On the origin and remediation of the Aznalcóllar toxic spill - an independent assessment

Summary of the discussions of the Doñana forum until May 1999

compiled by: L. Santamaría, J. Amézaga

(last updated July 14, 1999)



A large toxic spill occurred in Los Frailes (Aznalcóllar) last April 25 1998, following the rupture of a large tailings dam containing the effluents of a copper and zinc mine which produces pyritic waste rock, owned by Boliden APIRSA. Afterwards, Boliden reported that the failure of the dam was caused by an earth slide in the base of the dam1.

The accident sent 5,000,000 m3 of acid sludge into the Guadiamar River. The contaminated sludge contained high concentrations of arsenic, zinc, lead, copper, talium and cadmium2. During the clean-up operation that followed, the collected sediment was deposited in a former open-pit of the mine. The authorities approved the use of the open-pit based on a technical assessment from the Spanish ITGE, which concluded that there was an acceptably low risk of groundwater contamination. Under a recently approved scheme, the mine will re-open and the same open pit will be used for tailings disposal instead of the ruptured dam3.

During the last months, a group of researchers, engineers and other professionals have been interchanging opinions on the remediation of the Los Frailes spill through an e-mail discussion list (the 'Doñana forum'). The list was based on a free and informal exchange of information between interested parties, on the grounds of intellectual honesty and openness. The comments that follow here are a summary of the opinions and issues debated in the list. They are not necessarily the views of all the list members, though a majority of them contributed to the debate.


Causes, impacts and future risks

The problem behind the Los Frailes spill is certainly not a lack of technical know-how in the mining industry. Considerable progress has been made worldwide towards achieving a mining technology which is sustainable and does not harm people or the environment. Unfortunately, the drive to develop the Andalusian region has led the regional and national authorities to be repeatedly lenient with the implementation of environmental standards for the region's mines and industries4, including generous subsidies to contaminating industries and lack of rigour in technical inspections.

The most likely cause of the accident was defective construction and maintenance of the tailing dam. These deficiencies are likely to have led to over-pressuring of water-rich products within the tailings stock itself or onto the underlying recent sediments. An earth-slide as trigger for the failure seems unlikely for several reasons: the low topographic gradient in the vicinity of the toe of the dam, the highly localised nature of the failure and the lack of any causative process (e.g. earthquake) other than the loading by the tailings dam. The point that needs to be clearly answered (in terms of preventing a repetition) is whether the design and construction of the dam were adequate in the first place.

Boliden APIRSA bears a major responsibility for the pollution incident, through lack of care in the maintenance of the tailing pond and in the treatment of the mine's effluents. Though the EU 'polluter pays' principle should command that Boliden pays a major part of the clean-up operation5, the data reported by the media show that they have covered no more than 10 to 25 % of the expenses.

Some of us have defended the opinion that, without a previous impermeabilisation (such as a plastic film cover), it cannot be affirmed that the open-pit of Aznalcóllar will allow the storage of mining sludges and tailings without any risk of pollution of the underground waters. The open-pit is located within materials which have been strongly folded, thrusted, and shistosed (slates) during the Variscan orogeny. Due to the presence and abundance of strongly dipping planar structures (schistosity, thrusts, diaclases), this is certainly not an impermeable zone (fracture permeability). Furthermore, the percolating waters will flow through sulphide-rich rocks and within this retention system (an anoxic acidic medium), arsenic could be solubilised in its more mobile and toxic form (As3+) and percolate downflow.

However, the ITGE has determined that impermeabilisation is not necessary as long as the waste is below the head of water in the surrounding strata. As long as the tailings are not allowed to accumulate above the decant levels to the valley aquifer, installing plastic would be an unnecessary expense. It could even be counter-productive, as plastic impermeabilisation could involve filling the hole to a higher level even though we all know plastic liners will eventually leak. What is important to stress, in view of recent claims to rise the level of the tailings deposit, is that the site is certainly permeable and polluted water will percolate if the level is too high.


The first phase of the remediation - a critical assessment

Considering the presence downstream of major agricultural areas, fisheries, tourist resorts and nature reserves of international importance, the lack of a risk assessment and an emergency plan for the mine was a major technical flaw. The situation was worsened by the lack of institutional co-operation and the conspicuous political efforts to minimise the public perception of the accident, rather than treating it. This, combined with the particular circumstances of the Spanish scientific-technical arena, resulted in a quite defective operation. We believe that it is time for a re-evaluation of the objectives and procedures, so that the second phase of the remediation becomes more effective.

When the toxic sludge reached Entremuros, the inundation wave was only a few centimetres high. At that time, the best solution would have been its containment by emergency dikes, allowing for a limited dispersal of the contaminated sediment that would have facilitated its treatment. The diversion of the toxic sludge to the Guadalquivir estuary, where it could not be treated, has created an additional hazard for the local tourism and fisheries. The pollution of the estuary has added to the continuous discharge of the Tinto-Odiel estuary, one of Europe's most polluted areas.

The analysis provided by the CSIC quickly demonstrated that the seafood of the Guadalquivir estuary was severely contaminated6. Fortunately, its consumption was quickly regulated. However, pollution limits for seafood are not stringent enough because they ignore the synergy of the different pollutants and the existence of multiple polluted food sources.

During the 6 months that followed the accident, the toxic sludge was mechanically removed6. This operation lacked an adequate assessment of the spill, proper co-ordination, adequate equipment, safety measures to prevent the intoxication of the workers, and the use of adequate technology. As a result, the removal was fairly chaotic, huge amounts of contaminated aerosols were resuspended, groundwater pollution was not properly controlled, the flow of contaminated particles to the estuary continued several weeks after the spill (bound to suspended solid matter) and the watershed suffered major modifications (with broad areas of exposed soil and the associated erosion). Adequate consolidation and capping has not been reported. In addition, the slow response to the disaster could have caused far worse problems if major rainfall had occurred following the original event.

The team that co-ordinates the remediation efforts has an unbalanced representation of researchers without direct experience in the customary procedures of pollution control and restoration in the mining industry. Conspicuous absences in the team are professionals with practical experience of remediation in the field, mine water specialists and system-scale ecologists.


Proposed guidelines for future remediation efforts

The lack of institutional transparency makes it difficult to assess which remediation projects are on their way and what technologies will be used. However, the media have repeatedly reported on the use of phytoremediation and proposals to develop transgenic plans for this purpose. We believe that dedicating the efforts to develop new technologies and/or re-investigating technologies that are already in the market, is not the right way to address the restoration. On the contrary, the urgency and scale of the problem requires the use of readily available technology.

A whole variety of methods are available for the soils and river-sediments contaminated by the Los Frailes spill. Depending on the degree of soil pollution, these might include (a) in-situ stabilisation/solidification, (b) phytoremediation, (c) complete excavation of soils/sediments, extraction of contaminants and re-deposition on site, (d) complete removal and off-site disposal (already done), followed by secondary/tertiary treatment and/or consolidation and capping, and (e) inundation to prevent further oxidation of the pyrite residues.

Phytoremediation is a promising new technique for remediating metals in soils, but full-scale applications are still limited. In particular, phytoremediation of certain metals (arsenic, copper) is still in the research stage7. Phytoremediation has also a high public acceptance, low cost, generates less secondary wastes and has a small impact on soil structure7,8. However, it is much slower than mechanical treatments, the toxicity and bioavailability of degradation products is not known, contaminants may be mobilised into the groundwater and there is a potential for contaminants to enter food chain through animal consumption7,8.

Proposals to develop genetically modified bacteria and plants to treat the polluted zones are, particularly for arsenic (no hyper-accumulator plant known, no detailed knowledge of the genetic control for arsenic accumulation), far too ambitious for the time scales proposed (3-5 years). Proposing a technology that will take years to develop, without a guarantee of success, is particularly irresponsible in front of the emergency of this concrete case of toxic pollution.

A thorough remediation plan is needed. It is necessary to define clearly the extension and the state of the polluted zones (soils, waters, life), to classify them and to decide what we want exactly, when we hope to recover them and for what use (agriculture, tourism, nature reserves, water resources, etc.). It should be evident that there is no unique solution and that each solution has its own cost and we must be careful considering all possible unexpected secondary effects (mainly on groundwater). In view of the urgency, the most simple and low cost solutions should be probably the best ones.

There seems to be little institutional agreement on the future use of the polluted zone. Plans to transform the area into a green corridor are strongly contested by the agricultural lobby, which claim low levels of soil pollution. Soil pollution legislation advanced by the regional government is also discouraging. Not only are the proposed pollution limits for agricultural soil very high when compared to the legislation in other countries (Europe, USA, Canada), but these limits are doubled for non-agricultural soils.


Institutional response and public perception

A subject of concern is the apparent resentment of the local population towards the publicity of the accident rather than towards the accident itself. Local perception is that the spill (or rather its publicisation) poses a threat to their income. At the same time, they have not received clear institutional information on the hazards posed by the spill (rather they have received information aimed to minimise the perception of these risks). In this situation public demand for proper remediation and tight public health standards is likely to be minimal. Institutional acknowledgement of the economic, social and public health problems created by the spill and proposal of long term solutions to these problems should be part of any regeneration plan.

The pollution of the Guadiamar River and Guadalquivir estuary imposes a further burden on the Gulf of Cadiz to that already arising from the nearby Tinto-Odiel river system. The regular annual pollutant loading coming from the Tinto-Odiel system is equivalent to the 'one-off' spill from Los Frailes9,10. The coastal waters of the Tinto-Odiel region have high metal concentrations when compared to any other European coastal areas9,10. These data indicate that the situation in SW Spain is the worst that we have to face in Western Europe. The danger is real for its populations, which are continuously exposed to water and air contamination. However, detailed epidemiological studies are still lacking. It is also ironic that two wetland areas of major national and international importance (the Doñana National and Natural Parks, and the Odiel marshes) are situated there.

The remediation of the Los Frailes spill is above all a regional problem of national dimensions, and should be solved at this level. There are major lessons to be learnt from this dreadful event that should have obvious effects on all the institutions involved in it. The company, local and regional authorities, and national administration have the public duty to introduce major adjustments, within a coordinated regional framework, to the systems and procedures that failed to prevent, and react adequately to, the accident.

The problem of the Western Andalusia mining region (Tinto-Odiel and Guadiamar- catchements) is also international. The Doñana park is protected by European legislation and a world reserve of UNESCO, many migrating birds stop in Doñana and its surroundings, large amounts of tourists utilise the place, and strawberries and other fruits are exported through Europe. Environmental pollution is a serious issue with immediate and long-term consequences. Not only because it threatens the health of people and ecosystems, but also because it has the potential to destroy their major sources of income (agriculture and tourism). We believe that the EU institutions have the capacity to foster an adequate restoration of the Doñana watershed, through a careful consideration of subsidised activities. This could have a real effect on the ground, helping to overcome regional and national inertia and facilitating definite improvements.



1. see a/o El País, 29-09-98, pg.30.

2. Geen, A. van & Z. Chase, 1998. Recent Mine Spill Adds to Contamination of Southern Spain. EOS, Transactions, American Geophysical Union: 79, 449-455.

3. see a/o El Pais, 05-02-1999.

4. For an example of reported pollution levels that preceded the larger spill discussed here, see: Arambarri et al., 1981, La contaminación en la zona minera y en la zona agrícola del río Guadiamar, Simposio sobre el agua en Andalucía, 2: 817-832, and Fournier et al., 1985, Niveles de metales pesados en los sedimentos actuales del Parque Nacional de Doñana, Comunicaciones INIA - Serie Recursos Naturales , 38: 67 pp., Madrid. For a toxic spill that preceded the one discussed here, see: El correo de Andalucía, 2-02-02-96 pg. 14, 15-03-96 pg. 11; El Pais, 02-02-96; ABC, 02-02-96; El Correo, 03-02-96. See also: Arambarri, P., Cabrera, F., and González Quesada, R., Quality evaluation of the surface waters entering the Doñana National Park (SW Spain), Sci. Total Envir., 191, 185-196, 1996.

5. see a/o. Framework for Community Action in the Field of Water Policy, EU DGXI, homepage http://europa.eu.int/comm/dg11/docum/9749sm.htm . See also: Council Directive on the Conservation of Natural Habitats and of Wild Fauna and Flora (92/43/EEC), at homepage http://www.ecnc.nl/doc/europe/legislat/habidire.html

6. CSIC, 1999. Coto de Doñana. Accidente de Aznalcóllar. [Reports of the 'experts comission' on the spill and its remediation]. Homepage: http://www.csic.es/hispano/coto/aznalco.htm .

7. EPA, 1998. A Citizen's Guide to Phytoremediation. United States Environmental Protection Agency, Office of Solid Waste and Emergency Response (5102G) EPA 542-F-98-011, August 1998

8. see a/o Chappell, J. 1998. Phytoremediation of TCE in Groundwater using Populus. Status Report prepared for the U.S. EPA Technology Innovation Office (TIO), 401 M Street, SW (5102G), Washington, DC 20460. Homepage: http://clu-in.org

9. Cossa, D., Elbaz-Poulichet, F., 1999. Mercury in an acidic river-estuarine system: consequences for the Atlnatic Shelf Waters. 5th International Conference on Mercury as a Global Pollutant. Rio de Janeiro, Brazil, May 23-28, 1999.

10. Morley, N.H., Elbaz-Poulichet, F., Becker, J.M., Nomerange, P., 1999. Trace metals in the waters of the Straits of Gibraltar. Oral presentation. 24th General Assembly of the European Geophysical Society in The Hague, 19-23 April 1999. Journal Geophysical Research Abstracts, Volume 1.

The 'Doñana forum' is an activity of the Spanish Association for the Advancement of Science and Technology (AACTE; http://www.econ.upf.es/~ninomora/aacte.html )

This page brought to you by:
WISE Uranium Project (home) · The Los Frailes tailings dam failure · Safety of Tailings Dams