Toxicology Letters 140 /141 (2003) 343 /351

www.elsevier.com/locate/toxlet

Review

Groundwater pollution and remediation options for multisource contaminated aquifers (Bitterfeld/Wolfen, Germany)
P. Wycisk a,*, H. Weiss b, A. Kaschl b, S. Heidrich b, K. Sommerwerk a
a

Department of Environmental Geology, Institute of Geological Sciences, Martin Luther University, Domstr. 5, D-06108 Halle (Saale),
Germany
b
Interdisciplinary Department of Industrial and Mining Landscapes, UFZ Centre for Environmental Research Leipzig-Halle,
Permoserstr. 15, D-04318 Leipzig, Germany
Received 15 September 2002; accepted 12 December 2002

Abstract
Large-scale contaminated megasites like Bitterfeld/Wolfen in the eastern part of Germany are characterized by a
regional pollution of soil, surface water and groundwater due to the long and varied history of the chemical industry on
location. The pollutants in groundwater may spread to uncontaminated areas and endanger receptors like surface water
and drinking water wells according to the site-specific hydrologic regime. In addition, the sheer extension of the
contamination at megasites as well as the existence of large densely populated areas and land of high-reuse value
prevent a simple risk management strategy of use restriction for the whole area. Since a complete clean-up of the
groundwater on a megasite is neither economically feasible nor technically possible within a reasonable time-frame, a
multi-approach remediation strategy is needed, taking into account the immediate risks for human health, ecosystem
and so-called protectable goods. Moreover, the contaminants at megasites typically represent a dangerous cocktail of
multiple harmful substances stemming from a variety of sources, which may interact with each other and complicate the
search for an appropriate remediation strategy. At the SAFIRA-project site in Bitterfeld approaches for in situ
remediation of multiple contaminants in groundwater are being tested. Alternatives in local implementation strategies
as well as consequences of long-term restrictions for megasites like Bitterfeld need an independent evaluation of the
situation using a risk-based approach. For this reason, a GIS-based 3D model of the area including geology,
contaminants, hydrogeology, land-use and protected areas has been built. The regional groundwater pollution is
characterized by contamination profiles of all monitored substances. In the area of investigation, e.g. threefold and
fourfold threshold levels of chlorinated methane, ethane and ethene as well as HCH-isomers, mono-, di- and
tetrachlorobenzene, DDT-isomers and benzene are frequently detected in groundwater, that means in at least 60% of
the wells that were sampled. High median values of more than 10 mg/l were calculated for cis /trans -1,2-dichloroethene,
1,2-dichloroethane, chloroethene and monochlorobenzene. In general, the regional distribution of contaminants reflect
the different sources and pathways, and give first results from a regional point of view, depending on a land-use
classification of specific areas.
# 2003 Elsevier Science Ireland Ltd. All rights reserved.

* Corresponding author. Tel.:
/49-345-55-26-134; fax:
/49-345-55-27-177.

E-mail address: wycisk@geologie.uni-halle.de (P. Wycisk).
0378-4274/03/$ - see front matter # 2003 Elsevier Science Ireland Ltd. All rights reserved.
doi:10.1016/S0378-4274(03)00031-6

344

P. Wycisk et al. / Toxicology Letters 140 /141 (2003) 343 /351

Keywords: Bitterfeld; Megasite; Contaminated groundwater; Chlorinated hydrocarbons; Remediation technologies

1. Introduction
So-called megasites of a regional scale represent
a complex problem in the field of integrated
contamination site management and risk assessment for the site owners and stakeholders involved. Megasites can result from former
industrial and mining areas, harbors and military
areas and are commonly found inside and outside
Europe, as we know today. Due to the large
amount of contaminants that have entered the
soil and groundwater on a regional scale, it must
be assumed that these sites will emit pollutants for
a long time to come, if preventive measures are not
undertaken. Megasites like Bitterfeld/Wolfen with
a long industrial history have a multi-source
origin. These, in general, intractable site history
of contaminant inputs in addition to a large
variety of contaminants of different chemical
properties complicate the assessment of potential
health effects and represent a challenging situation
for the remediator (Krapp and Ruske, 1992;
Walkow et al., 2000).
Post-research in the field of risk assessment and
groundwater remediation techniques has been
mainly restricted to single-source contamination
sites with individual plumes of contaminated
groundwater. Multi-source megasites are characterized, in the case study of Bitterfeld/Wolfen, by
an extended mixture of organic pollutants, which
is typical for the situation of the industrial site, and
possible synergistic and antagonistic effects, which
have not yet been examined to the necessary
details. Since a complete clean-up of groundwater
of the megasite including the removal of the
contaminated source areas /10 km2 is neither
technically nor economically feasible within an
intermediate time-frame, an effective remediation
strategy must focus on risk minimization (Wycisk
et al., 1997; Wycisk, 1998). Land-use restriction
measures as well as simple relocation are in many
cases not a desired option for a regional scale
contamination, due to the given high land-reuse
value within the region. Human and ecosystem

risk management must be compliant to the national and governmental (state) regulations and
will present an ongoing challenge within the
European Water Framework in the years to come.

2. Bitterfeld/Wolfen megasite
The industrial region of Leipzig /Halle /Bitterfeld in eastern Germany is characterized by overlapping environmental impacts of the chemical
industry, an extensive devastation of the landscape, and a lowering of the groundwater table
caused by lignite mining. Among the different
tasks of land reclamation, the management of
hazardous-waste deposits and groundwater remediation are the most urgent. Underground and
open-cast mining activities since 1830 and the
concentration of chemical industry sites led to an
extensive lowering of the groundwater table and a
change of the groundwater dynamics. The major
part of industrial-waste deposits is in contact with
the groundwater. As a consequence of this hydrogeological situation and the low-flow velocities,
the landfills currently show a stagnating emission
pattern.
An area of about 235 km2 is affected in the
Bitterfeld region, of which an area of 25 km2
shows a significant groundwater contamination,
containing a volume of about 200 million m3. The
restoration of pre-mining groundwater conditions
is the major task in re-development of post-mining
landscapes. This will have considerable consequences for the assessment of the polluted areas
and the landfills.
The increase of the groundwater level will also
cause considerable problems for the basements of
the buildings in the region. During the last
hundred years, the mean groundwater level was
kept noticeably below the surface. The consequences of the currently rising groundwater level
are not only damages to buildings, but also the

P. Wycisk et al. / Toxicology Letters 140 /141 (2003) 343 /351

mobilizing effect on contaminants and their transport into basements of inhabited structures (buildings, houses).
In northwest of Bitterfeld, several former opencast lignite mines have been used as landfills for
chemical wastes. Hydrogeclogically, the waste sites
are situated in a region with severely disturbed
groundwater conditions due to the mining activities. Since the industrial dumps were incompletely
sealed, the contaminates affected the groundwater
directly. Fig. 1 gives an overview of the groundwater-flow direction and the abundant distribution of large contaminated sites.
The Antonie landfill, for example, contains
about 6 million tons of various industrial residues,
including waste material from pesticide produc-

345

tion. The groundwater below the site is extremely

polluted by HCH-isomers (Walkow et al., 2000).
The groundwater pollution downstream, in contrast, is characterized by the more mobile contaminants like chlorinated aliphatic and aromatic
hydrocarbons. The concentrations reach several
hundreds of milligrams per liter, exceeding effluent
standards by several orders of magnitude. Conventional pump and treat technologies are
economically unfeasible in the long term, due to
the fact of the continuous output from the landfills
and the necessary volumes of about 4 million m3
per year. Presently, the groundwater is discharging
into the Goitzsche open-cast, but after complete
flooding, the whole floodplain area of the Mulde
River will be increasingly affected in the future.

Fig. 1. Overview of the Bitterfeld area with the distribution of contamination sites and the major groundwater-flow direction, 1995.

346

P. Wycisk et al. / Toxicology Letters 140 /141 (2003) 343 /351

2.1. Geological and hydrogeological background

The contaminated area of Bitterfeld is located in
the floodplain of the Mulde River and can be
described by the following generalized hydrogeological situation. Fig. 2 shows a schematic geological/hydrogeological overview of the area.
The upper aquifer consists of Quaternary sands
and gravels. The Quaternary unit can be divided
into a lower part, represented by lower terrace
sediments of the Weichselian Mulde and overlying
sediments, composed of braided river deposits of a
smaller tributary stream. Both are separated by a
hydraulically effective clay layer (Krapp and
Ruske, 1992; Ruske et al., 1997).
This aquifer is in parts underlain by the upper
oligocene lignite seam acting as a local aquitard.
The lignite seam has been intensively mined in the
southern part of Bitterfeld. The lower aquifer
consists of the upper oligocene micaceous sands
of different hydraulic conductivity in its upper and
lower part. The base of this hydrogeological
section is represented by middle Oligocene clays
(Rupelian clay). The latter unit is considered to be
the regional scale aquitard, hence corresponding to
the base of the groundwater pollution.

Identification, assessment and remediation of a

large-scale groundwater contamination require a
profound knowledge of the geological structure to
predict the fate of contaminants in the subsurface.
Due to the distinct heterogeneity of the upper
aquifer (Quaternary) and the large artificial
mining dumps, a detailed 3D digital geological
model of the subsurface geology was built (Wycisk
et al., 2002). Using 125 selected drill holes as a base
of 28 networked cross-sections, small-scale lithological and structural heterogeneities, in particular
of the Quaternary layers, could be assigned to 31
litho-stratigraphic sedimentation units and depicted using a 10/10 m2 GIS grid. An assignment
of hydraulic parameters to individual sedimentary
bodies allows a combination with flow and transport models. The structural model was generated
by combining gridding sections and additional
point and lateral information of sediment distribution and allows */ beyond visualization
purposes */volumetric calculations of distinct sedimentary units, which are relevant for an assessment of retardation processes in the remaining
lignite seam. The present structural model is the
base for a Spatial Model Bitterfeld (Wycisk et
al., 2002) and enables further investigations on the

Fig. 2. Schematic geological and hydrogeological setting of the pathway of contaminants at the Bitterfeld area, which has been
dominated by industrial and mining activities for more than 100 years.

P. Wycisk et al. / Toxicology Letters 140 /141 (2003) 343 /351

environmental impact of distinct scenarios of

toxicological exposure assessment.
2.2. Groundwater pollution at the local scale
In order to characterize the organic pollution of
the groundwater at various water levels, several
samples were examined at a local scale from
different depths at the remediation SAFIRA pilot
plant. The main components of organic contamination were found to be chlorobenzene, 1,2dichlorobenzene, 1,4-dichlorobenzene, benzene,
trichloroethene, cis -1,2-dichloroethene and trans 1,2-dichloroethene. Despite the fact that high
organic pollution with halogenated hydrocarbons
was confirmed in both aquifers, contamination
varied considerably in terms of quality and quantity. The lower aquifer is dominated by aliphatic
chlorinated hydrocarbons (trans -1,2-DCE, cis -1,2DCE and TCE), whereas the main component of
the upper aquifer is chlorobenzene. During depthoriented sampling, the contamination was found
to display a significant stratification, whereas the
sample from a depth of 7/8 m only contains very
low amounts of pollutants; samples from depths
between 16 and 20 m were found to contain high
levels of chlorobenzene (8 /51 mg/l), as well as
substantial concentrations of dichlorobenzenes (up
to 1 mg/l).
One striking aspect is that the aliphatic halogenated hydrocarbons and benzene chiefly occur in
the strata between 19.5 and 24.5 m, whereas
chlorobenzene and dichlorobenzene are found in
higher concentrations in the strata between 12 and
22.65 m. The main components are trans -1,2dichloroethene, cis -1,2-dichloroethene, chlorobenzene, trichloroethene and benzene. Bromobenzenes and perchloroethene were also identified by
mass spectrometry. Contamination of the groundwater with inorganic pollutants (e.g. heavy metals,
arsenic, etc.) has proven to be of minor importance. The only noteworthy feature is the high
levels of sulfate (up to 1000 mg/l) and chloride
(about 1300 mg/l).
To sum up the local situation of the Bitterfeld
site, the contaminants show a distinct vertical
stratification and high concentration levels alternate with low or zero levels. This shows very

347

clearly that such a differentiated situation can only

be described in detail by multi-level wells, which
are usually not state of the art. Secondly, due to
the hydrogeological heterogeneity in the subsurface, the experience during the last years of
groundwater sampling gives a clear evidence of a
strong local variability of the detected contaminants. These two facts make it very difficult to
state distinct concentration levels as well as certain
organic substances or compounds.
2.3. Groundwater pollution at regional scale
Due to the environmental importance of risk
assessment in the region of Bitterfeld, a groundwater-monitoring program has been implemented
by the site owners and carried out continuously
since 1990. Based on parts of this data set (290
wells, 1200 samples, each up to 180 contamination
parameters), a specific contamination profile for
this region was derived on the base of available
monitoring data within the SAFIRA project
(Thieken, 2001). Before doing that the influence
of important parameters like detection limits and
the statistical measures to estimate the average
detected concentration on the results of the ranking procedure were investigated and led to a
modification of the method. Furthermore, a cluster analysis with the two criteria, detection frequency and average-detected concentration,
reveals substances with similar behavior in the
environment in terms of persistence and mobility,
and improves the contamination profile at the
regional level. In the area of investigation, e.g.
threefold and fourfold detection frequency of
chlorinated methane, ethane and ethene as well
as HCH-isomers, mono-, di- and tetrachlorobenzene, DDT-isomers and benzene are detected
frequently in groundwater, that means in at least
60% of the wells that were sampled. Various
statistical measures were used to estimate the
average-detected concentration for the monitored
contamination data. Fig. 3 shows the regional
distribution and total frequency of the organic
compounds like BTEX, volatile chlorinated organic compounds and chlorobenzene from the
upper aquifer. It was concluded from the monitoring data that the median is the most suitable

348

P. Wycisk et al. / Toxicology Letters 140 /141 (2003) 343 /351

Fig. 3. Regional distribution and total frequency of selected organic compounds and substances from the upper aquifer, like BTEX,
volatile chlorinated organic compounds (LHKW), and chlorobenzene (CB). Groundwater contours from the year 2000.

measure for this purpose. High median values of

more than 10 mg/l were calculated for cis /trans -1,2dichloroethene, 1,2-dichloroethane, chloroethene
and monochlorobenzene. The regional distribution of contaminants reflects in general their
different sources and pathways and gives first
results from a regional point of view, depending
on the land-use classification of specific areas, e.g.
industry, mining, settlements, agricultural and
alluvial plains/meadows (see Table 1).
2.4. Remediation options for megasites
The large-scale contaminated megasite of Bitterfeld/Wolfen is characterized by a regional
pollution of soil, surface water and groundwater,
due to the long and varied history of the chemical
industry. The pollutants in groundwater may
spread to uncontaminated areas and endanger
receptors like the surface water of the Mulde River
and to some extent drinking water wells according

to the site-specific hydrologic regime. The extension of the contamination at the multi-source
megasite as well as the existence of large densely
populated areas and land of high-reuse value
prevent a risk management strategy of use restriction for the whole area. Since a complete clean-up
of the groundwater on a megasite is neither
economically feasible nor technically possible
within a reasonable time-frame, a multi-approach
remediation strategy is needed taking into account
the immediate risks for human health, ecosystem
and so-called protectable goods. Moreover, the
contaminants at the megasite typically represent a
cocktail of multiple harmful substances stemming
from a variety of sources, which may interact with
each other and complicate the search for an
appropriate remediation strategy. Fig. 4 gives a
schematic overview of possible remediation options at the megasite.
To stop the uncontrolled spreading and extension of the contaminants in the groundwater

P. Wycisk et al. / Toxicology Letters 140 /141 (2003) 343 /351

349

Table 1
Average detected concentration in mg/l of indicated organic compounds per classified land-use unit

LHKW
BTEX
CB
HCH
DDT/D/E

Alluvial plain

Mining areas

Urban areas

Industrial areas

Agricultural

Total area

n /RL

MEC

n /RL

MEC

n /RL

MEC

n /RL

MEC

n /RL

MEC

n /RL

MEC

51
40
46
28
27

16.7
4.8
17.0
1.8
0.2

107
62
84
58
32

22.1
2.7
15.8
2.6
0.1

36
24
34
11
10

66.0
11.2
13.2
2.0
0.2

155
130
149
124
97

89.4
15.7
89.2
2.6
0.3

16
13
16
1
0

5.8
2.5
10.2
/
/

365
269
329
222
166

42.7
8.2
26.5
2.5
0.2

The measured concentrations (MEC) reflect the source /receptor (Mulde River) relation of the pathway of contaminants. Selective
data set 9198M. MEC, mean emission concentration; RL, relevance level; LHKW, volatile chlorinated organic compounds; CB,
chlorobenzene (modified after Thieken, 2001).

towards uncontaminated areas and receptors,

including the surface waters, a common approach
consists of pump-and-treat measures. The sufficiently treated water can be channeled into the
surface water (channels, rivers, creeks). If the
quantity of water pumped is large enough, the
groundwater flow regime may be altered and a
further expansion of the contaminated zone is
avoided. However, this approach requires that
relevant technologies exist for treating the particular contamination encountered at the site.
Pump-and-treat measures are being currently employed and their expansion considered at the
koloBitterfeld site within the framework of the O
gisches Groprojekt (Ecological Mega Project)
Bitterfeld /Wolfen. Barriers consisting of pumping
wells are used to stop the groundwater flow in
northern, northeastern and eastern direction. The

NE barrier, for example, has been in operation

since 1994, using 20 pumping wells to extract an
annual amount of 1.3 /106 m3 contaminated
water from the Quaternary aquifer (Lucke and
Gromann, 2002).
In order to provide the cost-effective remediation schemes, efficient monitoring strategies need
to be developed. These strategies include integral
sampling methods such as time integrating sampling devices for the assessment of contaminant
fluxes over longer periods of time. In addition,
depth-specific information on contaminant distribution is required to efficiently place active
remediation measures or permeable reactive barriers (Schirmer et al., 1995; Merkel et al., 2001;
Wei et al., 2001).
At the SAFIRA-project site in Bitterfeld, different approaches for in situ remediation for multiple

Fig. 4. Sketch of the long-distance related fate processes of organic contaminants at megasites and alternative remediation options for
groundwater.

350

P. Wycisk et al. / Toxicology Letters 140 /141 (2003) 343 /351

contaminants in groundwater are being tested.

SAFIRA is the German acronym for Remediation Research in Regionally Contaminated Aquifers (Merkel et al., 2001; Wei et al., 2001, 2002).
The following systems are being tested as part of
in situ technologies:
. biodegradation of chlorinated contaminants in
an anaerobic/microaerobic system,
. adsorption and simultaneous microbial degradation on activated carbon,
. zeolite-supported palladium catalysts,
. membrane-supported palladium catalysts,
. oxidative solid metal catalysts,
. activated carbon filtration,
. anaerobic microbial degradation of pollutants,
. combination of redox reactors.
These technologies which have been used in the
pilot plant reactors are directly fed by groundwater from the Quaternary aquifer. The technologies may be upscaled and used subsequently in
the so-called sorptive barriers or permeable
reactive barriers, which are installed directly in
the aquifer (Fig. 4). Reactive barriers are located
in the main direction of contaminant flow and
decrease the contaminant loading by, e.g. adsorbing reactions (e.g. activated carbon) or through
redox reactions (Fe8, ORC) (Teutsch et al., 1996;
Wei et al., 2002). A variation of reactive barriers
is the funnel-and-gate approach, where the
contaminant plume is funneled by impermeable
funnel walls towards the reactive gate inside the
aquifer. The gate reacts as a zone, where the
contaminants are broken down or adsorbed.
For the area of the Bitterfeld megasite lying
downstream of the main contaminant zone, monitored natural attenuation (MNA) may prove to be
a valuable and additional remediation option and
should be included in the overall strategy. MNA is
based upon a variety of physical, chemical or
biological processes that, under favorable conditions reduce mass, toxicity, mobility, volume or
concentration of contaminants in soil or groundwater. After sufficient NA rates have been confirmed on-site, long-term monitoring must ensure
the persistence and relevance of these rates to
minimize all relevant risks, in order to guarantee a

permanent protection of man and the environment.

Acknowledgements
The SAFIRA project is funded by the German
Federal Ministry of Education and Research (E1.1
FKZ 02WT0023). We are grateful to the LAF,
GP (Bitterfeld/Wolfen) of Saxony Anhalt and
O
the Landratsamt Bitterfeld providing us with the
groundwater monitoring data used in the project.

References
Krapp, L., Ruske, R., 1992. Geologische Verhaltnisse von
Bitterfeld und ihre Relevanz zu Kontaminationen des
Grundwassers und Bodens. In: Hille, J., Ruske, R., Scholz,
R.W., Walkow, F. (Eds.), Bitterfeld, Modellhafte okologische Bestandsaufnahme einer Kontaminierten Industrieregion. 1. Bitterfelder Umweltkonferenz. Erich SchmidtVerlag, Berlin, pp. 85 /92.
Lucke, E., Gromann, J., 2002. Grundwasser Management am
Standort Bitterfeld /Sachstand und Optimierungskonzept.
In: Lal, M., Scholz, S. (Eds.), Berichte vom 18. Bochumer
Altlasten-Seminar 2002 und 13. Leipziger Altlasten-Seminar
2002, Schurmann
/Klages, Bochum, pp. 371 /384.
Merkel, P., Teutsch, G., Wei, H., Schirmer, M., 2001.
SAFIRA (Remediation Research in Regionally Contaminated Aquifers) */Reactive Permeable Barrier Approach.
Groundwater Quality 2001, Sheffield.
Ruske, R., Hubner, J., Bohme, O., Falke, P., 1997. Ergebnisse
der geologisch-hydrogeologisch-geotechnischen Standorterkundung. In: Wei, H., Teutsch, G., Daus (Eds.), Sanierungsforschung in Regional Kontaminierten Aquiferen
(SAFIRA), UFZ Bericht, Leipzig, 27/1997, pp. 14 /17.
Schirmer, M., Jones, I., Teutsch, G., Lerner, D.N., 1995.
Development and testing of multi-level sock samplers for
groundwater. J. Hydrol. 171, 239 /257.
Teutsch, G., Grathwohl, P., Schad, H., Werner, P., 1996. In
situ-Reaktionswande */ein neuer Ansatz zur passiven Sanierung von Boden- und Grundwasserverunreinigungen.
Grundwasser 1, 12 /20.
Thieken, A.H., 2001. Schadstoffmuster in der regionalen
Grundwasserkontamination der mitteldeutschen Industrieund Bergbauregion Bitterfeld-Wolfen. Ph.D. Thesis. Martin-Luther-Universitat, Fb. Geowissenschaften, Halle
(Saale), 147 pp.
Walkow, F., Enders, K., Peklo, P., 2000. The pollution of soil
and groundwater in Bitterfeld. In: Wei, H., Rinjnaarts, H.,
Staps, S., Merkel, P. (Eds.), SAFIRA */Abstracts of the
Workshop of November 17 /18, 1999, at Bitterfeld, Germany, UFZ-Bericht, 23/2000, pp. 7 /12.

P. Wycisk et al. / Toxicology Letters 140 /141 (2003) 343 /351

Wei, H., Teutsch, G., Fritz, P., Daus, B., Dahmke, A.,
Grathwohl, P., Trabitzsch, R., Feist, B., Ruske, R., Bohme,
O., Schirmer, M., 2001. Sanierungsforschung in regional
kontaminierten Aquiferen (SAFIRA). 1. Information zum
Forschungschwerpunkt am Standort Bitterfeld. Grundwasser 6 (3), 117 /122.
Wei, H., Schirmer, M., Teutsch, G., Merkel, P., 2002.
Sanierungsforschung in regional kontaminierten Aquiferen
(SAFIRA). 2. Projektuberblick und Pilotanlage. Grundwasser 7 (3), 135 /139.
Wycisk, P., 1998. Methodische Ansatze zur Prufung der
Umwelt- und Raumvertraglichkeit von Reinigungswanden.
In: Sanierung von Altlasten Mittels Durchstromter Rein-

351

igungswande /UBA-Fachgesprach, Umweltbundesamt Berlin, pp. 52 /69.

Wycisk, P., Thieken, A., Doring, U., Niessen, R., Lachmann,
S., 1997. Umwelt- und Raumvertraglichkeit der Sanierungsvorhaben von regionalen Grundwasserkontaminationen. In:
Wei, H., Teutsch, G., Daus, B. (Hrsg.), Sanierungsforschung in Regional Kontaminierten Aquiferen (SAFIRA), UFZ Bericht, 27/97, pp. 161 /171.
Wycisk, P., Fabritius, H., Ruske, R., Wei, H., 2002. Das
digitale geologische Strukturmodell Bitterfeld als neuer
Baustein in der Sanierungsforschung. Grundwasser 7 (3),
165 /171.