The Impact of Land Use/Land Cover Changes on LandDegradation Dynamics: A Mediterranean Case Study

S. Bajocco • A. De Angelis • L. Perini •

A. Ferrara • L. Salvati

Received: 10 August 2010 / Accepted: 25 January 2012 / Published online: 15 March 2012

� Springer Science+Business Media, LLC 2012

Abstract In the last decades, due to climate changes, soil

deterioration, and Land Use/Land Cover Changes (LUL-

CCs), land degradation risk has become one of the most

important ecological issues at the global level. Land deg-

radation involves two interlocking systems: the natural

ecosystem and the socio-economic system. The complexity

of land degradation processes should be addressed using a

multidisciplinary approach. Therefore, the aim of this work

is to assess diachronically land degradation dynamics

under changing land covers. This paper analyzes LULCCs

and the parallel increase in the level of land sensitivity to

degradation along the coastal belt of Sardinia (Italy), a

typical Mediterranean region where human pressure affects

the landscape characteristics through fires, intensive agri-

cultural practices, land abandonment, urban sprawl, and

tourism concentration. Results reveal that two factors

mainly affect the level of land sensitivity to degradation in

the study area: (i) land abandonment and (ii) unsustainable

use of rural and peri-urban areas. Taken together, these

factors represent the primary cause of the LULCCs

observed in coastal Sardinia. By linking the structural

features of the Mediterranean landscape with its functional

land degradation dynamics over time, these results con-

tribute to orienting policies for sustainable land manage-

ment in Mediterranean coastal areas.

Keywords Land sensitivity � Land management � Coastal

area � Multi-temporal land cover maps � Sardinia � Italy

Introduction

Land degradation is becoming one of the major environ-

mental issues all over the world and affects also developed

regions like North America, Australia and Southern Europe

(Romm 2011). Land degradation is hence an interactive

process involving multiple causal factors, among which

climate variability, soil quality, and land management play

a significant role (Lambin and others 2001; Reynolds and

Stafford Smith 2002; Geist and Lambin 2004). In the

Mediterranean region, both biophysical variables and

socioeconomic conditions affect land quality, and their

interaction may become extremely complex through space

and time, resulting in typical land degradation patterns

(UNEP 1994; MEA 2005a, b).

In the present study, land degradation is considered as a

process occurring not only in semi-natural areas, but also in

agricultural and peri-urban lands (Bajocco and others

2011). For instance, besides soil erosion, the major land

degradation processes in the Mediterranean basin are soil

sealing, compaction mainly due to agricultural intensifi-

cation, salinization, and contamination due to industrial

activities (Montanarella 2007). Agropastoralism is con-

sidered one of the most important land degradation driving

forces as it acts both directly (e.g., overgrazing) and indi-

rectly generating land cover changes to create new pastures

(Harrington 1981; Valentine 1990; Margaris 1992) and

S. Bajocco (&) � A. De Angelis � L. Perini

Unit for Climatology and Meteorology in Agriculture

(CRA-CMA), Italian National Agricultural Research Council,

Via del Caravita 7a, 00186 Rome, RM, Italy

e-mail: sofia.bajocco@entecra.it

A. Ferrara

Department of Crop Science, University of Basilicata,

Via dell’Ateneo Lucano, Potenza, PZ, Italy

L. Salvati

Centre for Soil-Plant Relationships (CRA-RPS), Italian National

Agricultural Research Council, Via della Navicella 2-4,

00184 Rome, RM, Italy

123

Environmental Management (2012) 49:980–989

DOI 10.1007/s00267-012-9831-8

promoting mechanical tillage to improve forage production

(Perez-Trejo 1994). This is particularly true in specified

Mediterranean areas, such as southern Spain, Sardinia,

Crete, and the Aegean islands, which suffer from long-term

human pressure (e.g., Marathianou and others 2000).

Ancillary land degradation drivers include drought, the

natural or human-induced reduction in vegetation cover,

poor agricultural practices determining soil organic matter

losses, as well as irrational irrigation practices leading to

salinization, which are all factors contributing to increasing

the level of land degradation sensitivity of a region

(Symeonakis and Drake 2004).

In the past, the impact of human activities on the Medi-

terranean landscapes has increased considerably causing,

among others, biodiversity loss, deforestation, and soil ero-

sion (Giordano and Marini 2008). Land Use/Land Cover

Changes (LULCCs) represent one of the most important

consequences of the increasing human pressure (e.g., Con-

acher and Sala 1998), since reflect changes in both the rural

system (e.g., crop intensification vs extensification) and set-

tlement patterns (e.g., littoralisation vs inland depopulation

with the consequent abandonment of cultivated land).

It has been widely studied and demonstrated that during

the last decades the Mediterranean region underwent major

LULCCs as a result of the relocation of people to the

coastal border, forest fires, the abandonment of farms and

grazing land, the rapid expansion of tourism-related

activities, urbanization, deforestation, as well as the

intensification of agriculture (Balabanis and others 2000;

Burke and Thornes 2004; Bonet 2004; Sluiter and de Jong

2007). Many studies demonstrated that LULCCs affect the

level of land sensitivity to degradation often accelerating

land degradation processes (Burke and Thornes 2004;

Drake and Vafeidis 2003; Symeonakis and others 2007).

Examples of the existing nexus between LULCCs and land

degradation sensitivity include

(i) crop intensification determining soil pollution due to

an increase in fertilizer use;

(ii) land abandonment leading to unmanaged vegetation

prone to fire and favoring soil erosion;

(iii) urbanization favoring soil sealing and the consequent

alteration of the hydrogeological cycle; and

(iv) deforestation causing biodiversity loss and habitat

fragmentation.

The European Environmental Agency (2006) published

a well-known report focusing on LULCCs in the Medi-

terranean region. According to this study, population den-

sity on coastal areas is on average 10% higher than inland,

reaching a peak of 50% in some countries. Even more

worrying is the conversion rate of natural coastal areas into

artificial ones, being faster than population increase (Alves

and others 2007). In Italy, especially in southern Italy, this

process impacts on dry landscapes determining loss of

natural areas and biodiversity, soil deterioration, and the

consequent decrease in land productivity and quality

(Salvati and Zitti 2008).

The aim of this study is therefore to analyze main land

degradation processes through the evaluation of the rela-

tionship between LULCCs and land degradation over a

Mediterranean coastal area using multitemporal land cover

(LACOAST (LC) and CORINE Land Cover map (CLC)) and

diachronic land sensitivity maps (according to Environ-

mental Sensitive Areas (ESA) procedure (e.g., Brandt 2005))

within a Geographic Information System (GIS) framework.

In particular, this paper focuses on the coastal area of Sardinia

(Italy), a typical Mediterranean region characterized by

natural and human pressures, including planning deregula-

tion, coastal erosion, tourism concentration, and pressing

urbanization. The methodology illustrated in this paper,

based on modest data requirements and a readily imple-

mented procedure, is a useful tool supporting regional plan-

ning and land management in Mediterranean drylands.

Methodology of the Study

Outline of the Methodology

As clearly stated by Thornes (2004), Otto and others (2007),

and Symeonakis and others (2007), the study of LULCCs

trajectories may provide a meaningful contribution to the

land degradation assessment. This is particularly true in the

Mediterranean region where land degradation is not only

driven by climate change (like in several other regions in

the world, see Sivakumar 2007 for a review), but it depends

specifically on anthropogenic processes (see Wilson and

Juntti 2005). Although a number of EU-funded projects

concentrated on the interrelation between LULCCs and

land degradation (Burke and Thornes 2004), they rarely

referred to land degradation as a dynamic process (Hawkes

2004). Assessing LULCCs trends and determining their

possible impact on the changing level of land sensitivity to

degradation is a relatively new research topic with crucial

policy implications (Geist and Lambin 2004; Otto and

others 2007; Symeonakis and others 2007). As far as we

know, there is no methodology of how to use LULCCs

analysis to assess changes in land degradation sensitivity.

The present study contributes to this deserving need.

Study Area and Data Sets

The Study Area

Sardinia is the second largest island in the Mediterranean

Sea, with an area of roughly 24,000 km2

. It is situated

Environmental Management (2012) 49:980–989 981

123

between 38�510 and 41�150 latitude north and 8�80 and

9�500 east longitude and is one of the twenty Italian

administrative regions (Fig. 1). The island shows a com-

plex geography with 1,840 km of coasts and a prevalently

hilly topography (Bajocco and others 2010). Meadows and

pastures cover nearly 40% of the Island, the Mediterranean

maquis occupies over 20% of the surface area, whereas

woodlands cover almost 10% of the total area (Santini and

others 2010). Sardinia’s climate is predominantly Medi-

terranean. The mean annual temperature ranges from 11 to

17�C, while rainfall varies from 400 to 1,100 mm

depending on the elevation. Dry periods are often followed

by heavy rainfall episodes, triggering soil erosion and

flooding events (Santini and others 2010). The main

human-related activity causing land degradation in Sardi-

nia is mainly represented by grazing (D’Angelo and others

2000; Enne and others 2002), drought (Fiori and others

2004), mismanagement and salinisation of groundwater

resources (Salvati and Zitti 2008), intense wildfire activity

(Bajocco and Ricotta 2008), and decrease in vegetation

cover as a result of deforestation and land abandonment

(Giordano and Marini 2008).

Land Cover Maps

As determined by the combined effect of biophysical and

socioeconomic factors, LULCC is a fundamental indicator

for integrated coastal zone management and land quality

assessment over time (Freire and others 2009). Land cover

reflects the structural state of the landscape. This is the

reason why land cover data are increasingly used for

deriving landscape attributes, such as its composition,

diversity, and changes, as well as for modeling its different

properties (Feranec and others 2010). In this study, we used

comparable land cover maps from 1975 to 2000. The

1975–1990 change map was derived from the LACOAST

(LAnd cover changes in COASTal zones) project, while

the map dated 1990 and 2000 from the CORINE (Coor-

dinated Information on the European Environment) Land

Cover project.

LACOAST (LC) is a research project aimed at quanti-

fying LULCCs in a 10 km land strip from the coastline

during 1975–1990. LC is based on Landsat satellite images

from 1970s (Perdigao and Christensen 2000) and uses

CORINE Land Cover (CLC) 1990 as its reference dataset.

It uses CLC hierarchical classification (three-level hierar-

chical nomenclature with 44 categories at the third classi-

fication level) at 1:100,000 scale (see Bossard and others

2000). This project was carried out in the mid 1990s by the

Agricultural Information Systems Unit (AIS) of the Space

Applications Institute (SAI) based at the Joint Research

Centre (JRC), Ispra (Italy).

The CLC project was aimed at providing land cover

maps at various times for the whole of Europe and was

coordinated by the European Environment Agency (EEA).

It contributes to the knowledge of the land cover and its

changes in 26 European countries between 1990 and 2000

(Feranec and others 2010), providing two CLC databases

(CLC1990 and CLC2000) with comparable technical fea-

tures (Buttner and others 2002).

The CLC inventory is based on Landsat satellite images

as primary information source. The choice of scale

(1:100.000), minimum mapping unit (MMU) (25 ha), and

minimum width of the polygons (100 m) represents a

trade-off between production costs and land cover infor-

mation details (EEA 2007). The geometrical accuracy is

100 m, that means that there are no shifts higher that

100 m between the Landsat images and the CLC polygon

Fig. 1 Location of the study

area (left) and the coastal area of

Sardinia with the changed

polygons of both LACOAST

and CORINE Land Cover maps

(right)

982 Environmental Management (2012) 49:980–989

123

edges. These basic variables are the same for CLC1990 and

CLC2000. The standard CLC nomenclature includes 44

land cover classes and it is standardized for all of Europe

which makes comparison and aggregation at the European

level easier. The classes are grouped in a three-level hier-

archy (Table 1). The five main (level-one) categories are:

(i) artificial surfaces, (ii) agricultural areas, (iii) forests and

semi-natural areas, (iv) wetlands, and (v) water bodies.

The approach of computer assisted visual interpretation

of satellite images was chosen as the CLC mapping

methodology. Raw satellite images were pre-processed and

enhanced to yield a geometrically correct document in

national projection. Ortho-corrected Landsat satellite ima-

ges were provided with an RMSE error below 25 m.

Detailed topographic maps and in some cases orthophotos

were used to achieve this accuracy. Geospatial information

were validated in the field according to sampling proce-

dures. The main CLC technical characteristics are sum-

marized in Table 2. As for the CLC change product

(CLC1990–2000), the technical features are the same as for

the CLC basic products (i.e., CLC1990 and CLC2000),

except for the MMU that is 5 ha (Bossard and others 2000).

Environmentally Sensitive Areas Maps

We used the ESA (Environmentally Sensitive Area)

framework to quantify land degradation sensitivity over the

investigated area (Basso and others 2000). This framework

was applied at both regional and local scale in Portugal,

Spain, Italy, and Greece. The procedure is capable of

integrating variables from different data sources. It was

extensively validated in the field in several target sites by

analyzing the correlation between the ESAI and some

indicators of soil quality and physical degradation (Kosmas

and others 1999; Basso and others 2000; Lavado Contador

and others 2009). The methodology is based on more than

ten variables covering different themes, including the

geological, topographical, and climatic conditions, human

pressure, and land mismanagement. A set of sensitivity

scores was assigned to each variable. Scores were derived

from statistical analyses and additional information gath-

ered from the available literature (e.g., Kosmas and others

2000a, b; Lavado Contador and others 2009). For each

theme, a quality indicator was calculated by averaging the

sensitivity scores of the selected variables. A composite

index was then calculated by averaging the values of the

quality indicators.

Climate characteristics were described in the ESA

framework by mean annual rainfall rate, aridity index

(defined as the ratio between annual average rainfall rate

and reference evapotranspiration), and aspect (Sivakumar

2007). The average annual reference evapotranspiration

was calculated using Penman–Monteith formula (Incerti

and others 2007). Climate analysis was carried out for the

periods 1961–1990 and 1971–2000 (Salvati and Zitti

2008). Soil information was obtained from variables

Table 1 CORINE Land Cover map legend

Code Land cover type

111 Continuous urban fabric

112 Discontinuous urban fabric

121 Industrial or commercial units

122 Road and rail networks and associated land

123 Port areas

124 Airports

131 Mineral extraction sites

132 Dump sites

133 Construction sites

141 Green urban areas

142 Sport and leisure facilities

211 Non-irrigated arable land

212 Permanently irrigated land

213 Rice fields

221 Vineyards

222 Fruit trees and berry plantations

223 Olive groves

231 Pastures

241 Annual crops associated with permanent crops

242 Complex cultivation patterns

243 Land principally occupied by agriculture, with significant

areas of natural vegetation

244 Agro-forestry areas

311 Broad-leaved forest

312 Coniferous forest

313 Mixed forest

321 Natural grasslands

322 Moors and heathland

323 Sclerophyllous vegetation

324 Transitional woodland-shrub

331 Beaches, dunes, sands

332 Bare rocks

333 Sparsely vegetated areas

334 Burnt areas

335 Glaciers and perpetual snow

411 Inland marshes

412 Peat bogs

421 Salt marshes

422 Salines

423 Intertidal flats

511 Water courses

512 Water bodies

521 Coastal lagoons

522 Estuaries

523 Sea and ocean

Environmental Management (2012) 49:980–989 983

123

including soil texture, depth, slope, and drainage.

According to Basso and others (2000) some variables can

be considered as static in the ESA model as they change

slowly or rarely and by their nature are infrequently mea-

sured or mapped. This is the case for soil quality, which

was regarded as constant in the following analyses (Salvati

and Zitti 2008).

Vegetation quality in the ESA model was assessed by

considering four variables: fire risk, vegetation protection

against soil erosion, vegetation resistance to drought, and

vegetation cover (Basso and others 2000). Such indicators

were obtained from the elaboration of land cover maps dated

1990 and 2000. A weight was attributed to each third-level

CLC category based on its different level of sensitivity

related to vegetation and landscape characteristics (see

Salvati and Bajocco 2011). Finally, human-derived land

degradation was assessed as a result of processes such as the

relocation of people along the coastal areas, increasing

population density around the major cities, and the intensi-

fication of agriculture. Population density measured at the

municipality level in 1991 and 2001 by the National Census

of Households was used as a proxy for human pressure

(Salvati and Zitti 2008). Moreover, a demographic variation

index calculated for a time horizon of 10 years was defined at

the same geographical scale (Salvati and Zitti 2005). An

index of agricultural intensification was further obtained

from land cover maps in 1990 and 2000; a weight was

attributed to each cover class in order to obtain a land clas-

sification based on crop intensity (Salvati and others 2007).

Four partial indicators quantifying the environmental

quality in terms of climate (Climate Quality Index, CQI),

soil (Soil Quality Index, SQI), vegetation (Vegetation

Quality Index, VQI), and land management (Land Man-

agement Quality Index, MQI) were then estimated as the

geometric mean of the different scores attributed to each

selected variable. The scores of each thematic indicator

ranges from 1 (the lowest contribution to land sensitivity to

degradation) to 2 (the highest contribution to sensitivity to

degradation). The final index of land sensitivity (ESAI)

was subsequently estimated in each i-th spatial unit and j-th

year as the geometric mean of the four partial indicators

(Basso and others 2000) as follows:

ESAIij ¼ SQIij � CQIij � VQIij �MQIij

� �1=4

The ESAI score ranges from 1 (the lowest land sensitivity

to land degradation) to 2 (the highest sensitivity to land

degradation). Based on the ESAI values, it is possible to

identify four classes of land sensitivity which refer to the

most used classification thresholds (Basso and others 2000;

Brandt 2005): (i) non-affected areas (ESAI \ 1.17), (ii)

potentially affected areas (1.17 \ ESAI \ 1.225), (iii)

‘fragile’ areas (1.225 \ ESAI \ 1.375), and (iv) ‘critical’

areas (ESAI [ 1.375). ESAI maps of 1990 and 2000 were

produced after the various elementary layers were registered

and referenced to an elementary pixel of 1 km2 (scale

1:250,000) (Basso and others 2000; Salvati and others 2007;

Lavado Contador and others 2009).

GIS and Statistical Analysis

A preliminary analysis was undertaken considering the first

CLC level to identify which LULCCs occurred along the

Sardinian coastal belt during the two reference periods

(1975–1990 and 1990–2000). This allowed us to reduce the

possible errors, caused by the use of different data sources,

and their implications. Working at such ‘‘coarse’’ nomen-

clature level ensures that only real processes are identified,

and avoids misinterpretation of the same land cover class

(possible when working at the second or third CLC level).

In order to assess the land degradation sensitivity trend

associated with the LULCCs observed from 1975 to 2000,

we matched the ESAI maps of 1990 and 2000, respectively,

with the LULCC maps of LACOAST (LC75-90) and

CORINE Land Cover (CLC90-00). Since the spatial scales

Table 2 Summary of the main technical features of the CORINE Land Cover (CLC) products used in this paper

CLC1990 CLC2000

Satellite data Landsat-4/5 TM Landsat-7 ETM

Time consistency 1986–1998 2000 ± 1 Year

Geometric accuracy satellite images B50 m B25 m

CLC mapping MMU 25 ha 25 ha

Geometric accuracy CLC data 100 m Better than 100 m

Thematic accuracy C85% (Probably not achieved) C85% (Achieved, see Buttner-Maucha 2006)

Change mapping – Change area for existing polygons C5 ha; isolated changes C25 ha

Production time 10 Years 4 Years

Documentation Incomplete metadata Standard metadata

Access to the data Unclear dissemination policy Dissemination policy agreed from the start

Number of European countries involved 26 28

984 Environmental Management (2012) 49:980–989

123

of LULCCs and ESAI maps are different (1:100,000 and

1:250,000 respectively), we re-sampled the ESAI maps

using ArcGIS ‘‘Resample’’ tool (ESRI Inc., Redwoods,

USA) in order to have a reliable comparison between them.

We thus obtained a minimum pixel size (100 m 9 100 m)

comparable with the minimum LULCC polygons width

(100 m).

In order to analyze the relationship between LULCCs

and land degradation sensitivity dynamics, we focused on

the multi-temporal trend of the ESAI values (ESAI90 and

ESAI00) referred only to the LC75-90 polygons. This

analysis enabled us to monitor how the land quality status

varied (improving or worsening) over time in a given area

where LULCCs occurred in the past (Symeonakis and

others 2007). We hence carried out a change detection

analysis of the ESAI trend during 1990–2000 and matched

the results with the LC75-90 map in order to quantify the

ESAI increase (or decrease) associated to each changed

polygon during the investigated time period.

Finally, through the ‘zonal statistics’ ArcGIS tool, we

derived the ESAI90 and ESAI00 average value associated,

respectively, to each LC75-90 and CLC90-00 polygon in

order to identify which LULCCs have led over time to a

different level of land sensitivity to degradation. We

excluded the polygons associated to null ESAI pixels and

to the land cover categories indicating wetlands and water

bodies. We elaborated the LULCC classes at the first and

second CLC levels in order to interpret the results by

reducing the number of records, but still keeping the

information on the landscape variability (Feranec and

others 2010).

Results and Discussion

A preliminary overview of LULCCs in coastal Sardinia

shows that during 1975–1990 (Table 3) a larger surface area

underwent land cover change compared to 1990–2000

(61.9 km2 in 1975–1990 vs 32.3 km2 in 1990–2000 on a

total surface area covering 7,780 km2). The greatest modi-

fications occurred in coastal Sardinia during 1975–1990

(45% of the area undergoing changes in land cover) involved

LULCCs within the semi-natural class 3 areas (change class

3 ? 3). However, a relatively high percentage of this con-

version type is recorded also in 1990–2000 (21%), possibly

indicating the impact of repeated burning on land cover.

Important land cover conversions of agricultural into semi-

natural areas (change class 2 ? 3), mainly shrublands,

burnt areas, or sparsely vegetated areas were also observed.

These transitions increased from 1975–1990 (10%) to

1990–2000 (45%) possibly indicating a progressive aban-

donment of cultivated land. By contrast, the conversion of

natural and semi-natural areas into urban (change class

3 ? 1) and agricultural areas (change class 3 ? 2) has been

reduced over time. In particular, the latter conversion path

decreased from 1975–1990 (20%) to 1990–2000 (1%). The

rate of land conversion into artificial areas (change class

2 ? 1) was relatively stable all over the investigated period.

This type of conversion may be considered a honest indi-

cator of urban sprawl and littoralization with consequences

on the soil sealing status of coastal areas. LULCCs within

two different agricultural land cover classes (change class

2 ? 2) were mainly observed from non-irrigated fields

towards permanent crop, rice fields, and heterogeneous

cultivations in both time periods and may indicate crop

intensification (10% in 1975–1990 and 15% in 1990–2000).

Based on the previous results and according to the

approach proposed by Feranec and others (2010), the major

LULCCs occurred in coastal Sardinia from 1975 to 2000

were: (i) urbanization (land conversion to CLC class 1), (ii)

agricultural intensification (conversion to CLC class 2 and

modifications within classes 2 to classes 21 and 22), and

(iii) deforestation (conversion from 31 second-level CLC

class to 1 and 2 first-level CLC classes). Feranec and others

(2010) did not consider two additional LULCCs that were

observed in our study area: wildfires and land abandonment

(land conversion to CLC classes 33 or 24).

The assessment of LULCCs in coastal Sardinia was

supplemented by the analysis of land sensitivity to degra-

dation according to the average ESAI calculated at each

changing land cover class. The most critical ESAI values in

1990 (Table 4) were associated to the LULCCs observed in

1975–1990 and related to:

(i) urbanization, from forested to urban areas (31 ? 11)

and from cropland to industrial areas (21 ? 12) or

mines, pits, and dumps (21 ? 13);

(ii) crop intensification, from heterogeneous agricultural

lands to arable land (24 ? 21) or permanent

Table 3 Total and percent surface of each change class (first

CORINE level) by period

Change

classes

LC75-90 CLC90-00

Surface

(ha)

Surface

(%)

Surface

(ha)

Surface

(%)

1-1 248.7 0.4 42.5 0.1

1-2 392.2 0.6 0.0 0.0

1-3 647.3 1.1 188.7 0.6

2-1 5988.4 9.7 4158.0 12.9

2-2 5106.8 8.3 5395.0 16.7

2-3 6551.4 10.6 14698.4 45.5

3-1 2990.8 4.8 687.0 2.1

3-2 12265.2 19.8 314.1 1.0

3-3 27731.8 44.8 6855.0 21.2

Total 61922.5 100.0 32338.6 100.0

Environmental Management (2012) 49:980–989 985

123

cultivations (24 ? 22), and from bare or sparsely

vegetated areas to arable land (33 ? 21);

(iii) deforestation, from forested to urban areas (31 ?11) or permanent crops (31 ? 22);

(iv) land abandonment, from mine, dumps, construction

sites (13) and arable land (21) to heterogeneous

agricultural areas (24); and finally

(v) wildfires, from cropland to bare or sparsely vegetated

areas (21 ? 33) due to stubble burning and pasture

renewal.

Interestingly, also those polygons that did not change

land cover class during 1975–1990 (i.e., CLC classes 21

and 31) are characterized by a nearly critical status of land

degradation sensitivity in 1990; this aspect is particularly

serious for those classes that should highly represent nat-

ural areas (e.g., CLC class 31), which means that even if no

structural changes occurred, the landscape functionality is

threatened by high environmental fragility.

The results of the diachronic analysis of the ESAI

(1990–2000) associated to the LC75-90 polygons are

shown in Table 5. The change classes associated to a

decreasing land degradation sensitivity were related to:

(vi) afforestation (24 ? 31), i.e. natural or human-

induced forest regeneration of cultivated areas; and

(vii) vegetation recovery (24 ? 33 and 21 ? 33), pos-

sibly indicating a process of natural or human-

induced land requalification after land abandonment

or fire, and (33 ? 32) related to natural recovery

after burning phenomena.

On the contrary, an increasing land degradation sensi-

tivity was associated to the following land cover transi-

tions: (i) urbanization (24 ? 11 and 24 ? 12), (ii) crop

intensification (24 ? 22), and (vi) afforestation (33 ? 31)

related to post-fire forestation. The last result demonstrates

that, on the one hand, forested areas that have suffered

structural damages can hardly be reclaimed from a land

degradation perspective, and that, on the other hand, the

management of the new forest resources has not been

adequately carried out.

Finally, Table 6 compares the ESAI dynamics in dif-

ferent years and LULCCs. Polygons undergoing urbani-

zation (conversion to CLC class 1) in 1975–1990 showed a

stable level of land sensitivity in 2000, while those poly-

gons that underwent edification in 1990–2000 showed an

increasing ESAI level. On the contrary, polygons that were

converted into agricultural areas (CLC class 2) revealed

worsened sensitivity levels in the following years, thus

indicating crop intensification that could have had a strong

impact on land quality especially in ecologically-fragile

sites. Polygons undergoing modifications within the semi-

natural lands (CLC class 3) in 1975–1990 showed a weak

increase in the level of land sensitivity to degradation in

2000, while those polygons that underwent re-naturaliza-

tion processes in 1990–2000 showed lower ESAI values in

2000 compared to those observed in the previous time

period. This suggests that in the past the ecological

recovery of burnt and abandoned areas (as also demon-

strated by the results presented in Table 5) was relatively

difficult to manage compared with that observed in recent

years probably due to the more effective land practices.

Conclusions

Land degradation is not a static process and needs approa-

ches capable of addressing its spatial and temporal dynamics

(Salvati and Zitti 2009). Land degradation cannot be faced

as a single problem since it impacts on water and soil quality,

public health, and biodiversity. A better knowledge of the

processes driving LULCCs is a key issue to promote a sus-

tainable land management system. In this context, moni-

toring LULCCs at regional scale represents a major concern

Table 4 Average ESAI90 values for each combination of LULC change classes of LC75-90 at the second CORINE level

Land cover class 1990 (Final state)

11 12 13 21 22 24 31 32 33

1975 (Initial state) 13 – – – – – 1.457 1.332 1.330 –

21 1.466 1.565 1.532 1.449 – 1.494 1.364 1.376 1.506

22 1.400 – 1.386 – – – – – 1.325

24 1.425 1.434 1.471 1.529 1.508 1.395 1.415 1.399 1.329

31 1.545 – – 1.419 1.503 1.359 1.439 1.336 1.407

32 1.435 – 1.431 1.457 1.409 1.387 1.375 1.365 1.406

33 – – – 1.526 – 1.414 1.384 1.414 1.329

The mean was weighted on the surface area of each category. Absent values correspond to LULCCs that do not occur in the reference period or

that are not enough representative (\5 ESAI pixels)

Values indicating high sensitivity to land degradation were marked in bold

986 Environmental Management (2012) 49:980–989

123

for the identification of areas threatened by land degradation

where mitigation actions should be carried out (D’Angelo

and others 2000). LULCCs are traditionally interpreted by

distinguishing two transformation types: conversion and

modification. Land use/land cover conversion refers to the

complete replacement of one land cover type with another,

while land use/land cover modification refers to the more

subtle changes that affect the character of the land cover

without changing its attribute classification (Leemans and

Zuidema 1995).

Identifying the causes of land cover changes (LULCCs)

requires understanding how people make land-use deci-

sions (decision-making processes) and how specific envi-

ronmental and social factors interact to influence these

decisions (decision-making context) (Geist and Lambin

2004). Hence, assessing the decision-making context rep-

resents a major concern when analyzing the mutual rela-

tionship between land management and land quality status.

Through the analysis of such changes, negative effects

strongly linked to land degradation could be highlighted,

and the spatial pattern of the degradation processes could

be evaluated (Maitima and others 2009).

This contribution strongly endorses the importance of

high to medium-resolution time-series land cover data.

Detailed land cover information is required in many

aspects dealing with sustainable land management, as a

prerequisite for monitoring environmental change and

modeling land use, and as a basis for land statistics at all

levels (Jansen and Di Gregorio 2004). A permanent

assessment of LULCCs and human-related causes and

responses is essential in land degradation studies. The

combined use of land cover and land degradation data, on

the one hand, allows to detect where certain changes occur,

what type of change, as well as how the land quality status

is changing. On the other hand, these data support decision-

makers to develop short- and medium-term plans for the

conservation and sustainable use of natural resources

(Jansen and Di Gregorio 2004). The evidence emerged in

this paper, linking the structural feature of theTa

ble

5D

iffe

ren

cean

dch

ang

era

te(%

)b

etw

een

mea

nE

SA

I90

and

mea

nE

SA

I00

for

each

com

bin

atio

no

fL

UL

CC

clas

ses

of

LC

75

-90

atth

ese

con

dC

OR

INE

lev

el

Lan

dco

ver

clas

s1

99

0(F

inal

stat

e)

11

(%)

12

(%)

13

(%)

21

(%)

22

(%)

24

(%)

31

(%)

32

(%)

33

(%)

19

75

(In

itia

lst

ate)

13

––

––

–0

.00

9(0

.6)

0.0

12

(0.9

)0

.01

6(1

.2)

–

21

0.0

12

(0.8

)0

.00

1(0

.1)

0.0

18

(1.2

)0

.00

6(0

.4)

–0

.00

7(0

.5)

0.0

00

(0.0

)0

.01

6(1

.2)

-0

.07

9(-

5.2

)

22

0.0

16

(1.1

)–

0.0

16

(1.2

)–

––

––

0.0

08

(0.6

)

24

0.0

22

(1.5

)0

.08

7(6

.1)

1.2

53

(0.9

)0

.00

9(0

.6)

0.0

46

(3.1

)0

.00

9(0

.6)

-0

.02

2(-

1.6

)0

.00

5(0

.4)

-0

.03

3(-

2.5

)

31

0.0

04

(0.3

)–

–0

.00

9(0

.6)

0.0

18

(1.2

)0

.01

0(0

.7)

0.0

17

(1.2

)0

.00

9(0

.7)

0.0

16

(1.1

)

32

0.0

03

(0.3

)–

1.3

98

(1.0

)0

.00

4(0

.3)

0.0

17

(1.2

)-

0.0

02

(-0

.1)

0.0

11

(0.8

)0

.00

7(0

.5)

-0

.00

3(-

0.2

)

33

––

–0

.01

8(1

.2)

–0

.01

(1.0

)0

.01

9(1

.5)

-0

.02

3(-

1.6

)0

.01

6(1

.2)

No

rmal

char

acte

rre

fers

toa

wo

rsen

ing

inla

nd

deg

rad

atio

nst

atu

s.B

old

char

acte

rre

fers

toan

ES

AI

incr

ease

rate

hig

her

than

1.5

%(a

rbit

rary

thre

shold

).It

alic

sre

fer

toan

ES

AI

dec

reas

era

te

hig

her

than

1.5

%(a

rbit

rary

thre

sho

ld).

Ab

sen

tv

alu

esco

rres

po

nd

toL

UL

CC

sth

atd

on

ot

occ

ur

inth

ere

fere

nce

per

iod

or

that

are

no

ten

ou

gh

rep

rese

nta

tiv

e(\

4E

SA

Ip

ixel

so

f1

km

2)

Table 6 Weighted mean of the ESAI90 and ESAI00 values related to

each change class (final state) of LC75-90 (both ESAI90 and ESAI00)

and CLC90-00 (only ESAI00), at the first CORINE level

Periods Land cover classes (final state)

1 2 3

LC75-90 versus ESAI90 1.483 1.459 1.385

LC75-90 versus ESAI00 1.477 1.474 1.386

% Change between -0.4 1.0 0.07

CLC90-00 versus ESAI00 1.482 1.480 1.367

The mean was weighted on the relative surface of each category.

Normal character indicates an increase of the weighted mean. Italics

indicate a reduction of the ESAI weighted mean

Environmental Management (2012) 49:980–989 987

123

Mediterranean landscape with its functional dynamics over

time, provide a simple framework to foresee the land

sensitivity response to changing LULCCs scenarios, and

can effectively contribute to land management policies

targeted to preserve the environmental quality of Medi-

terranean coastal areas. Finally, land cover mapping and

documentation may not provide the ultimate explanation

for all problems related to land degradation and cannot be

an end in itself. Nevertheless, it serves as a stepping stone

for understanding trends and possible causes of LULCCs

and their implications.

Acknowledgments Thanks are due to T. Ceccarelli (CRA-CMA)

who provided technical support and critical reading of the manuscript.

Authors of this paper were partly financed by ‘Agroscenari’ project

(research unit 6a) funded by the Italian Ministry of Agricultural and

Forestry Policies.

References

Alves FL, Silva CP, Pinto P (2007) The Assessment of Coastal Zone

Development at a Regional Level—the case study of the

Portuguese Central Area. Journal of Coastal Research SI50:72–76

Bajocco S, Ricotta C (2008) Evidence of selective burning in Sardinia

(Italy): Which land cover classes do wildfires prefer? Landscape

Ecology 23:241–248

Bajocco S, Rosati L, Ricotta C (2010) Knowing fire incidence

through fuel phenology: a remotely sensed approach. Ecological

Modelling 221:59–66

Bajocco S, Salvati L, Ricotta C (2011) Land degradation vs. fire: a

spiral process? Progress in Physical Geography 35:3–18

Balabanis P, Peter D, Ghazi A, Tsogas M (2000) Mediterranean

desertification research results and policy implications. Proceed-

ings of the international conference, Nov 1996. European

Commission, Luxemburg

Basso F, Bove E, Dumontet S, Ferrara A, Pisante M, Quaranta G,

Taberner M (2000) Evaluating environmental sensitivity at the

basin scale through the use of geographic information systems

and remotely sensed data: an example covering the Agri basin—

Southern Italy. Catena 40:19–35

Bonet A (2004) Secondary succession of semi-arid Mediterranean

old-fields in south-eastern Spain: insights for conservation and

restoration of degraded lands. Journal of Arid Environments

56(2):213–233

Bossard M, Feranec J, Otahel J (2000) CORINE land cover technical

guide—addendum 2000. Technical Report No. 40. European

Environment Agency, Copenhagen, p 105

Brandt J (2005) Land degradation information system to support

National Action Programmes in the Mediterranean (DISMED).

DIS4ME, Land degradation Indicator System for Mediterranean

Europe. www.unibas.it/desertnet/dis4me/using_dis4me/dismed.

htm. Accessed Aug 2010

Burke SM, Thornes JB (2004) A thematic review of EU Mediterra-

nean desertification research in Frameworks III and IV: preface.

Advances in Environmental Monitoring and Modelling 1:1–14

Buttner G, Feranec J, Jaffrain G (2002) CORINE land cover update.

I&CLC2000 project. European Environment Agency. Technical

Guidelines, p 56

Buttner G, Maucha G (2006) The thematic accuracy of Corine land

cover 2000. Assessment using LUCAS (land use/cover area frame

statistical survey). Technical report 7. European Environment

Agency, Copenhagen. http://reports.eea.europa.eu/technical_

report_2006_7/en. Accessed Mar 2012

Conacher AJ, Sala M (1998) Land degradation in Mediterranean

environments of the world. Wiley, Chichester

D’Angelo M, Enne G, Madrau S, Percich S, Previtali F, Pulina G, Zucca

C (2000) Mitigating land degradation in Mediterranean agro-

silvo-pastoral systems: a GIS-based approach. Catena 40:37–49

Drake NA, Vafeidis A (2003) Review of spatial and temporal

methods for assessing land degradation in the Mediterranean.

Advances in Environmental Monitoring and Modelling 1:1–51

EEA (2007) CLC2006 technical guidelines. Technical report 17/2007.

European Environmental Agency, p 66

Enne G, Pulina G, D’Angelo M, Previtali F, Madrau S, Caredda S,

Francesconi AH (2002) Agropastoral activities and land degra-

dation: the case study of Sardinia. In: Thornes J, Brandt J,

Geeson N (eds) Mediterranean desertification—a mosaic of

processes and responses. Wiley, Chichester, pp 71–82

European Environmental Agency (2006) The changing faces of

Europe’s coastal areas. European Environment Agency, Report

No. 6/2006, Copenhagen, p 107

Feranec J, Jaffrain G, Soukup T, Hazeu G (2010) Determining

changes and flows in European landscapes 1990–2000 using

CORINE land cover data. Applied Geography 30:19–35

Fiori M, Motroni A, Duce P, Spano D (2004) A daily water balance

estimate for climate risk evaluation at a local scale. Acta

Horticulturae, vol 664. ISHS. IV international symposium on

irrigation of horticultural crops

Freire S, Santos T, Tenedorio JA (2009) Recent urbanization and land

use/land cover change in Portugal the influence of coastline and

coastal urban centers. Journal of Coastal Research Special Issue

56:1499–1503

Geist HJ, Lambin EF (2004) Dynamic causal patterns of desertifica-

tion. Bioscience 54:817–829

Giordano F, Marini A (2008) A landscape approach for detecting and

assessing changes in an area prone to desertification in Sardinia

(Italy). International Journal of Navigation and Observation. doi:

10.1155/2008/549630

Harrington GN (1981) Grazing arid and semiarid pastures. In: Morley

FHW (ed) Grazing animals. Elsevier, Amsterdam, pp 181–201

Hawkes JC (2004) A review of European Union funded research into

Mediterranean desertification processes. Advances in Environ-

mental Monitoring and Modelling 1:1–39

Incerti G, Feoli E, Giovacchini A, Salvati L, Brunetti A (2007)

Analysis of bioclimatic time series and their neural network-

based classification to characterize drought risk patterns in south

Italy. International Journal of Biometeorology 51:253–263

Jansen LJM, Di Gregorio A (2004) Land Cover Classification System:

Basic concepts, main software functions and overview of the

‘‘land system’’ approach. In: Groom G (ed) Developments in

strategic landscape monitoring for the Nordic countries. Nordic

Council of Ministers of Environment. ANP 2004/705:64–73

Kosmas C, Kirkby M, Geeson N (1999) Manual on key indicators of

desertification and mapping environmental sensitive areas to

desertification. Directorate General, European Commission,

Bruxelles. http://www.kcl.ac.uk/projects/desertlinks/downloads/

publicdownloads/ESA%20Manual.pdf. Accessed Aug 2010

Kosmas C, Danalatos NG, Gerontidis S (2000a) The effect of land

parameters on vegetation performance and degree of erosion

under Mediterranean conditions. Catena 40:3–17

Kosmas C, Gerontidis S, Marathianou M (2000b) The effect of land

use change on soil and vegetation over various lithological

formations on Lesvos. Catena 40:51–68

Lambin EF, Turner BL, Geist HJ, Agbola SB, Angelsen A, Bruce JW,

Coomes OT, Dirzo R, Fischer G, Folke C, George PS,

Homewood K, Imbernon J, Leemans R, Li XB, Moran EF,

Mortimore M, Ramakrishnan PS, Richards JF, Skanes H, Steffen

988 Environmental Management (2012) 49:980–989

123

W, Stone GD, Svedin U, Veldkamp TA, Vogel C, Xu JC (2001)

The causes of land-use and land-cover change: moving beyond

the myths. Global Environmental Change: Human and Policy

Dimensions 11:261–269

Lavado Contador JF, Schnabel S, Gomez Gutierrez A, Pulido

Fernandez M (2009) Mapping sensitivity to land degradation

in Extremadura, SW Spain. Land Degradation and Development

20:129–144

Leemans R, Zuidema G (1995) Evaluating changes in land-cover

and their importance for global change. Trends in Ecology and

Evolution 10:76-81

Maitima JM, Mugathal SM, Reid RS, Gachimbi LN, Majule A,

Lyaruu H, Pomery D, Mathai S, Mugisha S (2009) The linkages

between land use change, land degradation and biodiversity

across East Africa. African Journal of Environmental Science

and Technology 3(10):310–325

Marathianou M, Kosmas C, Gerontidis S, Detsis V (2000) Land-use

evolution and degradation in Lesvos (Greece): a historical

approach. Land Degradation and Development 11(1):63–73

Margaris NS (1992) Primary sector and environment in the Aegean.

Environmental Management 16(5):569–574

Millennium Ecosystem Assessment (MEA) (2005a) Ecosystems and

human well-being: synthesis. World Resources Institute. Island

Press, Washington, DC

Millennium Ecosystem Assessment (MEA) (2005b) Ecosystems and

human well-being: desertification synthesis. World Resources

Institute, Washington, DC

Montanarella L (2007) Trends in land degradation in Europe. In:

Sivakumar MV, N’Diangui N (eds) Climate and land degrada-

tion. Springer, Berlin

Otto R, Krusi BO, Kienast F (2007) Degradation of an arid coastal

landscape in relation to land use changes in southern Tenerife

(Canary Islands). Journal of Arid Environments 70:527–539

Perdigao W, Christensen S (2000) The LaCOAST atlas: land cover

changes in European coastal zones. Joint Research Centre, Ispra

Perez-Trejo F (1994) Desertification and land degradation in the

European Mediterranean, EUR 14850 EN. Report prepared for

the Commission of the European Communities, Brussels

Reynolds JF, Stafford Smith M (eds) (2002) Global desertification:

Do humans cause deserts?. Dahlem University Press, Berlin

Romm J (2011) Desertification: the next dust bowl. Nature

478:450–451

Salvati L, Bajocco S (2011) Land sensitivity to desertification across

Italy: past, present, and future. Applied Geography 31(1):

223–231

Salvati L, Zitti M (2005) Land degradation in the Mediterranean

basin: linking bio-physical and economic factors into an

ecological perspective. Biota 5:67–77

Salvati L, Zitti M (2008) Regional convergence of environmental

variables: empirical evidences from land degradation. Ecological

Economics 68:162–168

Salvati L, Zitti M (2009) Assessing the impact of ecological and

economic factors on land degradation vulnerability through

multiway analysis. Ecological Indicators 9:357–363

Salvati L, Macculi F, Toscano S, Zitti M (2007) Comparing indicators

of intensive agriculture from different statistical sources. Biota

8:51–60

Santini M, Caccamo G, Laurenti A, Noce S, Valentini R (2010) A

multicomponent GIS framework for desertification risk assess-

ment by an integrated index. Applied Geography 30(3):394–415

Sivakumar MVK (2007) Interactions between climate and land

degradation. Agriculture and Forest Meteorology 142:143–155

Sluiter R, de Jong SM (2007) Spatial patterns of Mediterranean land

abandonment and related land cover transitions. Landscape

Ecology 22:559–576

Symeonakis E, Drake N (2004) Monitoring desertification and land

degradation over sub-Saharan Africa. International Journal of

Remote Sensing 25:573–592

Symeonakis E, Calvo-Cases A, Arnau-Rosalen E (2007) Land use

change and land degradation in southeastern Mediterranean

Spain. Environmental Management 40:80–94

Thornes JB (2004) Stability and instability in the management of

Mediterranean desertification. In: Wainwright J, Mulligan M

(eds) Environmental modelling: findings implicity in complex-

ity. Wiley, Chichester, pp 303–315

UNEP (1994) United Nations Convention to combat desertification in

those countries experiencing serious drought and/or desertifica-

tion, particularly Africa. UNEP, Geneva

Valentine JF (1990) Grazing management. Academic Press, San

Diego

Wilson G, Juntti M (2005) Unravelling desertification: policies and

actor networks in southern Europe. Wageningen Academic

Publishers, Wageningen

Environmental Management (2012) 49:980–989 989

123