004
—
the
the
pace
1ank
tive
sing
ring
uma
ent,
ing:
data
-Jun
nia.
"tral
und.
ence
) do
ntal.
Rio
and
Inc,
ring-
IMPACT ANALYSIS AND SAMPLING DESIGN IN THE POLLUTION
MONITORING PROCESS OF THE AZNALCOLLAR ACCIDENT
USING GEOSTATISTICAL METHODS
C. Franco^*, J. Delgado "*, A. Soares?
* CMRP- Grupo do Ambiente. Instituto Superior Tecnico. Universidade Tecnica de Lisboa.
Av. Rovisco Pais — 1096-001 Lisboa, Portugal — cfranco@ist.utl.pt, ncrmp@alfa.ist.utl.pt
? Dpto. Ingeniería Cartográfica, Geodésica y Fotogrametría, Universidad de Jaén
c/ Virgen de la Cabeza, 2 — 23071 Jaén, Spain — jdelgado@ujaen.es
KEY WORDS: Spatial Information Systems, Cartography, Pollution, Impact analysis, Sampling, Statistics
ABSTRACT:
On the morning of the 25 of April 1998, an accidental spill of the waste damp basin of the Aznalcóllar mine (Seville, SW Spain)
occurred. The basin, containing mud and acid waters, suffered a rupture spilling 4 hm? of acid waters and 2 hm? of acid mud directly
into the Agrio and Guadiamar rivers. This accidental spill spread to an area of about 49 km? and they had very important ecological
consequences due to those rivers are the basic water resource of the Dofíana National Park aquifer (located 60km downstream the
mine) that was declared as Biosphere Reserve by UNESCO in 1994. Remediation actions were taken immediately in order to protect
the population health and agriculture and minimize the ecological impact in the Park. Once the primary remediation actions were
applied, it was very important to characterize the residual contamination distribution patterns and to obtain environmental hazards
maps in which areas with high heavy metals concentration (Cu, Pb, Zn, Cd and As) can be identified. This paper presents the most
important results of the geostatistical treatment of this data. The basic objectives of this treatment were to establish the impact
analysis of the contaminant (environmental hazards maps) using the direct cosimulation technique and remediation levels that
defines if a certain zone can be considered as contaminated or clean and the design of a new sampling schema that would be applied
in future sampling campaigns and planning the optimum sampling for the next campaigns.
1. INTRODUCTION
On the morning of the 25" of April 1998, the waste damp basin
of the Aznalcéllar mine, West of Seville (South-West of Spain),
containing mud and acid waters, suffered a rupture spilling
4hm’ of acid waters and 2hm’ of acid mud directly into the
Agrio river and consecutively into the Guadiamar river
(C.M.A, 2000) (figure 1). This accidental spill spread to an area
of around 49 km”. The affected area, situated 60km downstream
the mine, is of great ecological importance because the
Guadiamar river is the main hydrological resource of the
National Park of Dofiana (Biosphere Reserve, UNESCO 1994)
(figure 2).
* Aracena
Figure 2. Acid mud deposits in Guadiamar river (CMA, 2000)
, Minas de
Rid X The evaluation and management of environmental impacts due
na to residual soil contamination, was a main concern. Despite the
STUDY AREA T A VILE direct remediation done to the whole area, primarily through
XE i mobilization and excavation of the acid mud, there still was a
e $5 o & laMayor | significant quantity of residual contamination, which can affect
TM P ; + negatively all ecosystems. With this study, we intended to
saut ‘| characterise the spatial dispersion of heavy metals — Cu, Pb, Zn,
nn d £3 Cd and As — on the Guadiamar river margins, to be able to
ATLANTIC Da M S / elaborate and create environmental hazards maps as basic tools
OCEAN Wu pv 7 for important decision-making, such as the delineation of target
We nticar de | areas for remediation or for additional sampling and the
Jerez de la Froftera optimum sampling grid design.
Figure 1. Geographical location map
* Corresponding author.
373
