The consequences of land-cover changes on soil
erosion distribution in Slovakia
Tomáš Cebecauer a
, Jaroslav Hofierka b,
a
Institute of Geography, Slovak Academy of Sciences, Štefánikova 49, 814 73 Bratislava, Slovak Republic
b
Department of Geography and Regional Development, Faculty of Humanities and Natural Sciences,
University of Prešov, 17. novembra 1, 081 16 Prešov, Slovak Republic
Received 31 October 2005; received in revised form 13 September 2006; accepted 20 December 2006
Available online 2 June 2007
Abstract
Soil erosion is a complex process determined by mutual interaction of numerous factors. The aim of erosion research at regional
scales is a general evaluation of the landscape susceptibility to soil erosion by water, taking into account the main factors
influencing this process. One of the key factors influencing the susceptibility of a region to soil erosion is land cover. Natural as
well as human-induced changes of landscape may result in both the diminishment and acceleration of soil erosion. Recent studies
of land-cover changes indicate that during the last decade more than 4.11% of Slovak territory has changed. The objective of this
study is to assess the influence of land-cover and crop rotation changes over the 1990­2000 period on the intensity and spatial
pattern of soil erosion in Slovakia. The assessment is based on principles defined in the Universal Soil Loss Equation (USLE)
modified for application at regional scale and the use of the CORINE land cover (CLC) databases for 1990 and 2000. The C factor
for arable land has been refined using statistical data on the mean crop rotation and the acreage of particular agricultural crops in the
districts of Slovakia. The L factor has been calculated using sample areas with parcels identified by LANDSAT TM data. The
results indicate that the land-cover and crop rotation changes had a significant influence on soil erosion pattern predominately in
the hilly and mountainous parts of Slovakia. The pattern of soil erosion changes exhibits high spatial variation with overall slightly
decreased soil erosion risks. These changes are associated with ongoing land ownership changes, changing structure of crops,
deforestation and afforestation.
 2007 Elsevier B.V. All rights reserved.
Keywords: Land-cover change; Soil erosion; Geographic information system; Slovakia
1. Introduction
Since the Second World War the Central European
countries (CEC) have undergone two main periods
associated with communist and post-communist historic
eras. These periods resulted in vast landscape changes
where the original landscape structure has been almost
completely modified. During the first four decades that
began in 1948, the landscape development in Slovakia
was driven by a planned communist economy. It significantly
influenced the agricultural land utilisation
and arrangement of individual agricultural plots. The
processes of collectivisation and agricultural mass
production transformed the original mosaic of small
Available online at www.sciencedirect.com
Geomorphology 98 (2008) 187­198
www.elsevier.com/locate/geomorph
 Corresponding author. Department of Geography and Regional
Development, Faculty of Humanities and Natural Sciences, University
of Prešov, 17. novembra 1, 081 16 Prešov, Slovak Republic. Tel.: +421
51 7570613; fax: +421 51 7725547.
E-mail addresses: tomas.cebecauer@savba.sk (T. Cebecauer),
hofierka@fhpv.unipo.sk (J. Hofierka).
0169-555X/$ - see front matter  2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.geomorph.2006.12.035
parcels of arable land and pastures to the landscape
structure composed of large parcels with a size ranging
from tens to several hundreds of hectares (Solin and
Cebecauer, 1998; Van Rompaey et al., 2003; Cebecauerova
and Cebecauer, in press). Moreover, the
amelioration of areas with unsuitable conditions for
intensive agricultural production resulted in the increase
of arable land (Feranec et al., 2000) often with
inappropriate crops thus accelerating land degradation
processes (Van Rompaey et al., 2003). Several local
studies, for example, (Hofierka and Šúri, 1996) or (Solin
and Cebecauer, 1998), showed that these structural
changes significantly accelerated soil erosion processes.
The second period of landscape development has
started in 1989 by a change of political system. The
society has undergone a deep economic transformation
including the agricultural sector. Restitutions and new
agricultural support measures and programmes are
influencing current landscape developments. The CORINE
land cover project revealed that during the 10-year
period between 1990 and 2000 more than 4.11% of
Slovak territory has changed (Feranec et al., 2005a). The
largest changes occurred in agricultural areas and
forestland as a result of several processes: extensification
of agriculture, deforestation, afforestation and
urbanisation. Restitution of arable land and forestland
by original owners or their descendants caused changes
in farming and forestry practices. Large parcels of arable
land especially those in the vicinity of rural settlements
were again split up, areas with unfavourable agricultural
conditions were abandoned (Feranec et al., 2005b).
Since the accession of Slovakia and other central
European countries to the European Union in 2004 the
landscape development has been influenced by the
Common Agricultural Policy (CAP). Therefore further
significant changes in the agricultural land can be
expected. Moreover, recently started comprehensive
land consolidation projects financially supported by the
European Union include many activities and measures
that strongly affect the landscape and landscape
processes including soil erosion.
Soil erosion and sediment transport problems resulting
from land-use/management changes attracted attention
of many researchers. The CAP set-aside regulation,
the main driving force of agricultural land change in
Europe, has a strong influence on actual land management.
This can be also expected for the CEC
transforming agriculture. The land-use changes resulting
from this regulation may have both positive (Kirkby
et al., 2000; Van Rompaey et al., 2001; Fullen et al.,
2005) and negative (Souchere et al., 2003; Boellstroff
and Benito, 2005) influence on soil erosion and sediment
transport because of changing climatic, environmental
and economic conditions over Europe. Souchere
et al. (2003) and Schoorl and Veldkamp (2001) point out
the importance of spatial configuration of set-aside
strategies on the off-site effects and overall sediment
runoff. The effect of the agro-environmental policy
measures in the context of soil erosion and sediment
transport is related not only to the percentage of setaside
arable land, but also to the local conditions and
spatial configuration of set-aside plots. The inverse
relation between soil erosion and land-use changes was
identified by Bakker et al. (2005) in the region of
Lesvos, Greece where soil erosion is considered one of
the main drivers of arable land abandonment.
The studies focusing on the specific land-use changes
and arable land transformations in the CEC and their
impact on soil erosion were presented by Van Rompaey
et al. (2003) and Jordan et al. (2005) both utilising the
WATEM/SEDEM model. While Jordan et al. (2005)
studied the effect of the land-use changes on sediment
transportation rates in the Balaton region in Hungary
during the past two centuries (1784­2002), the study of
Van Rompaey et al. (2003) analysed the effect of potential
land development under the CAP regulation and
land arrangement administration in the selected basins
in the Czech Republic. Both studies identified a strong
influence of land utilisation changes on sediment transport
rates. Soil erosion modelling using historic land-use
data offers a unique opportunity to study impacts of
actual land-use changes on erosion and sediment discharges
(Jordan et al., 2005). In this context, recent
landscape changes induced by political and economic
transformation are part of complex information needed
for future landscape development plans in Slovakia and
other CEC countries. Special attention should be given
to the spatial distribution of these processes and their
association with local environmental conditions, land
utilisation and agricultural production. This knowledge
is crucial for effective land management decisions,
mitigation of land degradation processes and optimisation
of landscape structure for a sustainable landscape
development.
In this paper, we assess the changes in soil erosion in
Slovakia caused by three main factors: land-cover
changes in the 1990­2000 period identified by the
CORINE Land Cover mapping project, changes in
arable land utilisation and management practices using
crop rotation statistics and changes in parcel morphology
identified using LANDSAT TM data. We identify areas
with increased/decreased soil erosion risks and discuss
main reasons influencing these changes in the context of
changing socio-economic conditions of Slovak regions.
188 T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198
2. Methods and data
2.1. Study area
The territory of Slovakia (49,030 km2
) is situated in
the Western Carpathians and, due to complicated
geological structure and tectonic development, it is
strongly dissected. About 25% of the territory, located
predominately in the south-west and south-east, represents
flat alluvial plains or smoothly undulated hillylands.
In the hillylands, the loamy and silty soils were
developed on the loess and loess-like deposits, forming
areas with high susceptibility to soil erosion, despite
relatively gentle slopes. The rest of territory is formed
by highlands and mountains with intra-mountainous
valleys, developed on the crystalline and limestone
rocks in the central part of the Carpathians and on shales
and claystones in the northern part of the Carpathians.
These areas are also highly susceptible to soil erosion
mainly because of steep slopes. Soil erosion by water is
a main process of soil degradation in 75% of Slovak
territory with favourable natural conditions. Almost
50% of Slovakia is utilised for agricultural production
mainly in alluvial plains, hillylands, intra-mountainous
valleys and partially in highlands with moderate slopes.
Forestlands cover almost 44% of the Slovak territory,
mostly in mountainous areas. Since forests have a high
protection function against soil erosion, these areas may
be severely threatened by soil erosion in case of
improper timber production or natural disasters (Šúri
et al., 2002).
2.2. Soil erosion assessment
A regional assessment of soil erosion is influenced by
the complexity of the soil erosion process and available
level of detail in data describing the soil erosion factors.
Regional assessments require simplified models because
there is a relation between data accuracy, model complexity
and a resulting model error (Van Rompaey and
Govers, 2002), which can lead to a decrease of model
reliability when simplified data are used in complex
models. In the last decade, different approaches were
adopted for national and continental assessment ranging
from indicator or factor-based approaches to processbased
models (Gobin et al., 2004). The assessment of
soil erosion risk in this study is based upon principles
and parameters defined by the Universal Soil Loss
Equation (USLE, Wischmeier and Smith, 1978). This
model and its modifications (e. g. RUSLE, Renard
et al., 1997) are still widely used in many countries,
including Slovakia. A set of experiments has been done
to adapt and validate this model to the geographical
conditions of the Czech and Slovak Republics (see e.g.
Pasák et al., 1983; Alena, 1991; Malíšek, 1990, 1992).
The advantage of this model over other process-based
models used in regional assessments (e.g. PESERA
(Kirkby et al., 2004)) is in its simplicity and a greater
availability of input parameters, especially in Slovakia.
As the USLE model was developed for a local-scale
assessment of soil loss by sheet and rill erosion, its
application to regional assessments needs some modifications
(Šúri et al., 2002). Thus the model of soil
erosion risk at the regional scale is defined by this
formula:
EAR Á K Á L Á S Á C; 1
where EA is the soil erosion risk, R is the rainfall
erosivity factor, K is the soil erodibility factor that is a
function of soil properties, LS is the topographic factor
(potential of relief), where L is the slope length factor
and S is the slope steepness factor, C is the land cover/
management factor that takes into account differences in
density and structure of the vegetation cover reflecting
its protective function and also the methods of land
management. The USLE conservation practice factor, P,
is not considered in this model.
The model has been implemented in the GRASS GIS
environment (Neteler and Mitasova, 2004) and ArcView
GIS. The GRASS GIS modules have been used for
preparing the model factors (for example, map algebra,
spatial interpolation and reclassification), while ArcView
GIS has been used for data management, model
results evaluation and visualisation. The datasets and
performed spatial analyses were based on the raster data
model with a spatial resolution of 50 m.
The effects of land-cover changes in this study are
twofold. The first effect results from the land-cover
change per se and it influences the C factor. The second
effect arises from the parcel morphology in agricultural
areas and this affects the L factor. Both these changes
create large part of variability of input parameters in the
one-decade time scale. The temporal variability of the
other factors (R, K and S) is not considered in this study.
They can be treated as quasi-constant factors in this
relatively short period of time.
The proposed methodology has been applied to the
territory of Slovakia in order to identify areas with
increased or decreased soil erosion risk as a result of land
utilisation and parcel morphology changes represented
by the C and L factors during the 1990 to 2000 period.
Two soil erosion risk maps derived using the model (1)
separately for the years 1990 and 2000 reflect a mutual
189T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198
interaction of changing and stable input parameters. The
output of the model in tonnes/ha/year can be reclassified
into soil erosion risk categories. In Slovakia, commonly
used soil erosion risk categories have been defined by
Zachar (1982). These categories help to analyse the
overall trend in soil erosion risk in Slovakia.
Fig. 1. Locations of 16 sample areas used in the analysis of average slope length.
Fig. 2. Identification of parcels using LANDSAT TM data.
190 T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198
2.3. Input parameters
The input parameters of the model (1) are represented
by factors R, K, L, S and C. A detailed description of
the factors and derivation from the existing databases
available in Slovakia was presented by Šúri et al. (2002).
In this section, we briefly recall main principles and
describe some differences in this presented approach.
Rainfall erosivity (R factor) is determined as a
function of total storm kinetic energy (E) and its
maximum 30-min intensity (Imax30) (Wischmeier, 1959;
Wischmeier and Smith, 1958). These two factors were
estimated using the regression analyses of a long-term
15-minute rainfall intensity map published by Šamaj
and Valovič (1980), and a long-term mean annual
precipitation map spatially interpolated by Regularised
Spline with Tension in GRASS GIS (Hofierka et al.,
2002) and the R factor values for selected locations and
areas published for Slovakia by Alena (1991) and
Malíšek (1990). The minimal values of the R factor (in
metric units) are associated mainly with lowlands and
basins (R20­25), while mountainous areas and
eastern parts of the country with more intense rainfall
events have higher values (R30­35).
Soil erodibility (K factor) is a function of soil texture,
organic matter content, structure and permeability. In
Slovakia, there is no national database of the K factor or
individual soil properties covering the whole territory
of Slovakia. Therefore the estimation of the K factor
was based only on the map of soil texture at scale
1:500 000 published by Fulajtár and Čurlík (1980) who
used the national textural classification based on the
content of clay and very fine silt fraction (0­0.01 mm).
Individual soil texture classes of soil texture map were
reclassified to the K factor values (Šúri et al., 2002).
Low values of the K factor are associated with high
clay-content soils (K0.1­0.25), whereas high values
are associated with sandy-loam soils (K0.66­0.75).
Most areas are covered by loamy soils with K0.51­
0.65.
Potential of relief (S and L factor) is a function of
slope inclination and slope length. The S factor was
derived from a digital model of relief (DMR) with a grid
resolution of 50 m. This DMR is based on the digitised
contours of topographic maps in scale 1:50 000. The
Regularised Spline with Tension method has been used
to interpolate the grid-based DMR. The root mean
square error of interpolation at given points is less than
2.3 m (Šúri et al., 1997; Hofierka et al., 1998). The
selected spatial resolution reflects the scale of primary
data and provide a sufficient level of detail for this type
of regional assessments. The S factor values for
individual grid cells were derived from the slope angle
A (%) map using the standard USLE formula:
S  0:065  0:045A  0:0065A2
2
The slope length factor is defined as the slope
distance from the point of origin of overland flow to the
point of concentrated flow or until deposition occurs
(Haan et al., 1994). At a regional scale, the computation
of the L factor cannot solely rely on the DMR because it
does not contain all information necessary to derive a
spatially-distributed L factor. Moreover, the CORINE
land cover database used in this study does not provide
information about individual parcels of arable land and
landscape features influencing the runoff concentration
and deposition. A similar problem was solved by Van
Dijk (2001) and Molnár and Julien (1998) by choosing a
fixed value of slope length for the entire study area
reflecting a typical or average value. The same approach
was used in this study, however, the average slope
length factor was determined separately for each time
horizon. To account for temporal changes in parcel sizes
and morphology, 16 sample areas in different regions of
Slovakia were selected (Fig. 1). Parcel borders usually
present lines were overland flow is concentrated or
deposition occurs, so they can be considered as bounding
lines when calculating the slope length L factor.
The parcels were visually identified and vectorised
Table 1
Land cover/management (C factor) derived from the CORINE land
cover classes
CORINE land cover classes Land cover/management (C factor)
141, 31×, 321,322, 41× 0.001­0.005
142, 231, 324 0.006­0.010
243 0.100
22× 0.350­0.450
333, 334 0.500­0.550
211 0.166­0.335
242 0.108­0.218
11×, 12×, 13×, 331, 332, 51× Not considered
The codes represent the following classes:
141 -- green urban areas, 31× -- forests, 321­322 -- scrub and/or
herbaceous vegetation associations, 41× -- inland wetlands.
142 -- sport and leisure facilities, 231 -- pastures, 324 -- transitional
woodland-scrub.
243 -- land principally occupied by agriculture, with significant areas
of natural vegetation.
22× -- vineyards, fruit trees and berry plantations.
333 -- sparsely vegetated areas, 334 -- burnt areas.
211 -- non-irrigated arable land, 242 -- complex cultivation patterns.
11× -- urban fabric, 12× -- industrial, commercial and transport units,
13× -- mine, dump and constructions sites, 331 -- sands, 332 -- bare
rocks, 51× -- inland waters.
Based on crop rotation statistics.
191T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198
using LANDSAT TM data for both the time horizons
(Fig. 2). Then the slope lengths in the individual
parcels were calculated using the DMR data by r.flow
command in GRASS GIS (Mitasova et al., 1996). The
analysis of sample areas led to the average slope length
of 191 m for 1990 and 171 m for 2000 that have been
applied as constants in agricultural areas. The seminatural
and forest areas showed no change in average
slope lengths and a constant value of 185 m has been
used for both the times. Despite the fact that the average
value does not contain spatial variability in parcel
sizes and morphology, the sampling helped us to assess
the temporal changes in the L factor for the given
period. Manual delineation of all parcels in the territory
of Slovakia would not be feasible within a reasonable
time frame.
Land cover/management (C factor) takes into account
differences in properties of the vegetation cover and
methods of land management, reflecting their protective
function. The C factor was determined from the
CORINE land cover (CLC) database and crop rotation
statistical data for 72 Slovak districts.
Fig. 3. The mean annual C factor in arable land areas based on crop rotation data for districts. a) 1990, b) 2000.
192 T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198
The CLC database (SAZP, 2004) contains land cover
classes mapped using the LANDSAT TM images by
methodology and nomenclature unitedly defined for the
whole Europe at scale of 1:100,000 (Heymann et al.,
1994). Currently, two time horizons for Slovakia are
available describing the land cover in the 1989­1992
(CLC90) and 1999­2001 (CLC2000) periods. The
vector version of the CLC databases was used, later
converted to a raster format at 50-m resolution in
accordance with the other data layers used in this study.
The resolution and level of detail of the CLC databases
with a minimum mapping unit size of 25 ha and
minimum linear features width of 100 m are acceptable
for this type of regional analysis. Moreover, currently no
other land cover database with a complete coverage of
the territory of Slovakia is available.
The unified CLC methodology of mapping for both
databases allows to study temporal changes in land cover
and their impact on soil erosion processes. However, these
databases are not meant primarily for soil erosion
assessments, therefore the C factor values for some
CLC classes are based rather on rough estimates using
available literature (Cebecauer et al., 2004). The C factor
values were attributed individually to each of the mapped
land cover classes (Table 1), taking into consideration
their internal heterogeneity and its specific manifestation
Fig. 5. Total area of soil erosion risk categories in 1990 and 2000.
Fig. 4. The C factor change in the 1990­2000 period.
193T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198
in landscape within the natural conditions of Slovakia
(Feranec and Oťaheľ, 2001). Values of the C factor were
derived using the vegetation classes published by
Wischmeier and Smith (1978), Pasák et al. (1983) and
for the territory of Slovakia by Alena (1991) and Malíšek
(1992). Built-up areas, bare rocks and water surfaces were
excluded from the evaluation.
The C factor is dependent not only on the actual land
cover class, but it also reflects applied utilisation and
management practices (Cebecauer et al., 2004). The
crop rotation on arable land strongly influences the
long-term value of the C factor. Therefore the C factor
for the CLC arable land classes (211 -- non-irrigated
arable land and 242 -- complex cultivation patterns)
was derived using the published statistical data on the
mean crop rotation and the acreage of used agricultural
crops for 72 Slovak districts (Pflüger, 1998; Škultéty,
1999, 2000). The C factor for the individual areas of
classes 211 and 242 of the CLC databases was
determined as a weighted average of the mean annual
values of the C factor for the main crops in the districts.
To compute the mean annual C factor for 1990 and
2000, statistical data for the periods 1989­1991 and
1998­2000 were used, respectively. The average of
3 years was used to minimise the short-term (yearly)
variation in the crop rotation. The districts with the most
intensively utilised agricultural areas (mostly in lowlands)
and crops with a low protective function (such as
maize, sunflower, sugar beet and vegetables) have
higher values of the C factor than districts in
submountainous and mountainous areas (Fig. 3a, b).
This holds for both the times horizons, however, the C
factor for the CLC arable land classes increased in
southern regions in 2000, whereas northern regions,
mostly mountainous, have experienced a decrease in the
C factor. Overall, the average C factor in arable land
areas was slightly lower in 2000 (C=0.26036) than in
1990 (C=0.26135) due to changes in individual crop
shares.
The analysis of the C factor for all CLC classes in
1990 and 2000 showed that the CLC changes occurring
in 4.11% of the territory and the crop rotation changes in
Fig. 6. Map of soil erosion changes in the 1990­2000 period.
Table 2
The extent of spatial changes in soil erosion risk categories in the
1990­2000 period
Erosion risk change
interval
Area Total increase/decrease
[km2
] [%] [%]
b-75 8.7 0.00
-75­-22.6 85.4 0.17
-22.5­-7.6 424.9 0.87 12.6
-7.5­-0.8 5,688.5 11.61
-0.7­0.7 42,174.3 86.05 86.1
0.8­7.5 459.5 0.94
7.6­22.5 109.6 0.22 1.3
22.6­75 63.1 0.13
75N 5.4 0.01
Total area 49,010.6
194 T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198
arable land areas induced the C factor changes in
17,518 km2
(35.73%) of the territory. The C factor
decreased in 15.54% and increased in 20.20% of the
territory. Lowlands and hillylands in the south are
affected predominately by the increase in the C factor,
whereas submountainous and mountainous areas in the
north benefited mostly from the decrease in the C factor
with small area exceptions (Fig. 4). The overall
protective function of land cover slightly decreased
during the 1990 to 2000 period. Almost 65% of the
territory has not changed or experienced the land-cover
change with no impact on its protective function.
3. Results and discussion
The comparison of the total soil erosion risk
categories extent in 1990 and 2000 (Fig. 5) shows a
slight decrease in the high and very high risk categories
(239.90 and 74.21 km2
, respectively), decrease in the
low risk category (216.05 km2
) and increase in the none
or very low risk category (435.22 km2
) and moderate
risk category (32.75 km2
). Total areas for soil erosion
risk categories in Slovakia in 1990 and 2000 are almost
stable with only minor changes that mostly reduce soil
erosion risk in higher categories. However, this figure
shows only a cumulative effect of changes that does not
exhibit the full extent of spatial changes in soil erosion
risk during the considered period in Slovakia. The
comparison of soil erosion risk maps on a per-cell basis
gives us better information on the spatial extent and
intensity of undergone changes (Table 2). The identified
changes reveal that the affected area covers around
13.9% of the territory. A decrease in the erosion risk
category occurred in 12.6% of the territory and an
increase in the erosion risk category affected only 1.3%
of the territory.
The spatial extent of changes in soil erosion risk due
to considered factors has a character of small patches
distributed over the whole territory and exhibits strong
spatial differences (Fig. 6). The soil erosion risk changes
are the strongest in mountainous and submountainous
regions with high topographic potential for soil erosion.
In lowlands in the south-west and the south-east, the
studied changes had only a subtle impact on soil erosion,
even though the C factor increased in these areas (Fig. 4).
The unfavourable changes in C factor (especially in
arable land areas) were mitigated by the small susceptibility
of area to soil erosion given by the other factors.
The increase in erosion rates in mountainous and
submountainous areas is attributed mainly to the
ongoing local deforestation caused by timber production
and to the intensification of agricultural production in
vulnerable areas (e.g. conversion of pastures to arable
land). This type of land-use change stands for the most
dangerous changes as the significant decrease of the
protective function may initiate strong erosion processes.
For example, the area of 102 km2
of the moderate and
high erosion risk increase (N7.5 tonnes/ha/year) in the
northern part of Slovakia is related to the intensification
of agricultural production. The decreases in erosion risk,
mainly in the east, are attributed to the conversions of
arable land to semi-natural and forest land areas and
changes in the crop rotation system towards a higher
share of crops with a higher protective function against
soil erosion.
These changes with substantial regional differences
are also related to a diverse regional development under
varying socio-economic and environmental conditions.
Fig. 7. Example of soil erosion risk changes in agricultural area.
195T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198
The transformation of economy strongly hit the agriculture
sector. The animal production fell by more than 30%
during the studied period and consequently the acreage of
fodder crops decreased, resulting in the shifts in the
proportion of planted crops within the crop rotation
system. Thus the acreage of potatoes decreased by 50%,
fodder maize by 42% and annual and perennial fodder by
20% (source: Eurostat). The agricultural production in
areas with low soil production potential has shifted
towards extensive utilisation (e.g. pastures) or even land
abandonment followed by the succession of shrub -- tree
formations. The original large-parcel pattern of arable
land was locally changed to small parcels together
forming complex cultivation patterns. These changes
have increased the protective function of land cover and
reduced the soil erosion risk predominantly in submountainous
regions. On the other hand, the land ownership
changes (mainly via land restitution) induced many local
changes in the land utilisation often with a significant
decrease in the protective function of land cover (e.g.
areas of pastures were changed to arable land (Fig. 7)).
Forests have the highest protective function against
soil erosion. The comparison of CLC databases for the
considered time horizons showed deforestation in large
areas of Slovakia. During the 1992­2000 period, the
roundwood production had increased 22% (from 4.755
to 5.788 million m3
/year; source: Eurostat). This trend is
the consequence of large timber production exports as
new owners seek quick profits. The emergence of
clearcuts in the areas susceptible to soil erosion resulted
in the soil erosion risk increase (Fig. 8), predominately
in the central and northern parts of Slovakia (Fig. 6). On
the other hand, the afforestation of transitional woodland-scrub
areas altered the protective function of the
land cover positively towards the reduction of soil
erosion risk (Fig. 8). The overall result of these
contradictory processes leads to a slight increase in
soil erosion risk in forestland areas.
From the results is clear, that the transformation of
political system followed by the transformation of
society, and economic environment may induce significant
transformation of the land utilisation with
consequences on the soil erosion risk pattern. The
regional disparities in socio-economic status can be
increased by the inappropriate land management
practices in vulnerable areas with generally low soil
production potential. In these areas, the increase of soil
erosion may lead to even a further loss of soil production
potential with a negative impact on the local economy
and overall regional development. In the next decades
this may induce further land-use changes towards land
abandonments and afforestations.
Fig. 8. Example of soil erosion risk changes in semi-natural and forest areas.
196 T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198
4. Conclusions
Our study showed that land-cover changes in Slovakia
over the period 1990­2000 lead to changes in soil erosion
risk with a mixed character. While some parts of Slovakia
have experienced an increased protective function of land
cover, other, dominantly mountainous, parts were affected
by deforestation, natural disasters and conversion to
arable land at locations that are inherently erosion prone
and showed increased erosion risks. However, the overall
effect of land-cover change over the 1990­2000 period
has been slightly positive, lowering soil erosion risks in
most areas across the country. The analysis of regional
differences in land cover and crop rotation system showed
main causes of changes and helped the explanation of
ongoing changes in soil erosion pattern in Slovakia.
The changes in land cover and consequently in soil
erosion are primarily associated with human activities
and changes in landscape utilisation. We expect that a
recent accession of Slovakia to the EU and implementation
of common agricultural policy will lead to further
land-cover changes that will have positive effects with
increased protective function of land cover and thus
reduced erosion rates. However, the global climate
changes may offset these effects at least partially. The
complex interaction of evolving factors will be a subject
of our further study.
Acknowledgement
The authors wish to thank the reviewers for their
comments and suggestions that significantly improved
the original manuscript. This research was supported by
VEGA Grant Agency of the Slovak Republic within the
scientific project No. 1/3049/06 and Slovak Research
and Development Agency under the contract No.
COST-0016-06. Authors also thank the GeoModel s.r.
o. company for providing the digital model of relief of
Slovakia.
References
Alena, F., 1991. Protierózna ochrana na ornej pôde. Štátna melioračná
správa, Bratislava.
Bakker, M.M., Govers, G., Kosmas, C., Vanacker, V., Oost, K.,
Rounsevell, M., 2005. Soil erosion as a driver of land-use change.
Agriculture, Ecosystems & Environment 105, 467­481.
Boellstroff, D., Benito, G., 2005. Impacts of set-aside policy on the
risk of soil erosion in central Spain. Agriculture, Ecosystems &
Environment 107, 231­243.
Cebecauer, T., Šúri, M., Hofierka, J., Fulajtár, E., 2004. CORINE land
cover in the context of soil erosion assessment at a regional scale.
In: Mohaupt-Jahr, B., Keil, M., Kiefl, R. (Eds.), CORINE LAND
COVER 2000 in Germany and Europe and its use for environmental
applications. Proceedings, Berlin 20­21 January,
Umweltbundesamt, pp. 74­80.
Cebecauerova, M., Cebecauer, T., in press. The assessment of major
landscape change types in the southern part of Borska nizina
lowland and Male Karpaty mountains during the period 1954­
2003. Ekologia (Bratislava).
Feranec, J., Oťaheľ, J., 2001. Land cover of Slovakia. Veda, Bratislava.
Feranec, J., Šúri, M., Oťaheľ, J., Cebecauer, T., Kolář, J., Soukup, T.,
Zdeňková, D., Waszmuth, J., Vajdea, V., Vijdea, A.M., Nitica, C.,
2000. Inventory of major landscape changes in the Czech
Republic, Hungary, Romania and Slovak Republic 1970s­1990s.
International Journal of Applied Earth Observation and Geoinformation
2, 129­139.
Feranec, J., Otahel, J., Cebecauer, T., Machkova, N., Novacek, J.,
2005a. Land cover changes of Slovakia in period 1990­2000. In:
Kleinn, Ch., Nieschulze, J., Sloboda, B. (Eds.), Schriften aus der
Forstlichen Fakultät der Universität Gottingen und der Niedersächsichschen
Forstlichen Versuchsanstalt, 138. Frankfurt am Main
(J. D. Sauerländers Verlag), pp. 139­147 (http://www.ggrs.uni-
goettingen.de/ggrs2004/CD/Applications_in_Forestry/bd138.pdf).
Feranec, J., Otahel, J., Novacek, J., 2005b. Changes of agricultural
landscape in the administrative regions of Slovakia in 1990­2000.
In: Florianczyk, Z., Czapiewski, K. (Eds.), Rural areas and
development, 3. Warsaw (Institute of Agricultural and Food
Economics and Institute of Geography and Spatial Organization),
pp. 153­165.
Fulajtár, E., Čurlík, J., 1980. Pôdne druhy, skeletovitosť a zamokrenie.
In: Mazúr, E. (Ed.), Atlas SSR Atlas SSR. Geografický ústav SAV,
Úrad geodézie a kartografie, Bratislava, pp. 75­77.
Fullen, M.A., Booth, C.A., Brandsma, R.T., 2005. Long-term effects
of grass ley set-aside on erosion rates and soil organic matter on
sandy soils in east Shropshire, UK. Soil & Tillage Research 89,
122­128.
Gobin, A., Kirkby, M., Govers, G., 2004. Pan-European soil risk
assessment. In: Francaviglia, R. (Ed.), Agricultural Impacts on Soil
Erosion and Soil Biodiversity: Developing Indicators for Policy
Analysis. Proceedings from an OECD Expert Meeting - Rome,
Italy, March 2003, pp. 1­15.
Haan, C.T., Barfield, B.J., Hayes, J.C., 1994. Design hydrology and
sedimentology for small catchments. Academic Press, San Diego.
Heymann, Y., Steenmans, Ch., Croisille, G., Bossard, S., 1994.
CORINE land cover: Technical guide. Office for Official
Publications of the European Communities, Luxembourg.
Hofierka, J., Šúri, M., 1996. Modelling spatial and temporal changes in
soil water erosion. Geografický časopis 48, 255­269.
Hofierka, J., Šúri, M., Cebecauer, T., 1998. Rastrové digitálne modely
reliéfu a ich aplikačné možnosti. Acta Facultatis Studiorum
Humanitatis et Naturae Universitatis Prešoviensis, Prírodné vedy
30, 208­217.
Hofierka, J., Parajka, J., Mitasova, M., Mitas, L., 2002. Multivariate
interpolation of precipitation using regularized spline with tension.
Transactions in GIS 6, 135­150.
Jordan, G., Van Rompaey, A., Szilassi, P., Csillag, G., Mannaerts, Ch.,
Woldai, T., 2005. Historical land use changes and their impact on
sediment fluxes in the Balatom basin (Hungary). Agriculture,
Ecosystems & Environment 108, 119­133.
Kirkby, M.J., Le Bissonais, Y., Coulthard, T.J., Daroussin, J.,
McMahon, M.D., 2000. The development of land quality
indicators for soil degradation by water erosion. Agriculture,
Ecosystems & Environment 81, 125­135.
Kirkby, M.J., Jones, R.J.A., Irvine, B., Gobin, A., Govers, G., Cerdan,
O., Van Rompaey, A.J.J., Le Bissonnais, Y., Daroussin, J., King,
197T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198
D., Montanarella, L., Grimm, M., Vieillefont, V., Puigdefabregas,
J., Boer, M., Kosmas, C., Yassoglou, N., Tsara, M., Mantel, S., Van
Lynden, G.J., Huting, J., 2004. Pan-European Soil Erosion Risk
Assessment: The PESERA Map, Version 1 October 2003. Explanation
of Special Publication Ispra 2004 No.73 (S.P.I.04.73).
European Soil Bureau Research Report No.16, EUR 21176, 18pp.
and 1 map in ISO B1 format. Office for Official Publications of the
European Communities, Luxembourg.
Malíšek, A., 1990. Zhodnotenie faktora eroznej ucinnosti privalovej
zrazky. Geografický časopis 42, 410­422.
Malíšek, A., 1992. Optimalna dlzka svahu v zavislosti na vodnej erozii.
Vedecké práce Výskumného ústavu pôdnej úrodnosti 17, 203­220.
Mitasova, H., Hofierka, J., Zlocha, M., Iverson, R.L., 1996. Modeling
topographic potential for erosion and deposition using GIS.
International Journal of Geographical Information Systems 10,
629­641.
Molnár, D.K., Julien, P.Y., 1998. Estimation of upland erosion using
GIS. Computer & Geosciences 24, 183­192.
Neteler, M., Mitasova, H., 2004. Open Source GIS: A GRASS GIS
Approach, 2nd ed. Kluwer Academic Publishers, Boston.
Pasák, V., Janeček, M., Šabata, M., 1983. Ochrana zemědelské půdy
před erozí. Metodiky pro zavádění výsledků výzkumu do
zemědelské praxe 11. Ústav vědeckotechnických informací pro
zemědělství, Praha.
Pflüger, A., 1998. Register of the acreage seeded by agricultural crops
at May 25, 1998 (in Slovak). Štatistický úrad Slovenskej republiky,
Bratislava.
Renard, K.G., Foster, G.R., Weesies, G.A., McCool, D.K., Yoder, D.C.,
1997. Predicting soil erosion by water: a guide to conservation
planning with Revised Universal Soil Loss Equation (RUSLE).
Agric. Handb. 703. U.S. Gov. Print. Office, Washington, DC.
SAZP, 2004. Corine land cover. SAZP, Slovak Republic.
Schoorl, J.M., Veldkamp, A., 2001. Linking land use and landscape
processes modelling: a case study for the Álora region (south
Spain). Agriculture, Ecosystems & Environment 85, 281­292.
Solin, L., Cebecauer, T., 1998. Vplyv kolektivizacie polnohospodarstva
na vodnu eroziu pody v povodi Jablonka. Geografický
časopis 50, 35­57.
Souchere, V., King, Ch., Dubreuil, N., Lecomte-Morel, V., Le
Bissonnais, Y., Chalat, M., 2003. Grassland and crop trends: role
of the European Union Common Agricultural Policy and
consequences for runoff and soil erosion. Environmental Science
& Policy 6, 7­16.
Šamaj, F., Valovič, Š., 1980. Intensity of 15-minute rainfall. In: Mazúr,
E. (Ed.), Atlas SSR. Geograficky ustav SAV, Urad geodezia a
kartografie, Bratislava, p. 63.
Škultéty, I., 1999. Register of the acreage seeded by agricultural crops
at May 25, 1999 (in Slovak). Štatistický úrad Slovenskej republiky,
Bratislava.
Škultéty, I., 2000. Register of the acreage seeded by agricultural crops
at May 25, 2000 (in Slovak). Štatistický úrad Slovenskej republiky,
Bratislava.
Šúri, M., Cebecauer, T., Hofierka, J., 1997. Tvorba digitálneho modelu
reliéfu Slovenskej republiky. Geodetický a kartografický obzor 43,
257­262.
Šúri, M., Cebecauer, T., Hofierka, J., Fulajtár, E., 2002. Soil erosion
assessment of Slovakia at a regional scale using GIS. Ekológia
(Bratislava) 21, 404­422.
Van Dijk, P.M., 2001. Soil erosion and associated sediment supply to
rivers. Seasonal dynamics, soil conservation measures and impacts
of climate change. Ph.D. Thesis, University of Amsterdam. The
Netherlands.
Van Rompaey, A., Govers, G., 2002. Data quality and model
complexity for regional scale soil erosion prediction. International
Journal of Geographical Information Science 16, 663­680.
Van Rompaey, A., Govers, G., Van Hecke, E., Jacobs, K., 2001. The
impacts of land use policy on the soil erosion risk: a case study in
central Belgium. Agriculture, Ecosystems and Environment 83,
83­94.
Van Rompaey, A., Krasa, J., Dostal, J., Govers, G., 2003. Modelling
sediment supply to rivers and reservoirs in Eastern Europe during
and after the collectivisation period. Hydrobiologia 494, 169­176.
Wischmeier, W.H., 1959. A rainfall erosivity index for a Universal Soil
Loss Equation. Soil Science Society of America Proceedings 23,
246­249.
Wischmeier, W.H., Smith, D.D., 1958. Rainfall energy and its relation
to soil loss. Transactions of American Geophysical Union 39,
285­291.
Wischmeier, W.H., Smith, D.D., 1978. Predicting rainfall erosion
losses -- a guide to conservation planning. Agriculture Handbook,
No. 537. US Department of Agriculture, Washington DC.
Zachar, D., 1982. Soil erosion. Elsevier, Amsterdam.
198 T. Cebecauer, J. Hofierka / Geomorphology 98 (2008) 187­198