Carpathian Journal of Earth and Environmental Sciences, November 2013, Vol. 8, No. 4, p.

195 - 208

ASSESSMENT OF CHANGES IN LAND-USE AND LAND-COVER

PATTERN IN ROMANIA USING CORINE LAND COVER DATABASE

Elena-Ana POPOVICI, Dan BĂLTEANU & Gheorghe KUCSICSA

Institute of Geography, Romanian Academy, Dimitrie Racoviţă Str., no. 12, sector 2, Bucharest, RO-023993,
e-mail: [email protected]; [email protected]; [email protected]

Abstract: Romania has a wide range of landforms, land use and land cover categories being unevenly
distribute. However, agricultural land and forest land are prevailing. Important spatial changes related to
land use dynamic occurred in the 1990s (transition period) and got momentum after 2000. One of the
major changes during the transition period was the expansion of private ownership over agricultural and
forest land against the collective and state property that has been characteristic of the previous period. So,
the large collective and state farms from the communist period gave way to small, peasant-type family
farms, leading to noticeable fragmentation of the agricultural land. As to the land use pattern, two main
directions of change have been of interest. Firstly, the intensification and the extensification of the
agricultural systems, mostly in the plains and plateaux and secondly, the afforestation and deforestation,
largely in the mountains and plateaux have been described. The pack of urbanisation and industrialisation
slowed down, yet still on going in the large urban centres as Bucharest, Timișoara, Iași, Constanța, Cluj
Napoca etc., and in their neighbourhood. The present study aims to analyse the temporal and spatial
changes of the main land use and land cover categories by resorting to CORINE Land Cover (CLC)
database, while identifying and looking into the main factors controlling the evolution of land-use/cover
change. Insights into the regional level change resulted in studying it by major relief units, namely, the
Carpathian Mountains, hills, plateaux and plains, as well as the Danube Delta.

Keywords: Land-use and land-cover changes; CORINE Land Cover database; Romania.

1. INTRODUCTION considered relevant information source on landscape

processes (Weber, 2007). Timely detection and
Land-use and land-cover changes are central precise information about land-use and land-cover
to the interests of the global environmental research, changes are extremely important for understanding
especially in the regions with high dynamics of the the relationships and interactions between human
social and environmental pressures. This significant and natural phenomena if the management of
cumulative type of change influences and is affected decision-making is to be improved (Lu et al., 2004).
by climate change, loss of biodiversity, and To this end, remote sensing and GIS software are the
sustainability of human-environment interactions best methods to study the spatial distribution and
(Lambin et al., 1999; Lambin & Geist, 2006). Over evolution of geographical phenomena related to
the past decades, land-use and land-cover changes land-use and land-cover changes. CORINE Land
have become a priority research-topic of Cover database contains useful information for
international projects, given their implications for detecting land-use and land cover changes, as well
global environmental change (Turner et al., 1993). as for constructing land cover account. Land cover
The main changes, basically conversion and represents a concrete set of natural and
modification, considered to be cumulative, may have anthropogenic features, resulting largely from its use
detrimental local and regional environmental effects (Haines-Young & Weber, 2006) and of being an
on soil, vegetation, waters and air, and if indivisible part of the landscape (Feranec et al.,
cumulating, the quality of these factors at global 2010). Based on GIS processing of the CORINE
level could be degraded. Land Cover data layer some studies at national, as
Land-use and land-cover changes can be well as regional level have been achieved (Feranec

195
et al., 2000, 2006, 2007, 2010; Feranec & Otahel, referring only to regional level (Kuemmerle at al.,
2001; Kuemmerle et al., 2006; Otahel et al., 1993, 2009; Lakes et al., 2009; Müller & Kuemmerle,
2002; Willems et al., 2005, etc.). 2009; Dutca & Abrudan, 2010; Popovici, 2010;
The fall of the communist regime has led to a Grigorescu et al., 2012; Knorn et al., 2012).
series of radical political and socio-economic
changes in many Central and East European 2. STUDY AREA
countries. Transition in the economy meant the
replacement of the old centralised system by the free Romania is situated in the south-eastern part of
market system (Bălteanu et al., 2004, 2005). In the Central Europe at the contact with Balkan Europe
previous studies on assessing and quantifying land- (south of the Lower Danube), and at the crossroads
use and land-cover changes, in the context of post- of the major European thoroughfares (Niculescu,
communist period (Ptáček, 2000; Bičík et al., 2001; 2006). With an area of 238,391 km2 and 19,042,936
Zemek et al., 2005; Kuemmerle et al., 2006; inhabitants (Population Census 2011), Romania
Václavík & Rogan, 2009; Bičík & Jeleček, 2009, ranks among the medium-sized European states, but
Szilassi et al., 2010 etc.), it is generally recognized is the largest country in the south-east of Central
that the socio-economic and political conditions of Europe. The moderate temperate-continental climate
the transition period represent the major drivers of and the varied landforms favour a great diversity of
these changes. The socio-economic and political the vegetation cover. The mountains, hills and
changes have affected land use in Romania after plateaux are dominantly covered by the forest belt
1990 through a restructuring process and basic (coniferous and broad-leaved forests), while the
mutations in the ownership of land increasing the alpine and subalpine meadows are developed in the
influence of market forces in shaping the rural highland. In the plain areas the greatest part of oak
landscapes. Some of the negative effects of the Act forests, steppe and sylvo-steppe vegetation have
18/1991 resulted in the excessive fragmentation of been replaced by croplands.
farmland and in the marked degradation of Romania is one of the European countries
production services in agriculture and land quality with important land resources (0.68 ha agricultural
(Bălteanu & Popovici, 2010). The new land-use land and 0.43 ha arable land per capita). In terms of
system that emerged was by far more vulnerable to structure, agricultural land represents 61.2% of
extreme environmental perturbations and less Romania’s area, forest land 28.5%, terrains under
resilient (Fraser & Stringer, 2009). Moreover, water and ponds 3.5%, built-up areas 3.1%, roads
transition has led to significant changes in the land and railways 1.6% and degraded and unproductive
use/land cover pattern by conversion from one grounds 2.1% (Romanian Statistical Yearbook
category to another. The wide range of factors 2012). According to CORINE Land Cover database
responsible for these changes comprises the political (2006), agricultural areas amounted 58%, forest and
associated with economic, technological, semi-natural areas 33%, artificial surfaces 6%,
demographic and natural ones. Political and wetlands and water bodies 3% (Fig. 1).
institutional drivers have in time been directly
involved in land-use and land-cover change, 2.1. The geographical distribution of land
affecting primarily the regime of property rights and use and land cover categories
the decision-making systems involved in the
management of natural resources (Lambin et al., The diversity of natural conditions, as well as
2003). In Romania, the four major land reforms the general and regional particularities of a social and
enacted in the 19th and 20th centuries (1864, 1918- economic history, have resulted in land-fund structure
1921, 1945 and 1991) had decisively influenced with agricultural terrains having the highest
subsequent land-use dynamics, access and control of percentage (over 61%). The agricultural land use
land, as well as social relationships (Bălteanu et al., categories (arable, pastures and hayfields, vineyards
2004). The type and amplitude of modifications and orchards) hold different shares in every landform
varied in terms of the socio-economic and political unit. Hence, in the plain regions, below 200 m a.s.l.
conditions of each period. The general trend (the Romanian Plain and the Banat and Crişana Plain)
consisted in the enlargement of agricultural land and agricultural land represents over 80% of the total, 60-
built-up areas to the prejudice of forestland and 70% in the plateaux and about 54% in the hillsides
grassland. The significance of this study consists in (the Banat and Crişana Hills and the Subcarpathians).
the detailed analysis of the land-use/land cover The lowest percentages are found in the mountains
change pattern at national level, few such studies (12.1%) and in the Danube Delta (only 23.5% total
having been performed so far, and the existing ones surface-area) (Fig. 1).

196
 Terrains covered with natural and semi- mining waste-dumps, household refuse and
natural vegetation (forests and natural grassland) industrial waste etc. That land occupies 6.3% of the
represent 33.6% of the national territory. Their country’s territory in all types of landform, covering
distribution is inversely proportional to agricultural between 1.15% in the Danube Delta and 9.28% in
land. Thus, the largest forests occur in mountainous the hillsides.
and hilly regions (68.7% and 34.8%, respectively of Wetlands have the widest spread in the
their overall areas). With the decrease of altitude, Danube Delta, representing together with water
soils and climatic conditions become more surfaces, over 75% of the Delta area. Other notable
propitious to the development of farming land, wetlands are Balta Mică from Brăila located in the
forested areas shrinking to 22.5% in the hillsides and Danube Floodplain, and more than 30 such areas
to about 6% in the plains. alongside the main big rivers (the floodplains of the
Land severely modified by man (Artificial Danube, Prut, Siret, Buzău Mureş, the Three Criş
land) includes settlements (rural and urban), various Rivers, Someş, Timiş with its tributary the Bega
agricultural or industrial constructions (nurseries, etc.), oligotrophic and eutrophic boggy soils in the
silos, industrial estates etc.), commercial units, Eastern Carpathians, etc.
sporting and leisure facilities, roads and railways,

Figure 1. Distribution of the land use and land cover categories by major landform units (CORINE Land Cover, 2006)

197
 3. DATA AND METHODS Additionally, statistical data supplied by the
National Institute of Statistics (Romanian Statistical
There are two major types of land-use and Yearbooks 1990-2010; General Agricultural Census
land-cover changes, namely: conversion (that is 2002, 2010; Agricultural Farm Survey 2005;
radical changes involving the replacement of one TEMPO-Online database, etc.), as well as field
type of land cover by another) (Meyer & Turner II, surveys were taken into consideration.
1994, Turner & Meyer, 1991), and modification Changes in the land use and land cover pattern
(that is, maintaining a certain type of intensive use were detected for two relevant time-intervals in
without changing the terrain’s attributes) (Gregorio Romania (1990-2000 and 2000-2006). The first
& Jansen, 2000). The authors of the present paper interval (1990-2000) represents the period of
had in view to identify and analyse the first category transition to the market economy. The second
of changes in land use and land cover, thus interval (2000-2006) largely overlaps that of pre-
highlighting the main land cover flows (LCF) that accession to the European Union, when important
took place in Romania over the last two decades. land-use changes, mainly processes of
The concept of LC flows was applied in many intensification/extensification of agriculture,
studies by Stott & Haines-Young (1998), Haines- deforestation and urbanization took place.
Young & Weber (2006), Feranec et al. (2000, 2010). CORINE Land Cover is a database of the
The land-cover changes analysed herein represent European environmental landscape derived from the
the result of the main conversion processes that took interpretation of satellite imagery: Landsat -4/5 TM
place between the 15 classes of the second CLC data (in a few cases, Landsat MSS) and SPOT 2/3 (for
level (Table no. 1), the losses of initial land cover CLC 1990); Landsat -7ETM (for CLC2000) and
(consumption) and the making of new land type SPOT4 and/or IRS LISS III images (for CLC 2006)
(formation) (Weber, 2007). According to the (Bossard et al., 2000; Feranec et al., 2012). The
previous methodology, seven land-cover flows were complete classification, nomenclature and
identified and analysed: urbanisation and methodology are available in the official CORINE
industrialisation (conversion from agricultural and portal (http://www.eea.europa.eu/publications). A
natural land to artificial land); intensification of number of 39 out of the 44 classes in the CLC
agriculture (internal conversion of agriculture from nomenclature were identified in Romania, and
lower-to-higher intensity of use, and also conversion grouped into 15 classes (level 2) (Table 1).
from natural land (32, 33) to agriculture);
extensification of agriculture (internal conversion of 3.2. Data uncertainty
agriculture, transition of classes, associated with
higher-to-lower intensity of use); agricultural land Considering the minimum mapping area (25
abandonment (conversion of agricultural land to ha), the accuracy provided by CORINE Land Cover
semi-natural areas); deforestation (conversion from for identifying a certain type of conversion is not
forest land to other LU categories); afforestation high enough to catch all locations affected by land
(forest regeneration, transition from agriculture, use/land cover change. Since satellite images are not
semi-natural vegetation and wetlands to forests); received within the same period of the year,
water bodies construction and management sometimes a gap of several months existing among
(conversion from agriculture and natural lands to them, certain plant species are in different stages,
water bodies). hence they give a distinctive spectral response that
may induce possible errors of interpretation (Ursu et
3.1. Data al., 2006). Also, the resolution of the satellite images
differs over the two time-periods analysed: ≤50 m
Two sets of spatial data underlie the basic for CLC 1990 and ≥25 m for CLC 2000 and CLC
information used in this study: 2006. Land-use changes included vast water-covered
1) Data on land use and land cover structure areas, especially arable terrains, left after the 2005-
and data on land use and land cover changes 2006 floods had affected the western part of
(classification level 3) elaborated under the Romania (Banat-Crisana Plain) and the Danube
CORINE Programme (European Environment Floodplain. As a result, confusions may arise in
Agency; www.eea.europa.eu); analysing certain land-cover flows (e.g. water bodies
2) Data on major landform units (Romania. construction and management), as well as a certain
Space, Society, Environment, 2006) used in the incertitude in the visual interpretation of satellite
regional analysis of land-use changes. images (vectorization accuracy).

198
 Table 1. CORINE Land Cover nomenclature (identified in Romania)
LEVEL 1 LEVEL 2 LEVEL 3
111 Continuous urban fabric
11 Urban Fabric
112 Discontinuous urban fabric
121 Industrial or commercial units
12 Industrial, commercial and 122 Road and rail networks and associated land
1 Artificial surfaces transport units 123 Port areas
124 Airports
131 Mineral extraction sites
13 Mine, dump and construction
132 Dump sites
sites
133 Construction sites
14 Artificial, non-agricultural 141 Green urban areas
vegetated areas 142 Sport and leisure facilities
211 Non-irrigated arable land
21 Arable land 212 Permanently irrigated land
213 Rice fields
221 Vineyards
22 Permanent crops
2 Agricultural areas 222 Fruit trees and berry plantations
23 Pastures 231 Pastures
241 Annual crops associated with permanent crops
242 Complex cultivation patterns
24 Heterogeneous agricultural
243 Land principally occupied by agriculture, with
areas
significant areas of natural vegetation
244 Agro-forestry areas
311 Broad-leaved forests
31 Forests 312 Coniferous forests
313 Mixed forests
321 Natural grasslands
32 Scrub and/or herbaceous 322 Moors and heathland
3 Forest and semi- vegetation associations 323 Sclerophyllous vegetation
natural areas 324 Transitional woodland-scrub
331 Beaches, dunes, sands
33 Open spaces with little or no 332 Bare rocks
vegetation 333 Sparsely vegetated areas
334 Burnt areas
4 Wetlands 411 Inland marshes
41 Inland wetlands
412 Peat bogs
42 Maritime wetlands 421 Salt marshes
511 Water courses
51 Inland waters
512 Water bodies
5 Water bodies 521 Coastal lagoons
52 Marine waters
523 Sea and ocean

3.3. Description of GIS technique CLC level 2 (Fig. 2), used to establish the 8 flows
previously described. The regional analyses
The analysis has been facilitated by the consisted in intersecting the layer of change
Geographical Information System (GIS). In this (generalized form, level 2) with that designating the
case, land use and land cover changes are developed boundary of major landform units in Romania:
under ArcMap 9.3 and, additionally, in Excel. mountains, hills, plateaux and lowland areas, as well
Initially, the land-use changes layer was re- as the Danube Delta.
projected from ETRS_1989_LAEA_52N_10E into The obtained information allowed us to assess
Stereo-70 system of coordinates, suitable to our changes in space-and-time, identify the types of
study. A number of 126 types of change were change, build the matrix of change (flows), evaluate
identified across Romania during 1990-2000, and 71 loss-and-gain within each land-use class (level 2),
over 2000-2006 (CLC classification level 3). the rate of change and intensity of flows.
Subsequently, these changes were generalised to

199
 Figure 2. The generalisation of land-use and land cover changes spatial data classification level 3 to level 2

4. RESULTS AND DISCUSSIONS

Two distinct studying periods have been
4.1. Detection of land-use and land cover distinguished:
changes 1) The 1990-2003 transition, marked by basic
changes in agriculture, when the collective and state
December 1989 in Romania marked the property was replaced by private property (Fig. 3),
beginning of a period of transition to the market and
economy, a new stage in the evolution of agriculture 2) The post-transition period (2003-to-date),
and implicitly in land use (Popescu, 2001). The most corresponding to Romania’s pre- and post-accession
important changes of that period appeared in the to the European Union, associated to several land-
space dynamics of the main land use/cover use changes connected with the adoption and
categories and their quality, a new type of landed implementation of the Common Agricultural
property and land exploitation (Popovici, 2008, Policies (CAP).
2010).

Figure 3. Land fund by categories of use and form

Figure 4. Gains and losses by land-use and land cover categories over the 1990-2000 and 2000-2006 intervals

200
 The political and socio-economic
changes begun in the early 1990s had a
strong impact on lands, especially on the
quantity and quality of agricultural
terrains. Hence, over the 1990-2006
period, a total of 354,765 ha (1.48% of the
country’s area) underwent changes (Fig.
5), at an annual rate of 0.09%. By 2000,
the scale of the phenomenon was higher
(279 thousands ha and 0.11% annual rate
of change) than after 2000 (76 thousands
ha and 0.05% annual rate of change).
Among the land-use categories with
a negative record in both study periods
(1990-2000 and 2000-2006) were
permanent crops (vineyards and orchards),
their area shrinking by over 27,099 ha.
Then follows the pastures (13,633 ha),
open spaces with little or no vegetation
(143 ha) and inland marshes (376 ha). On
the other hand, the built-up areas,
heterogeneous agricultural areas and scrub
and/herbaceous vegetation associations
have significantly increased (Fig. 4).
In the first period, forests, shrub
and/herbaceous vegetation associations
and agricultural land use classes (21, 22,
23 and 24) registered major spatial
changes, while in the second period it
was classes 31 and 32 (forests and shrub
and/herbaceous vegetation associations)
that had the highest gains and losses (Fig.
4). Looking at the change detection
matrix (Fig. 5), the most frequent land
use/cover flows in Romania over the
1990-2006 were deforestation,
afforestation, extensification and
intensification of agriculture (Fig. 6).
Processes of urbanisation and
industrialisation would develop largely at
the expense of the agricultural land-use
categories (arable land, heterogeneous
agricultural areas, pastures and
permanent crops).
Deforestation. The retrocession of
forest land contributed to illegal logging,
especially active over the past ten years.
In 2008, more than 45% of the forest
land was in private property. The high
number of owners with small forest plots,
as well as the uncertainty of maintaining
these lands led to massive deforestations
(illegal logging), especially on the
Figure 5. Change detection matrix for the 1990-2006 period (hectares)
privately-owned terrains.

201
 In this way, the forested area shrank by up to property) were afforested, the cost being covered
39,406 ha. During the first period (1990-2000), from the Land Reclamation Fund (Ministry of
gains were higher than losses. The losses were by far Environment and Climate Change, 2010). Since
greater than gains in the second period (2000-2006) 2000, non-reimbursable European funds contributed
as a result of implementing the Act no.1/11 January to tree-planting on over 25,000 ha (2009). Measure
2000 (Fig. 4), when deforestation process 3.5 of the SAPARD Programme (Special Accession
represented about 74% of total changed area. Most Programme for Agriculture and Rural Development)
conversion phenomena involved forest land and stipulated the funding of afforestation projects. In
transitional woodland-shrub classes. The forested 2009, Measure 221 of the National Rural
area lost over 130,967 ha in favour of the latter Development Programme had in view the First
classes and about 1,607 ha in favour of the artificial Afforestation of Agricultural Lands.
classes (Fig. 5). Since 2000, natural areas were Farming underwent significant spatial changes
declared protected and deforestation practices on either in the land-use agricultural categories (internal
their territory somehow diminished. conversions) or between these categories and other
According to CORINE Land Cover database land-use classes, such as built-up terrains, forest land
(2006), the forested area of Romanian’s national and etc. (Fig. 5). The expansion of private property, the
natural parks amounts 720,495 ha (about 10% of all excessive fragmentation of agricultural land, the
forest land in this country) mostly in the Carpathian high percentage of very small subsistence farms, the
Mountains (8.6% of all of Romania’s forests). Since dismantling of big animal breeding farms and the
most protected areas had been delimited beginning degradation of production services in agriculture
with the year 2000, the trend of deforestation rate have contributed to the mitigation of cultivated
has decreased mostly inside national parks rather areas, the abandonment of pastures and permanent
than outside their boundaries. Thus, logging in crops over large surfaces. All these resulted in the
Romania’s national and natural parks fell by 36.8% enhancing of extensive agriculture, basically
(from 7,637 ha over 1990-2000 to 4,824 ha in 2000- conversion from agricultural classes with higher-to-
2006) (Fig. 7), and by only 22.6% in unprotected lower intensity of use (ex. arable and permanent
areas (from 66,865 ha to 51,753 ha). crops turned into pastures).
As forested area began shrinking, the quality The processes of extensification and
of land started deteriorating significantly through intensification of agriculture were analysed only in
torrential events, erosion, landslides and the terms of quantity and not of quality (inputs in
intensification of extreme climatic phenomena: agriculture were not taken into consideration).
floods, snowstorms and droughts. In the lowland During the post-communist time, the cultivated area
regions, climate change, the systematic destruction decreased from 9.8 million ha in 1989 to 7.8 million
of irrigation systems and the cutting of protection ha in 2009 because much of the arable land (11.0
forest belts (wind breaks) have facilitated the onset million ha over the 1990-2009 period) remained
of frequent and lengthy dry periods that had negative uncropped (Bălteanu & Popovici 2010).
effects on crop production, the environment and Extensification of agriculture gets more
living conditions, generally. prominent than intensification, it representing over
Afforestation is the second significant process 16% of the total changed area involving conversion
that took place in Romanian during the studied from arable land and permanent crops to pastures and
period, it representing over 26% of the total changed heterogeneous agricultural surfaces. In this period,
area (Fig. 6). The expansion of the forest area is due about 25,141 ha of arable land and 14,604 ha of
primarily to natural regeneration, particularly in the permanent crops were turned into low-productive
mountainous and Subcarpathian regions. This classes (23 and 24). The permanent crops area had
process unfolded largely on deforested terrains, but steadily shrunk mostly through abandonment or
also on abandoned farm land, pastures in particular clearing after having been recovered by their former
which developed in the wake of declining owners. At the same time, the new plantations
shepherding. The most common type of transition covered usually small, dispersed areas, most of them
was transitional woodland-shrub to forests (93,290 around peasant households. Also, in the transition
ha) (Fig. 5). period, the conversion of arable land to pastures
Artificial reforestation (planting after logging, resulted from the abandonment of the former,
calamities or establishment of a new forest on particularly in low-productive regions. If arable land
degraded lands) took place on small areas, directly stays fallow for several years, it gets covered with
depending on financial sources. In the years 1990- vegetation; since identifying them on satellite images
2000, 5,200 hectares of degraded land (public state is sometimes difficult, they are listed under grassland.

202
 Figure 6. Main land use and land cover flows in Romania over the 1990-2006 period

Figure 7. Disparities in the process of deforestation inside and close to a few national and natural parks
in Romania during the two studied periods.

203
 Several other causes for abandoning arable rural areas. In the former case, the main land-use
land is their high degree of fragmentation, money changes consist in the conversion of farming terrain
shortage with the small farmers, the absence of into built-up areas (Grigorecu et al., 2012), while in
markets to sell the products, few, if any, irrigation the latter case, farming land is abandoned because of
systems etc. Pastures and hayfields would expand to poverty, external migration in search for jobs and
the detriment of vineyards and orchards, lose in severe ageing of the rural population.
surface-area in favour of arable land (intensification Agricultural land abandonment was much
process), especially in the post-transition period. more obvious in the Subcarpathians and the
Intensification of agriculture represents about mountains, regions traditionally engaged in animal
14% of the total changed area (Fig. 6). In Romania’s husbandry, and once the animal stock was shrinking
transition period, the intensification of agriculture large pastures and hayfields were abandoned, their
was more reduced (15.8%) than extensification place being taken by semi-natural vegetation (class
(20%), because many arable lands and permanent 32). During the 1990-2000, agricultural land lost
crops had been abandoned. After 2000, the about 1,675 ha to transitional woodland-shrub class
cultivation of land got momentum, intensification (Fig. 5).
process holding a greater share (about 8% of total Water bodies construction and management
changed area) than extensification (2.9%), as some encompassed over 2,363 thousands ha (only 1% of
arable areas, left fallow in the period of transition, total changed area), out of which 2,333 thou ha
started being re-cropped. affected in the years 1990-2000. What emerged was
Urbanisation and industrialisation are primarily conversion of agricultural land (706 ha)
characteristic of all the large cities of post- and wetlands (899 thou ha) to inland waters (class
communist Romania. Over the 1990-2006 period, 51) (Fig. 5). However, interpreting these transitional
artificial surfaces would enlarge by some 17,681 ha phenomena raises a series of questions, because they
(5% of the total change-affected area), especially to refer not only to proper hydrotechnical works, but
the detriment of all agricultural land-use categories also to temporarily covered with water by floods.
(Figs 4, 6). However, intensive urban sprawl (the
housing-market boom) in the outlying rural areas got 4.2. Spatial distribution of land-use and
momentum after 2006-2008, when vast expanses of land-cover change processes
agricultural land were turned mainly into built-up
areas: new dwelling complexes, industrial and Looking at the spatial distribution of the main
commercial units (logistic parks, super-markets, land-use and land-cover change processes one may
etc.), sporting and leisure facilities, etc., essentially notice significant disparities both with regard to the
altering their spatial and functional structure. The two intervals (1990-2000 and 2000-2006) and to the
financial crisis of the past five years has also scope-and-breadth of change as well as to the main
influenced the land-use pattern in the peri-urban land-use and land-cover flows within the major
areas as many developers slowed down their landform units. Most conversions from one land use
activity, and the majority of the terrains bought for and land cover category to another took place in the
housing projects were abandoned. Most of these Carpathian Mountains, the Romanian Plain, the
terrains have fertile soils. Looking forward to seeing Moldavian Plateau and the Transylvanian Plateau,
the boom of the housing-market, certain investors less affected being the Danube Delta, the Banat and
chose to license the land to some farmers and have it Crişana Plain and Hills and the Getic Piedmont
re-introduced into the agricultural cycle (vegetables (Table 2, Fig. 8).
or cereal crops), others preferred abandoning them The 1990-2000 interval coincides with the
altogether. The urbanisation process, also connected period of transition to the market economy in which
with the evolution of the main demographic agriculture was intensified and extensified within all
indicators (depleted birth and fertility rates, higher relief units, less so in the mountain regions where
external migration, negative natural balance plus afforestation and deforestation prevailed. Records
some social problems), reveals a steep demographic speaking of over 42% of total changed area, the
decline which will obviously affect the medium-and- logged area being of 60,000 ha, that is more than
long-term land-use pattern. True enough, Romania’s had been replanted. On the other hand, extensive
population is in general declining numerically. versus intensive agriculture was higher in all the
However, some territorial differentiations do exist. other units excepting the Moldavian Plateau, the
Thus, there are regions (e.g. the urban sprawls) in Dobrogea Plateau and the Romanian Plain.
which the population is steadily increasing. Others However, some territorial disparities within one and
experience a steep decline such as poorly-developed the same landform unit do exist, e.g. the eastern half

204
of the Romanian Plain, certain areas in the next in line coming the Moldavian Plateau (11.4%)
Moldavian Plateau or in the Transylvanian Plateau. and the Romanian Plain (8.8%) (Table 3). Although
Here, the rate of transition of arable land and of lower scope-and-breadth, deforestation was the
permanent crops to pastures was remarkably high, dominant process in the mountains, hills and
mostly because the poverty, migration abroad of the plateaux, followed by urbanisation and
rural population, severe fragmentation of arable land intensification of agriculture, particularly in the
and droughts frequently raged in the south and lowlands (Romanian Plain, Banat and Crişana Plain)
south-east of the country. and the Moldavian Plateau (Figs. 6, 8). As regards
The 2000-2006 period overlaps Romania’s the intensification and extensification of agriculture,
pre-accession to the European Union, so that the the former process is undoubtedly the dominant one
total area changed was by far smaller, again because the areas left fallow for years on end started
comprehensive change being registered in the being cultivated, a sign that the farming sector was
mountain regions (58.8% of total changed area), revigorating.

Figure 8. Main land use and land cover flows in Romania by landforms

Table 2. Areas affected by land use and land cover changes

1990-2000 2000-2006
Landform Total % of % of Total % of % of
Landform
area (ha) changes landform total changes landform total
(ha) area changes (ha) area changes
I Carpathian Mountains 6,628,64 119,747 1.81 42.45 44,855 0.68 58.75
II Subcarpathians 1,658,94 21,703 1.31 7.69 2,280 0.14 2.98
III Transylvanian Plateau 2,529,74 29,916 1.18 10.61 5,634 0.22 7.38
I
Moldavian Plateau 2,295,48 32,639 1.42 11.57 8,702 0.38 11.40
V
Getic Piedmont and
V 1,462,52 10,607 0.73 3.76 2,857 0.20 3.74
Mehedinţi Plateau
V
Dobrogea Plateau 1,017,16 13,167 1.29 4.67 961 0.09 1.26
I
V
Banat and Crişana Hils 1,283,38 6,644 0.52 2.36 2,402 0.19 3.15
II
V Romanian Plain and Danube
4,896,84 36,395 0.74 12.90 6,690 0.14 8.76
III Floodplain
I
Banat and Crişana Plain 1,619,46 10,016 0.62 3.55 1,550 0.10 2.03
X
X Danube Delta and Razim-
447,012 1,242 0.28 0.44 418 0.09 0.55
I Sinoie Lagoon Complex
Total 282,076 9.90 100 76347 2.22 100

205
 In the Danube Delta, over 0.28% of the area highest urbanisation rates registered the north-east,
had been affected by change in the first study period south and south-east of the country. The peak
and only 0.09% in the second one (Table 2), the deforestation values recorded the Carpathians
main land cover flows being deforestation (31%) Mountains, especially the northern half of the
and afforestation (38%). The process of urbanisation Eastern Carpathians, mainly by implementing the
affected 10% of the total change area in the first “Lupu” Act no.1/11 January 2000.
period and 16% in the second one. If certain limitations or uncertainties of
CORINE Land Cover database (minimum mapping
5. CONCLUSIONS area is 25 ha) and possible interpretation errors of
satellite images are overlooked, this could be a
The land-use and land-cover changes reliable source for monitoring and quantifying
experienced during the 1990-2006 period were spatial and temporal land-use and land-cover
particularly complex both across the country and in changes at national and regional levels, in particular.
the landform units themselves. Since 1991 there
were major changes in the type of property over
agricultural and forest lands and in the type of farms. AKNOWLEDGEMENTS
The severe fragmentation of farm land, the
emergence of numerous individual farms practicing This paper was elaborated in the framework of the
project ”Atlas of Environment” - fundamental study made
subsistence agriculture, poor services for agriculture
under the research plan of the Romanian Academy’s
(irrigation, fertilisation, mechanisation etc.) Institute of Geography. The authors have an equal
contributed to significantly decreasing of the quality contribution to this paper.
and quantity of land-use. We would like to thank the European Environment
GIS-based assessment of land-use changes Agency for CORINE Land Cover database.
over the 1990–2006 period, according to CORINE
Land Cover database, highlighted a wide range of REFERENCES
modifications in the land-use and land-cover pattern,
basically transition from one category of use to Bălteanu, D., Popescu, M. & Urşanu (Popovici), E.A.,
another. Firstly, it was the structure of agricultural 2004. Land Tenure and Land Relations in
land that has been mostly affected (from intensive to Romania, cap.in International Encyclopedia of
extensive pattern), in that permanent crops and Land Tenure and Land Relations in the World, vol.
IV, ed. Edwin Mellen Press, New York, 357-453.
heterogeneous agricultural areas shrank, while
Bălteanu, D., Popescu, M. & Urşanu (Popovici), E.A.,
pastures and arable land would expand. 2005. Land use in Romania under the Transition
In some regions, arable areas were sharply to the Market Economy. Analele Universităţii
reduced, being abandoned by their new owners Bucuresti, Anul LIV, 99-112.
unable to work their plots regained under Land Bălteanu, D. & Popovici, E.A., 2010. Land use changes
Reform (the Act 18/1991). The same happened to and land degradation in post-socialist Romania.
vineyards and orchards, many being abandoned or Romanian Journal of Geography, 54, 2, 95-107.
cleared off. Since 2000, the period of Romania’s Bičík, I., Jeleček, L. & Štìpánek, V., 2001. Land-Use
pre-accession to EU, the situation would change, Changes and Their Social Driving Forces in
arable land expanding, while vine-and-fruit areas Czechia in the 19th and 20th centuries. Land Use
Policy, 18, 65–73.
continued to subside.
Bičík, I. & Jeleček, L., 2009. Land use and landscape
It was a time when mountain forests and changes in Czechia during the period of transition
lowland wind breaks were systematically cut, 1990–2007. Geografie – Sbornik ČGS, 114, 4,
impairing the quality of land and intensifying 263–281.
extreme climatic phenomena. In the post-communist Bossard, M., Feranec, J. & Otahel, J., 2000. CORINE
period, but especially after 2006, urban sprawl got land cover technical guide – Addendum 2000,
momentum largely in the neighbourhood of the big European Environment Agency, Copenhagen,
urban centres. This process entailed the conversion 105p.
of vast farm land to built-up terrains. Dutca, I. & Abrudan, I.V., 2010. Estimation of Forest
The evolution of the main land-cover flows Land-Cover Change in Romania, between 1990
and 2006, Bulletin of the Transilvania University
was distinctively different in each of the two study-
of Brasov Vol. 3, (52), 33-36.
periods and in landform units. Thus, extensification Feranec, J., Šủri, M., Otahel, J., Cebecauer, T., Kola,
and intensification of agriculture prevailed in the J., Soukup, T., Zdenkova, D., Waszmuth, J.,
plains and plateaux, the first process being more Vajdea, V., Vijdea, A., Nitica, C., 2000.
evident over 1990-2000, and the last after 2000. The Inventory of major landscape changes in the Czech

206
 Republic, Hungary, Romania and Slovak Republic 2009. Land use change in Southern Romania after
1970s–1990s. International Journal of Applied the collapse of socialism, Regional Environmental
Earth Observation and Geoinformation, 2, 129– Change, 9, 1–12.
139. Lakes, T., Muller, D. & Kruger, C., 2009. Cropland
Feranec, J. & Otahel, J., 2001. Land cover of Slovakia. change in southern Romania: a comparison of
Veda, Bratislava, 124 p. logistic regressions and artificial neural networks,
Feranec, J., Stoimenov, A., Otahel, J., Vasteva, R., Landscape ecology, 24, 9, 1195-1206.
Ropecka, M., Betak, J. & Husar, K., 2006. Lambin, E., Baulies, X., Bockstael, N., Fischer, G.,
Changes of rural landscape in Slovakia and Krug, T., Leemans, R., Moran, E., Rindfuss, R.,
Bulgaria in 1990–2000 identified by application of Sato, Y., Skole, D., Turner II, B. & Vogel, C.,
CORINE land cover data (case studies – Trnava 1999. Land-Use and Land-Cover Change (LUCC)
and Plovdiv region). In: Braun, M. (ed): EARSeL Implementation Strategy of IGBP, No 48,
eProceedings of the Second Workshop of the Scientific Steering Committee and International
EARSeL SIG on Remote Sensing of Land Use and Project Office of LUCC, 11-12.
Land Cover Application and Development. Lambin, E.F., Geist, H.J. & Lepers, E., 2003. Dynamics
EARSeL and Universität, Bonn, 441–454. of land use and cover change in tropical regions.
Feranec, J., Hazeu, G., Christensen, S. & Jaffrain, G., Annual Review of Environment and Resources 28,
2007. CORINE land cover change detection in 205–241.
Europe (case studies of the Netherlands and Lambin, E.F. & Geist, H.J., 2006. Land-Use and Land-
Slovakia). Land Use Policy, 24, 234–247. Cover Change. Local Processes and Global
Feranec, J., Jaffrain, G., Soukup, T. & Hazeu, G., Impacts. Springer Berlin Heidelberg, New York,
2010. Determining changes and flows in European 222 p.
landscapes 1990–2000 using CORINE land cover Lu, D., Mausel, P., Brondizio, E. & Moran, E., 2004.
data, Applied Geography, 30, 19–35. Change detection techniques. International Journal
Feranec, J., Soukup, T., Hazeu, G. & Jaffrain, G., of Remote Sensing, 25, 12, 2365–2407.
(2012) Land cover and its change in Europe: Meyer, W.B. & Turner II, B.L., 1994. Changes in Land
1990-2006. In Giri, Ch. (Ed), Remote sensing of Use and Land Cover: A Global Perspective.
land use and land cover. Principles and Cambridge University Press, Cambridge, 537 p.
applications. CRC Press, Taylor & Francis Group, Ministry of Environment and Climate Change (2010),
Boca Raton, 285-301 National Afforestation Programme, (in
Fraser, E. & Stringer, C.L., 2009. Explaining Romanian).
agricultural collapse: Macro-forces, micro-crises Müller, D. & Kuemmerle, T., 2009. Causes of Cropland
and the emergence of land use vulnerability in Abandonment during the Post-socialist Transition
southern Romania, Global Environmental Change in Southern Romania. In: Groisman, P.Y. and S.V.
19, 45–53. Ivanov (Eds.) Regional aspects of climate-
Gregorio, A. & Jansen, L., 2000. Land Cover terrestrial-hydrologic interactions in non-boreal
Classification System. Classification concepts and Eastern Europe. New York, Springer, 205-214.
user manual. FAO, Rome, 179. National Institute of Statistics, Romanian Statistical
Grigorescu, I., Mitrică, B., Kucsicsa, Gh., Popovici, Yearbooks 1990-2012, Bucharest.
E.A., Dumitraşcu, M. & Cuculici, R., 2012. Niculescu, Gh., 2006. Romania. Geographical position,
Post-communist land use changes related to urban surface-area and borderlines. In Bălteanu D. et al.
sprawl in the Romanian Metropolitan Areas. (eds.): Romania. Space, Society, Environment.
Human Geographies - Journal of Studies and The Publishing House of the Romanian Academy,
Research in Human Geography, 6, 1, 2012, 35-46. Bucharest, 17-19.
Haines -Young, R. & Weber, J.L., 2006. Land accounts Otahel, J., Zigrai, F. & Drogona, V., 1993. Landscape
for Europe 1990–2000. Towards integrated land use as a basis for environmental planning (case
and ecosystem accounting, European Environment studies Bratislava and Nitra hinterlands).
Agency, Copenhagen, 107 p. Geographical Studies, 2, 7–83.
Knorn, J., Kuemmerle, T., Szabo, A., Mindrescu, M., Otahel, J., Feranec, J., Suri, M. & Cebecauer, T.,
Keeton, W.S., Radeloff, V.C., Abrudan. I., 2002. Land cover of Slovakia: assessment of its
Griffiths, P., Gancz, V. & Hostert, P., 2012. changes. In: Bicík I et al. (eds.): Land Use/Land
Forest restitution and the protected area Cover Changes in the Period of Globalization.
effectiveness in post-socialist Romania. Biological Proceedings of the IGU-LUCC International
Conservation, 146, 204-212. Conference, Prague, 2001. KSGRR PrF UK,
Kuemmerle, T., Radeloff, V.C., Perzanowski, K. & Prague, 100–105.
Hostert, P., 2006. Crossborder Comparison of Popescu, M., 2001. The lessons of transition. Agriculture
Land Cover and Landscape Pattern in Eastern 1990-2000. Edit. Expert, Bucureşti, 250 p., in
Europe Using a Hybrid Classification Technique, Romanian.
Remote Sensing of Environment, 103, 449–464. Popovici, E.A., 2008. The main drivind forces of land use
Kuemmerle, T., Muller, D., Griffiths, P. & Rusu, M., change in Romania over the post-socialist period.

207
 Revista Geografică, T. XIV-XV – 2007-2008, International Biosphere-Geosphere Program: A
Institutul de Geografie, Edit. Academiei Române, study of global change and the human dimensions
Bucureşti, 123-128, in Romanian. of global environmental change programme,
Popovici, E.A., 2010. Cotmeana Piedmont. Land use Stockholm, 65.
dynamics and environmental quality. Edit. Ursu, A., Stoleriu, C., Sfîc, L. & Boșca, B., 2006.
Academiei Române, Bucureşti, 213 p., in Adjusting of CORINE Land Cover nomenclature to
Romanian. the Romanian land use particularities. Geographia
Ptáček P., 2000. Networking and Local Culture: Local technica , No. 1, 193-198, in Romanian.
Community Transformation after 1989 on the Václavík, T. & Rogan, J., 2009. Identifying Trends in
Example of Olomouc, Czech Republic. Acta Land Use/Land Cover Changes in the Context of
Universitatis Palackianae–Geographica, 36, 59–64. Post-Socialist Transformation in Central Europe:
Stott, A. & Haines-Young, R., 1998. Linking land cover, A Case Study of the Greater Olomouc Region,
intensity of use and botanical diversity in an Czech Republic. GIScience & Remote Sensing, 46,
accounting framework in the UK. In K. Uno, & P. 1, 54–76.
Bartelmus (eds.), Environmental accounting in Zemek, F., Heøman, M., Mašková, Z. & Kvìt, J., 2005.
theory and practice, Dordrecht: Kluwer, 245–260. Multifunctional Land Use—a Chance of Resettling
Szilassi P., Jordan G., Kovacs F., Van Rompaey A. & Abandoned Landscapes? (A case study of the
Van Dessel W., 2010. Investigating the link Zhùøí territory, the Czech Republic). Ecology,
between soil quality and agricultural land use 24(1), 96–108.
change. A case study in the Lake Balaton Weber, J.L., 2007. Implementation of land and
catchment, Hungary, Carpathian Journal of Earth ecosystem accounts at the European environment
and Environmental Sciences, 5, 2, 61-70. agency. Ecological Economics, 61(4), 695–707.
Turner II, B.L. & Meyer, W.B., 1991. Land use and Willems, E., Lemments, T. & Buffaria, B., 2005.
land cover in global environmental change: Utilisation of CLC 90&2000 data for monitoring
Considerations for study. International Social the impact of CAP developments on the rural
Science Journal, 43, 4(130), 669-679. landscape. In: Trends of some agro-environmental
Turner II, B.L., Moss, R.H. & Skole, D.L., 1993. indicators in the European Union (Report EUR
Relating land use and global land-cover change: A 21565), 15–349.
proposal for an IGBP-HDP core project,

Received at: 21. 06. 2013

Revised at: 12. 10. 2013
Accepted for publications at: 17. 10. 2013
Published online at: 24. 10. 2013

208