concentration value exceeding 0.100 mg/l is classified as heavily polluted (category V). The maximum limit value permitted by the Hungarian standard on drinking water sets a maximum of 0.100 mg/l. In most cases the limit values of the European Union are more stringent than the Hungarian ones. Directive 75/440/EEC prohibits the water abstraction from surface water for drinking water supply above cyanide concentration of 0.05 mg/l. According to the directive 98/83/EEC the limit value of cyanide concentration of potable water is 0.05 mg/l.
Despite the complex chemistry, regulators generally require that mine operators monitor for only three categories of cyanide: free cyanide, weak-acid-dissociable (WAD) cyanide, and total cyanide. Furthermore, the analytical procedures used to determine these categories of cyanide fail to indicate the presence of many of the other toxic breakdown products of cyanide. For example, routine analyses of cyanide fail to identify cyanates and thiocyanates, two significant cyanide breakdown products found at mine sites. Water samples from mining sites where cyanide is used as a process chemical may have WAD and/or total cyanide concentrations that are quite low or undetected, yet when the same samples are analyzed specifically for cyanates and thiocyanates, they may show tens of milligrams per liter (mg/L) or more of these compounds.
Hynes, T.P. J. Harrison, E. Bonitenko, T.M. Doronina, H. Baikowitz, M. James and J.M. Zinck: The International Scientific Commission’s Assessment of the Impact of the Cyanide Spill at Barkasaun, Kyrgyz Republic, May 20, 1998. In: Mining and Mineral Sciences Laboratories, Report MMSL 98-039(CR), August 1999
Ingles, J.C. 1982. Toxic of Cyanide., Presentation at a Seminar on "Alkaline Chlorination for Gold Mill Operators". May 26, Vancouver, Canada.
Heavy metals
Heavy metals on a mining site may be present either as a result of the mining operation or occur naturally. In both cases, heavy metals may be mobilized especially if acid mine drainage is present and subsequently cause potential health and environmental problems resulting in high burdens to soil and water. The acidic mine drainage contains sulfuric acid which is derived from the oxidation of the heavy metal sulfides.
Among the heavy metals abundant in the mining industries, arsenic, cadmium, lead, nickel, manganese and molybdenum are the most harmful to human life as they are bioaccumulative and relatively small doses can seriously affect health. Copper and chromium are detrimental to aquatic life. Zinc, lead, aluminum, boron, and iron may rapidly become available either in acid soils or as salts precipitated from neutralizing solutions (in the mine). They are all, to a greater or lesser extent, toxic to plant growth.
Copper is toxic to most aquatic vegetation. Contained in sediments, copper is better soluble than the other heavy metals and thus, better available for the aquatic food chain.
Mercury is highly toxic as a liquid, a vapor and as organic complexes. It is bio-accumulative and is a special risk to workers handling mercury. Detrimental effects on animals and humans are irreversible. Mercury can cause major environmental damage to all types of animal and plant life. In the present accident, mercury did not play a role.
Heavy metals have a toxic effect on living organisms primarily as a result of the bioaccumulation process. In Hungary, the category of heavily polluted water (category V) in the classification system of surface waters, contains the limit value 0.100 mg/l for copper and 0.300 mg/l for zinc.
In order to classify and to put into perspective the data that are shown in this report, Table A 2 gives water quality standards as established by the World Health Organization in 1993 and the Standard by the European Commission in 1998.
Water quality criteria for heavy metals - Please note that some concentrations are in milligram (mg) whereas others are in micrograms (ug) per liter)
|
Chemical |
WHO Guideline (1993) |
EU Standard |
|
Arsenic |
10 µg/L |
10 µg/L |
|
Cadmium |
3 µg/L |
5 µg/L |
|
Chromium |
50 µg/L |
50 µg/L |
|
Copper |
2 mg/l |
2 mg/L |
|
Lead |
10 µg/L |
10 µg/L |
|
Manganese |
0.5 mg/L |
50 µg/L |
|
Nickel |
20 µg/L |
20 µg/L |
|
Mercury |
1 µg/L |
1 µg/L |