ArticlePDF Available

Abstract and Figures

Land cover and its configuration in the landscape are crucial components in the provision of biodiversity and ecosystem services. In Mediterranean regions, natural landscapes mostly covered by evergreen vegetation have been to a large extent transformed into cultural landscapes since long time ago. We investigated land cover changes in Central Chile using multi-temporal satellite imagery taken in 1975, 1985, 1999 and 2008. The major trends in this highly dynamic landscape were reduction of dryland forest and conversion of shrubland to intensive land uses such as farmland. The average net annual deforestation rate was −1.7%, and shrubland reduction occurred at an annual rate of −0.7%; agriculture, urban areas and timber plantations increased at annual rates of 1.1%, 2.7% and 3.2%, respectively, during the 1975–2008 period. Total forest and shrubland loss rates were partly offset by passive revegetation. However, most of the areas that were passively revegetated remained as shrubland and did not turn into forests due to a low capacity of forest recovery. This resulted in a progressive loss and degradation of dryland forest over the entire region. Overall, the documented land cover changes increase provisioning services such as crops, cattle, and timber that are characteristic of cultural landscapes in the area but may cause an irreversible loss of biodiversity and a depletion of other ecological services provided by forests and shrubland. The implications for conservation of this area and the need for territorial planning and adapted land-use strategies are discussed.
Content may be subject to copyright.
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Monitoring land cover change of the dryland forest landscape of Central
Chile (1975–2008)
Jennifer J. Schulz
a
,
*
, Luis Cayuela
b
, Cristian Echeverria
c
, Javier Salas
d
,
Jose
´Marı
´a Rey Benayas
a
a
Departamento de Ecologı
´a, Edificio de Ciencias, Universidad de Alcala
´, Carretera de Barcelona km 33,600, E-28871 Alcala
´de Henares, Madrid, Spain
b
Departamento de Ecologı
´a, Centro Andaluz de Medio Ambiente, Universidad de Granada – Junta de Andalucı
´a, Avda. del Mediterra
´neo s/n,
E-18006 Granada, Spain
c
Departamento Manejo de Bosques y Medio Ambiente, Facultad de Ciencias Forestales, Universidad de Concepcio
´n, Casilla 160-C, Concepcio
´n, Chile
d
Departamento de Geografı
´a, Universidad de Alcala
´, E-28801 Alcala
´de Henares, Madrid, Spain
Keywords:
Deforestation
Mediterranean
Sclerophyllous forest
Remote sensing
Vegetation recovery
abstract
Land cover and its configuration in the landscape are crucial components in the provision
of biodiversity and ecosystem services. In Mediterranean regions, natural landscapes
mostly covered by evergreen vegetation have been to a large extent transformed into
cultural landscapes since long time ago. We investigated land cover changes in Central
Chile using multi-temporal satellite imagery taken in 1975, 1985, 1999 and 2008. The
major trends in this highly dynamic landscape were reduction of dryland forest and
conversion of shrubland to intensive land uses such as farmland. The average net annual
deforestation rate was 1.7%, and shrubland reduction occurred at an annual rate of
0.7%; agriculture, urban areas and timber plantations increased at annual rates of 1.1%,
2.7% and 3.2%, respectively, during the 1975–2008 period. Total forest and shrubland loss
rates were partly offset by passive revegetation. However, most of the areas that were
passively revegetated remained as shrubland and did not turn into forests due to a low
capacity of forest recovery. This resulted in a progressive loss and degradation of dryland
forest over the entire region. Overall, the documented land cover changes increase
provisioning services such as crops, cattle, and timber that are characteristic of cultural
landscapes in the area but may cause an irreversible loss of biodiversity and a depletion of
other ecological services provided by forests and shrubland. The implications for conser-
vation of this area and the need for territorial planning and adapted land-use strategies are
discussed.
Ó2009 Elsevier Ltd. All rights reserved.
Introduction
Natural landscapes, i.e. those unaffected or hardly affected by human activities, are being transformed into cultural
landscapes throughout the world (Feranec, Jaffrain, Soukup, & Hazeu, 2010; Foley et al., 2005; Lo
´pez & Sierra, 2010). This
transformation trades off the biodiversity and ecosystem services which are characteristic of both types of landscapes, e.g.
higher levels of biodiversity and supporting and regulating services in natural landscapes vs. higher levels of provisioning
services such as crop and timber production in cultural landscapes (Millennium Ecosystem Assessment, 2005; Rey Benayas,
*Corresponding author.
E-mail addresses: jennifer.schulz@uah.es (J.J. Schulz), lcayuela@ugr.es (L. Cayuela), cristian.echeverria@udec.cl (C. Echeverria), javier.salas@uah.es
(J. Salas), josem.rey@uah.es (J.M. Rey Benayas).
Contents lists available at ScienceDirect
Applied Geography
journal homepage: www.elsevier.com/locate/apgeog
0143-6228/$ – see front matter Ó2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apgeog.2009.12.003
Applied Geography 30 (2010) 436–447
Author's personal copy
Newton, Dı
´az, & Bullock, 2009). As the characteristics of land cover have important impacts on climate, biogeochemistry,
hydrology and species diversity, land cover change has been indicated as one of the high priority concerns for research and for
the development of strategies for sustainable management (Turner, Moss, & Skole, 1993; Vitousek, 1994). In recent years,
special attention has been given to land-use changes and dryland degradation (Reynolds, Maestre, Kemp, Stafford-Smith, &
Lambin, 2007). Vegetation cover in these ecosystems with limited primary productivity plays a crucial role in providing
services such as climate and water regulation (Millennium Ecosystem Assessment, 2005).
Mediterranean ecosystems are a particular type of dryland that account for less than 5% of the Earth’s surface but host
about 20% of the world’s plant species, many of which are endemic (Cowling, Rundel, Lamont, Kalin Arroyo, & Arianoutsou,
1996). Land degradation, i.e. the substantial decrease in the biological productivity of the land system, resulting from human
activities rather than natural events (Johnson & Lewis, 2006) is an important issue in many Mediterranean regions (Conacher
& Sala, 1998; Geri, Amici, & Rocchini, 2010). Loss of natural vegetation cover is often a precedent to soil erosion and dete-
rioration of the water storage capacity; these modifications of the land system may lead to desertification due to longer term
factors such as climate change, triggering short term degradation of ecosystems by humans (Reynolds & Stafford Smith,
2002). Nevertheless, in the Mediterranean basin and California, the loss of vegetation cover has been partly counterbalanced
by vegetation recovery over the last decade (Carmel & Flather, 2004; Lasanta, Gonza
´lez-Hidalgo, Vicente-Serrano, & Sferi,
2006; Mouillot, Ratte, Joffre, Mouillot, & Rambal, 2005; Pueyo & Beguerı
´a, 2007; Romero-Calcerrada & Perry, 2004), which
occurred mostly due to concentration of crop production and abandonment of less productive farmland. The polarisation
between more intensive and more extensive use of land has been described as the main trend of current landscape changes
(Antrop, 2005).
In Mediterranean Central Chile, however, this trend is less clear. Whereas land abandonment is occurring in some areas as
a result of soil degradation, threats to sclerophyllous forests and shrublands, such as urban and agricultural expansion, cattle
grazing, logging for firewood, and introduction of alien species, still persist throughout the region. Some studies have
described vegetation degradation in Central Chile concerning disturbances of successional trajectories at a rather local scale
(e.g. Balduzzi, Tomaselli, Serey, & Villasen
˜or, 1982; Fuentes, Avile
´s, & Segura, 1989; Holmgren, 2002) and pressures on
vegetation due to land occupation patterns (Ovalle, Avendan
˜o, Aronson, & Del Pozo, 1996). Common patterns of landscape
change throughout Central Chile, including the description of severe reduction of natural vegetation, have been described
rather qualitatively (e.g. Armesto, Arroyo, Mary, & Hinojosa, 2007; Aronson et al., 1998). So far, these processes have not been
mapped and quantified at a regional scale, and change trajectories among land cover types have not been systematically
evaluated and explained.
To address the issue of gaining a systematic understanding of the magnitude of land cover changes at the regional scale we
considered the advantages of remote sensing data to detect, measure and monitor land cover change due to this system’s
ability to capture an instantaneous synoptic view of a large part of the Earth’s surface and acquire repeated measurements of
the same area on a regular basis (Donoghue, 2002). Land cover detection and monitoring are especially useful in those regions
where there is a lack of available cartographic information with sufficient spatial resolution to examine how humans change
land cover and to provide a basis for conservation and restoration planning. To our knowledge, this is the first study that has
explored the recent historical and current extent of land cover types, as well as the changes that have occurred in Central Chile
over a 33-year period (1975–2008). The main goal of this study is to investigate the dynamics of land cover change, focussing
on the dynamics of natural vegetation cover as a result of land-use pressure, particularly a hypothesised expansion of
cropland, pastures and timber plantations. The specific scope of this paper is embedded in the dimension of land change
science that focuses on observation, monitoring, and land characterization (Turner, Lambin, & Reenberg, 2007). We specif-
ically address: (1) area change and change rate; (2) spatial distribution of changes; (3) change trajectories of land cover types;
and (4) accuracy assessment of change detection. The information generated in this study will be a useful basis for analyzing
underlying causes of change and designing management strategies, as it identifies the spatio-temporal patterns associated
with landscape processes that might affect policy making, conservation and restoration programs.
Material and methods
Study area
The study area is located in the Mediterranean bioclimatic zone of Central Chile (Amigo & Ramı
´rez, 1998) and extends over
13,175 km
2
, covering parts of the Valparaı
´so, Libertador Bernando O’Higgins and Metropolitan administrative regions (Fig. 1).
The area includes characteristic landscapes of the Mediterranean zone, like parts of the coastal plains, the coastal mountain
range and the Central Valley. The rationale for defining the boundaries of the study area based on the bioclimatic zone was
that vegetation has a similar response to biotic and abiotic factors. For example, vegetation recovery is constrained within this
bioclimatic area by water availability. Additional criteria used for defining the study area were that it shares a relatively
common pattern of land use and that it concentrates a large population that may put high pressure on natural resources.
Altogether the area is characterised by dry summers and wet winters with strong inter-annual variability due to the El
Nin
˜o–Southern Oscillation (ENSO) phenomenon. The mean annual temperature is 13.2
C, and the mean annual precipitation
is 531 mm. Temperature and moisture patterns are primarily a function of topography (Badano, Cavieres, Molina-
Montenegro, & Quiroz, 2005), with elevations ranging from sea level to 2260 m in the coastal mountain range. The climatic
variability and varied topography result in a spatially heterogeneous mosaic of vegetation. At present, Acacia caven shrubland
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447 437
Author's personal copy
is predominant and covers most of the lower hillslopes, whereas evergreen sclerophyllous forest remains in drainage
corridors and on steeper slopes with southern aspects.
The pre-Columbian vegetation of Central Chile is described to have been dense and diverse, with a predominance of
evergreen sclerophyllous trees and shrubs (Balduzzi et al., 1982). Historical records indicate that Central Chile has experi-
enced profound landscape transformations since the mid-sixteenth century resulting from logging, agriculture expansion and
livestock overgrazing (Vogiatzakis, Mannion, & Griffiths, 2006). Land use is mostly concentrated in flat valleys, where the
major agricultural activities are vineyard and fruit cultivation as well as corn and wheat cropping. Two of the most important
uses of native forest resources by local communities are extraction of fuelwood and extensive livestock husbandry.
Conversions to commercial timber plantations with exotic species like Pinus radiata and Eucalyptus globulus have also
occurred in the flat coastal zone since the 1970s (Aronson et al., 1998).
The study area is home to circa 5.2 million inhabitants, which represents around 34% of the Chilean population. The
population increased by 53% between 1970 and 2002, and the percentage of urban population as compared to rural pop-
ulation remained high (93% urban population in 1970 and 96% in 2002). The region is acknowledged as one of the world’s 25
biodiversity hotspots (Myers, Mittermeier, Mittermeier, da Fonseca, & Kent, 2000) and is home to approximately 2400 plant
species, 23% of which are endemic (Cowling et al., 1996).
Multi-temporal land cover classification
Land cover change was evaluated with a post-classification procedure in order to obtain a matrix of change directions
among land cover classes (Lu, Mausel, Brondizio, & Moran, 2004). A time series of four pairs of unprocessed Landsat images
(path 233, row 83, and path 233, row 84) for the years 1975 (MSS), 1985 (TM), 1999 (ETMþ), and 2008 (TM) was used. Each
pair comprised two neighbouring scenes from the same date taken under relatively clear sky conditions (<10% cloud cover).
Due to a prevailing cloud cover in the southern coastal range of the study area, it was impossible to obtain pairs of images
from the same month for the whole 33-year period. To avoid major differences in phenology, all images obtained were taken
during the dry season (image dates from November to March) and in La Nin
˜a years. As fluctuations in precipitation are
relevant for the spectral response of biomass, all selected scenes represent similar drought conditions and correspond closely
to mean monthly values throughout the study period. All images were pre-processed, including geometric, atmospheric and
topographic corrections (Appendix A).
Fig. 1. Location of the study area in Central Chile, between 3351 00000–340705500 S and 712200000–710004800 W.
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447438
Author's personal copy
Nineteen land cover classes were initially defined. Four hundred and ninety-eight field points were taken with a GPS in
order to train the spectral signature of the selected land cover classes in a supervised classification scheme. Informal inter-
views of land owners and land managers were conducted during field survey to obtain information on previous and current
land cover and land use. This information was complemented with high-resolution imagery obtained from Google Earth to
account for areas with restricted accessibility. To train the classification, a region growing approach with the ‘‘seed’’ function
(PCI, 2000) was used. This approach starts with a set of ‘‘seed’’ points and appends neighbouring pixels to the seeds that have
same spectral properties. Training areas for the spectral signatures of older images were selected in those sites where land
cover remained unchanged or by using areas with similar spectral characteristics.
Signature separability of the initial classes for all images was evaluated using the Bhattacharyya distance. Based on this
distance, classes were iteratively merged until reasonably high signature separability was achieved. Specific separability
values for forest vs. timber plantations and for urban vs. bareland remained low (values of Bhattacharyya distance >1.3), but
these classes were not merged due to their importance for land planning. The iterative process of selecting consistent land
cover classes throughout the time series resulted in eight final land cover classes (Table 1). The average separability for all
signatures reached a value of Bhattacharyya distance >1.9, indicating good overall separability (PCI, 2000).
Classification of the resulting eight land cover classes was performed using the maximum likelihood algorithm. This
procedure has proven to be a robust and consistent classifier for multi-date classifications (e.g. Shalaby & Tateishi, 2007; Wu
et al., 2006;Yuan, Sawaya, Loeffelholz, & Bauer, 2005). To better discriminate between forests and timber plantations as well
as between the bareland and urban classes, post-classification processing was applied using ancillary data (Appendix A). The
MSS classification was re-sampled to a 30 30 m pixel size to allow multi-temporal comparison with the rest of the series.
Pre-processing and classification of remote sensing data were performed with PCI 7.0 (2000). Post-classification procedures
were performed with ArcMap 9.2 (ESRI, 2006).
Analysis of land cover change
The extent of the original satellite images varied slightlyand there were areas of shadows in the 1975 scenes and clouds in
the 1985 scenes. Therefore areas without data were subtracted from the whole time series before comparisons, and change
calculations were made. To account for these differences, a mask was created containing all pixels with no data from any of
the four classifications. This mask was applied to the entire set of time series.
To calculate the extentof each land cover class, we analysed classified maps using ArcGIS 9.2 (ESRI, 2006) and its extension
Spatial Analyst. A cross-tabulation procedure between the classifications was processed with IDRISI Andes (Clark Labs, 2006);
area change, gains, losses and persistence were calculated as proposed by Pontius, Shusasand, and McEachern (2004).
Analysis of change via cross-tabulation is a statistical method to identify signals of systematic processes within a land change
pattern (Pontius et al., 2004). Systematic transitions among classes were calculated and examined through the off-diagonal
entries of the cross-tabulation matrix. Analysis and mapping of the spatial distribution of transitions, persistence, gains and
losses were elaborated with the IDRISI Extension Land Change Modeler (Clark Labs, 2006). To create a map of persistence and
changes for the study area, binary change/no change maps were processed for each period. The three resulting maps were
added to obtain a map of persistence and change occurrence (1, 2 or 3) for the whole study period.
The annual rate of change for each class was calculated with the formula proposed by Puyravaud (2003):
r¼ð1=ðt2t1ÞÞlnðA2=A1Þ;
where A
2
and A
1
are the class areas at the end and the beginning, respectively, of the period being evaluated, and tis the
number of years spanning that period.
Accuracy assessment
Accuracy assessment involves identifying a set of sample locations (ground verification points) that are visited in the field.
The land cover found in the field is then compared to that mapped in the image for the same location by means of confusion
Table 1
Description of major land cover types defined in this study.
Class Description
Forest 75–100% canopy cover, advanced stage of succession of sclerophyllous forest with species like Cryptocarya alba,Peumus boldus,
Quillaja saponaria,Lithrea caustica and deciduous forest, mainly Nothofagus macrocarpa and Ribes punctatum
Shrubland 25–75% cover of shrub species, such as Acacia caven,Maytenus boaria,Prosopis chilensis,Trevoa trinervis,Colliguaja odorifera
Agriculture Rainfed and irrigated agriculture, wine yards, fruit orchards
Urban Urban and industrial areas
Bareland Rocks, beach and dunes, bare river beds, permanently degraded land, newly cleared land
Water Rivers, lakes, water reservoirs
Pasture Grassland with less than 25% shrub cover
Plantation Timber plantations with Pinus and Eucalyptus in advanced growth stage
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447 439
Author's personal copy
matrices. Validation of the 1999 ETMþand 2008 TM land cover maps was accomplished using 280 independent ground control
points. For plantations, only points from stands older than 12 years were considered. The 1975 MSS and 1985 TM land cover
maps were verified based on interpretation of the ground control points that had not changed over time (219 and 255 points,
respectively) using expert knowledge. For the 1975 MSS land cover maps, an additional map of plantations dated 1970 (INFOR,
1970) was georeferenced and used to identify control points for this class. Classification accuracy was first validated after
maximum likelihood classification and then again after post-classification modifications using cartographic information in GIS.
Overall accuracy and Cohen’s Kappa Index of Agreement (KIA) were calculated for each classification (Lu et al., 2004; Shao & Wu,
2008). Confusion matrices were processed using the Arc View Extension Kappa Tools 2.1a (Jenness & Wynne, 2006).
Results
Area change and change rates of land cover types
Over the whole study period, shrubland was the predominant land cover type, although it declined at an annual rate of
0.7%, from 43.3% of the study area in 1975 to 33.9% in 2008 (Fig. 2). Forest showed the largest decline in relation to its area,
with only about 58% (113,605 ha) of its extent in 1975 (195,773 ha) remaining in 2008, and an annual decline of 1.7%. Pasture
declined slightly at an annual rate of 0.2%, with around 94% (169,216 ha) of the 1975 extent (178,232 ha) remaining in 2008.
Other land cover types experienced an overall expansion. Thus, agriculture increased annually by 1.1% and expanded to
144% (265,102 ha) of the area occupied in 1975. Bareland reached 157% of its 1975 extent in 2008, with an annual growth rate
of 1.4%. A large increase was detected for urban areas as well, with an annual growth rate of 2.7%; these areas occupied 5.6% of
the study area in 2008, 241% of its 1975 extent. Timber plantation cover increased byover 288% in 2008 compared to the area
in 1975; although the annual change rate was the highest among all classes (3.2%) for the 1999–2008 period, timber plan-
tations covered only 3.4% of the study area in 2008 (Fig. 2).
Land cover changes did not occur at equal rates during all time intervals (Fig. 3). Between 1975 and 1985, forest experienced
a strong loss at an annual rate of 3.7%. This annual rate declined to 0.3% and 1.5% for the 1985–1999 and 1999–2008 periods,
respectively. Overall, forest losses during the three study periods were offset by about one third by forest gains. During the
period of highest forest loss (1975–1985), shrubland cover increased at an annual rate of 0.2%. From 1985 to 1999, the amount of
shrubland decreased at an annual rate of 0.6%, reaching a maximum annual loss of 2.0% during the 1999–2008 period. Overall
shrubland losses during the three study periods were offset by shrubland gains by about two thirds. Nonetheless, half of these
offsets came from forest to shrubland conversion and should therefore not be considered vegetation gain.
Agriculture rose very slightly between 1975 and 1985, but expanded at annual rates of 1.7% and 1.2% during the 1985–1999
and 1999–2008 periods, respectively. Urban areas spread at an annual rate of 5.7% between 1975 and 1985 and continued
expanding at annual rates of 1.0% and 2.0% during the 1985–1999 and 1999–2008 periods, respectively. Bareland increased at
annual rates of 3.3% and 4.0% during the 1975–1985 and 1999–2008 periods, respectively, but decreased at an annual rate of
1.7% during the 1985–1999 period. Gains and losses in pasture cover were high but compensated for each other over the
whole study period, resulting in annual rates of 1.3%, 1.1%, and 0.9% during the periods 1975–1985, 1985–1999 and 1999–
2008, respectively. The area of timber plantations remained relatively stable between 1975 and 1999, but experienced an
important expansion between 1999 and 2008, with an annual rate of 10.6%.
Spatial distribution of changes
The spatial distribution of intensity and patterns of land cover changes and persistence is shown in Fig. 4. During the entire
study period, we found that 28.2% of the study area was subject to only one change, 31.7% was subject to two changes, 16.8%
changed in all three time periods, and only 23.3% of the pixels remained unchanged. The majority of the unchanged pixels
(1975–2008) were shrubland (42.5%) and agriculture (25.3%), followed by forest (11.1%), urban (8.2%) and pasture (6.6%). An
extent of 2.5% of the study area was identified as permanent bareland.
The mostintense change dynamics were located in the coastal zone, where frequent exchanges between pasture and shrubland,
as well as between pasture, bareland and agricultural areas (particularly rotations between pastures, herbaceous crops and fallow
cycles), were found. Timber plantations, generally located in the flat coastal zone, increased and showed relatively high spatial
variability due to rotations betweenplantations and logged areas at the north-west coast and further south on the coastal plains. In
mountainous areas, changeswere less frequent andmainly consisted of theconversionof forest toshrubland and,to a lesser extent,
shrubland to forest. Agriculture expanded across the entire study area, particularly in the flat valleys from the coast to around
Santiago. Although the bottoms of some valleys remained as agriculture throughout the 33 years, an increase in agriculture
occurred in the foothills at the expense of shrubland and pasture. The increase in urban areas was related to the rapid growth of
Chile’scapital, Santiago, and the urban agglomeration of major cities located in the north-west partof the studyarea. Expansion of
urban areas was characterised by only one change throughout the three time periods and an aggregated spatial pattern.
Change trajectories among land cover types
The most consistent trend of inter-class change between 1975 and 2008 was a progressive loss of natural vegetation cover
(Fig. 5). Between 1975 and 1985, the major changes were a conversion of forest to shrubland (50,351 ha) and of shrubland to
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447440
Author's personal copy
Fig. 2. Land cover maps of the study area in Central Chile for the years 1975,1985, 1999 and 2008 and comparison of the respective extents of land cover classes
by percentage of study area (study area ¼1,265,204 ha).
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447 441
Author's personal copy
bareland (21,689 ha), pasture (14,022 ha) and urban areas (9356 ha, Fig. 5). Between 1985 and 1999, the major contributions
to net change were the conversion of shrubland to agriculture (40,338 ha) and pasture (18,726 ha). Shrubland regained some
area from bareland (14,647 ha), whereas bareland contributed to a gain in agriculture (12,589 ha). Between 1999 and 2008,
the foremost change was a loss of shrubland that contributed to increases in bareland (42,132 ha), agriculture (19,356 ha) and
timber plantations (14,973 ha). In contrast to previous periods, shrubland regained a small area from forests (9968 ha, Fig. 5).
Accuracy assessment
Classification accuracy increased notably after applying post-classification procedures, from overall agreements of 56.2%,
60.8%, 60.9%, and 72.3% to 65.8%, 77.3%, 78.9%, and 89.8% for the 1975 MSS, 1985 TM, 1999 ETMþ, and 2008 TM images,
respectively (results not shown). Cohen’s Kappa Index of Agreement (KIA) was 63.4%, 73.8%, 75.8% and 88.3% for the set of
post-processed images.
Discussion
Patterns of landscape change
Human interactions with ecosystems are inherently dynamic and complex, and any categorisation of these is an
oversimplification. However, there is little hope of understanding these interactions without such simplifications (Ellis &
Ramankutty, 2008). Working at the broad scale of this study has the advantage of providing general trends at the regional
scale that are useful for landscape planning and serve as a basis for analyzing drivers of land cover change. However, it has the
disadvantage of lower detectability of pattern and processes at the scale of land-use units in the real world (e.g. a field).
Nevertheless, informal interviews that were conducted accompanying the field survey give an important complementary
source of information to interpret detected changes at this regional scale and reinforce that observed changes followed
similar ground level land-use patterns throughout the study area.
Our analysis of land cover change in Mediterranean Central Chile reveals a general trend of a continuous reduction in
natural vegetation, i.e. forest and shrubland cover, that in turn has led to an increase in provisioning ecosystem services such
Fig. 3. Net change (i.e. gains minus losses), gains and losses for each land cover class as a percentage of the study area for the periods 1975–1985, 1985–1999 and
1999–2008 and for the whole study period, 1975–2008.
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447442
Author's personal copy
as food and timber production. This process takes place as a progressive modification from forest to shrubland vegetation, and
a highly dynamic conversion between shrubland and human-induced types of land cover. Nevertheless, deforestation rates in
this region are relatively low compared to rates in temperate forests in south-central Chile (Echeverria et al., 2006). This is
probably due to the fact that Central Chile has been densely populated since the times of colonisation and major conversions
of forest cover had taken place long before the 1970s (Conacher & Sala, 1998).
However, a relatively high amount of shrubland, the predominant vegetation cover in this semiarid landscape, was lost as
a consequence of conversions to intensive land uses, chiefly expanding agriculture and timber plantations and, to a lesser
extent, urbanization. This can be explained by an increase in local demand due to population growth and an open market
policy initiated after Chile’s economic crisis at the beginning of the 1970s (Camus, 2000; Silva, 2004). Agriculture and forestry
thereafter became the most important competitive producers (Camus, 2000). The strong increase in agriculture has been
stimulated by a combination of market liberalisation, incentives for newexport-oriented crops, introduction of new irrigation
technologies, and improvements in road infrastructure (Valde
´s & Foster, 2005). The expansion of timber plantations was
mostly a result of a government subsidy for tree-planting initiated in 1974 (Decree 701), which stimulated the planting of
P. radiata and E. globulus (Aronson et al., 1998). In the case of Central Chile, the rate of increase of timber plantations was the
highest of all classes in the 1999–2008 period. However, the expansion of timber plantations did not result in major
conversions of forest, as it did in southern Chile (Echeverria et al., 2006). Rapid expansion of urban areas, chiefly in the 1975–
1985 period, coincided with the abolishment of the urban limits by the Ministry of Housing and Urbanism (Decree 420) in
1979 and the liberalisation of the urban land market, both until 1985 (Kusnetzoff, 1987).
In our study region, forest loss patterns consisted mainly of the conversion of forest to shrubland and the reduction of
forest to remnants located on steep hills, where intensive use by humans is constrained by topography. The transformation of
forest to shrubland has been described as a continuous degradation of sclerophyllous forest, mostly driven by permanent
grazing pressure, firewood collection and charcoal production (Armesto et al., 2007; Balduzzi et al., 1982; Fuentes, Hoffmann,
Poiani, & Alliende, 1986; Rundel, 1999). In addition, successional recovery of forest is largelyconstrained by water availability,
soil erosion, lack of seed banks, disturbance by human-induced fires and limited regeneration capacities of forest species as
Fig. 4. Distribution of persistent pixels (i.e. those that never changed land cover type) and of pixels that showed one, two or three changes across the three
periods analysed from 1975 to 2008.
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447 443
Author's personal copy
compared to shrubland species (Armesto et al., 2007; Balduzzi et al., 1982; Conacher & Sala, 1998; Fuentes et al.,1986; Jimenez
& Armesto, 1992; Montenegro, Ginocchio, Segura, Keely, & Go
´mez, 2004; Rundel, 1999). This is evidenced in our analysis by
the large proportion of shrubland that remained unchanged over the whole study period. It has been argued that the loss of
forest and the change toward the predominance of shrubland cover represent a shift to an alternative ecosystem state
(Holmgren, 2002). In contrast to deforestation patterns in other dry forest regions, no direct conversion from forest to
agriculture (Izquierdo & Grau, 2009)orvice versa (e.g. Lasanta-Martı
´nez, Vicente-Serrano, & Cuadrat-Prats, 2005) occurred in
the study area.
We detected forest recovery through succession (on about 2.7% of the study area) over the whole study period as other
authors have documented in other Mediterranean areas (Serra, Pons, & Saurı
´,2008). Nonetheless, the latter process has been
halted, as mentioned above, by different physical and ecological factors. Thus, shrubland acts as a highly dynamic
compartment with large gains and losses over the three study periods as compared to other land cover changes. Bidirectional
changes (i.e. gains and losses) in shrubland cover resulted in the following spatial patterns: (1) exchanges between shrubland
and agriculture, agriculture and pasture, and pasture and shrubland took place in small patches scattered throughout the
relatively flat areas and partly explain the high dynamics detected in the studied landscapes; (2) such exchanges resulted in
a net loss of shrubland due to land-use intensification and agglomeration; and (3) from 1999 onward, these patterns have
spread to hillslopes, as indicated by the appearance of relatively large continuous patches of newly cleared bareland and
agriculture on the lower hillslopes, which were formerly covered by shrublands. This was motivated by important govern-
mental subsidies to encourage irrigation schemes (Maletta, 2001) and a collapse of the capacity of the flat areas to sustain the
increasing expansion of agricultural land.
Consequences of land cover change: implications for landscape planning and management
Vegetation loss and degradation reduce precipitation infiltration and runoff regulation, which promotes soil erosion and
has a negative impact on ground water recharge (Conacher & Sala, 1998; Millennium Ecosystem Assessment, 2005). In
addition, vegetation cover is highly correlated with water balance and regional climate regulation (Feddema et al., 2005; Foley
et al., 2005; Pielke, 2005). Changes in land use by humans and the resulting alterations in surface features and biogeophysical
processes influence weather and climate more immediately than the carbon cycle (Bonan, 2008; Pielke, 2005). Land-use
decisions therefore have consequences for the structure and function of ecosystems and affect environmental goods and
services; these decisions also affect humans in ways that go beyond the immediate land-use situation (Turner et al., 2007).
The continuous degradation of the vegetation cover could have a strong impact on human livelihood and well-being in
Central Chile as well as other dryland landscapes, as there are increasing water demands for agriculture (Cai, Ringler, & You,
2008) and human consumption due to large population increases.
Fig. 5. Major change trajectories and their contributions to net change in percentage of the study area (thick lines correspond to net change >3.2%, intermediate
lines correspond to net changes between 1.6% and 3.2%, and thin lines correspond to net change <1.6%; only net contributions to change >10,000 ha or 0.8% of the
study area are represented).
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447444
Author's personal copy
Environmental problems like degradation, loss of biodiversity and decreases in productivity accumulate over the long
term and have non-linear effects on regional to global scales (DeFries, Foley, & Asner, 2004; Foley et al., 2005). Consequently,
strategies for adapted land use, including the optimisation of the spatial configuration of uses and restoration of the natural
vegetation cover should be developed quickly. Vegetation cover within the landscape mosaic must be carefully considered in
planning to sustain habitat and regulation functions and enhance the productive capacities of the landscape. Strategies should
go beyond preservation within protected areas and logging restrictions along rivers and streams (Turner et al., 2007). For
instance, Rey Benayas, Bullock, and Newton (2008) proposed the ‘‘woodland-islet in agricultural seas’’ model to conciliate
agricultural production and conservation or restoration of native woodlands. Closer monitoring is needed for cattle grazing
stocks to establish guidelines for an adapted carrying capacity, as cattle graze on pastures, shrubland and in forests, which are
all mainly private land and do not have adequate use restrictions. The repercussions of firewood extraction and charcoal
production have hardly been quantified in Central Chile, but we know that firewood and derived charcoal provided around
18% of the national energy supply between 1990 and 2007, while firewood consumption doubled in this period (CNE, 2008).
Some estimates by Dubroeucq and Livenais (2004) in northern Chile show that the impact of firewood extraction on vege-
tation cover should not be underestimated.
Apart from the need for land-use planning, restoration and rehabilitation are important issues in drylands (Le Houerou,
2000; Vallejo, Aronson, Pausas, & Cortina, 2006). Holmgren and Scheffer (2001) postulated that there might be a window of
opportunity for passive restoration through the exclusion of herbivores in ENSO years due to higher water availability, which
Gutie
´rrez, Holmgren, Manrique, and Squeo (2007) have experimentally shown for drier zones further north in Chile. It could
be especially interesting to use this strategy to establish buffer zones and corridors between remaining old growth forest,
which were detected in this study as stable forest areas. Land-use planning to ensure the long-term maintenance of landscape
functions that are of common societal concern has not yet been established in Chile, where territorial planning legislation
mainly focuses on urban areas, infrastructure and industrial development. Recently, some efforts have been made in land-
scape planning for protected areas (Oltremari & Thelen, 2003). Also, forms of adaptive and multifunctional land use like
mixed agroforestry systems should be encouraged as an alternative to monoculture cropping and crop pasture rotations
(Aronson et al., 1998; Ovalle, Del Pozo, Casado, Acosta, & de Miguel, 2006).
Conclusion
Several uncertainty factors underlie the classification of satellite imagery into land cover types, and such classification is
never completely accurate (Shao & Wu, 2008). However, this work has estimated the extent of land cover in Mediterranean
Central Chile, characterised the respective changes, and assessed the dynamics and stepwise vegetation cover loss that has
taken place over the last 33 years. Our case study provides further evidence of how Mediterranean regions show a constant
transformation of their ecological systems. This analysis illustrates how natural vegetation cover tends to diminish in a very
subtle and slow fashion due to passive revegetation that partly counterbalances vegetation loss. Nevertheless, forest cover is
being degraded, and many areas do not recover, but remain as shrubland. Shrubland, in turn, is lost to intensive land uses like
agriculture and timber plantations. Land-use changes in Mediterranean regions must not always be interpreted as the loss of
a specific set of ecological conditions based on values established in the world of the West, but as a change in the ecosystem
services that dynamic land cover types provide to humans in culturally distinct regions. However, our interviews conducted
during the field survey revealed a high awareness of local population regarding the benefits that forests provide and the
consequences of forest loss, including reduced water provision and erosion problems. The successful identification of change
trajectories provides a critical component for land use, conservation and restoration planning in dry landscapes. The
investigation of regional dynamics provides a basis for future analysis of drivers and circumstances that enhance change or
stability of land cover.
Acknowledgements
This work was financed by the European Commission, REFORLAN Project, INCO Contract CT2006-032132. The manuscript
benefited from useful comments provided by M. Gonza
´lez-Espinosa, R. M. Navarro Cerrillo, and an anonymous reviewer.
Maps of plantations from 1970 (INFOR, 1970) were provided by the Ibero Amerikanisches Institut, Berlin.
Appendix. Supplementary material
Supplementary data associated with this article can be found in the online version, at doi:10.1016/j.apgeog.2009.12.003.
References
Amigo, J., & Ramı
´rez, C. (1998). A bioclimatic classification of Chile: woodland communities in the temperate zone. Plant Ecology, 136(1), 9–26.
Antrop, M. (2005). Why landscapes of the past are important for the future. Landscape and Urban Planning, 70(1–2), 21–34.
Armesto, J. J., Arroyo, K., Mary, T., & Hinojosa, L. F. (2007). The Mediterranean environment of Central Chile. In T. T. Velben, K. R. Young, & A. R. Orme (Eds.),
The physical geography of South America (pp. 184–199). New York: Oxford University Press.
Aronson, J., Del Pozo, A., Ovalle, C., Avendan
˜o, J., Lavin, A., & Etienne, M. (1998). Land use changes and conflicts in Central Chile. In P. W. Rundel, G.
Montenegro, & F. Jaksic (Eds.), Landscape disturbance and biodiversity in Mediterranean-type ecosystem (pp.155–168). Berlin, Heidelberg: Springer Verlag.
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447 445
Author's personal copy
Badano, E. I., Cavieres, L. A., Molina-Montenegro, M. A., & Quiroz, C. L. (2005). Slope aspect influences plant association patterns in the Mediterranean
matorral of Central Chile. Journal of Arid Environments, 62(1), 93–108.
Balduzzi, A., Tomaselli, R., Serey, I., & Villasen
˜or, R. (1982). Degradation of the Mediterranean type of vegetation in Central Chile. Ecologia Mediterranea, 7,
223–240.
Bonan, G. B. (2008). Forests and climate change: forcings, feedbacks, and the climate benefits of forests. Science, 320(5882), 1444–1449.
Cai, X., Ringler, C., & You, J. Y. (2008). Substitution between water and other agricultural inputs: implications for water conservation in a river basin context.
Ecological Economics, 66(1), 38–50.
Camus, P. (2000). Innovacio
´n agroproductiva y ordenamiento del territorio. El caso del desarollo forestal chileno. Scripta Nova, 69(85). Retrieved January 26,
2009 from: http://www.ub.es/geocrit/sn-69-85.htm.
Carmel, Y., & Flather, C. H. (2004). Comparing landscape scale vegetation dynamics following recent disturbance in climatically similar sites in California and
the Mediterranean basin. Landscape Ecology, 19(6), 573–590.
Clark Labs. (2006). IDRISI Andes (version 15.00). Worcester: Clark Labs, Clark University.
CNE. (2008). Encuestas balance de energı
´a a empresas del sector. Comisio
´n Nacional De Energı
´a. Retrieved January 26, 2009 from: http://www.cne.cl/
estadisticas/anuario/index.html.
Conacher, A. J., & Sala, M. (1998). Land degradation in Mediterranean environments of the world: Nature and extent, causes and solutions. Chichester: John
Wiley and Sons Ltd.
Cowling, R. M., Rundel, P. W., Lamont, B. B., Kalin Arroyo, M., & Arianoutsou, M. (1996). Plant diversity in Mediterranean-climate regions. Trends in Ecology
and Evolution, 11(9), 362–366.
DeFries, R. S., Foley, J. A., & Asner, G. P. (2004). Land-use choices: balancing human needs and ecosystem function. Frontiers in Ecology and the Environment,
2(5), 249–257.
Donoghue, D. N. M. (2002). Remote sensing: environmental change. Progress in Physical Geography, 26(1), 144–151.
Dubroeucq, D., & Livenais, P. (2004). Land cover and land use changes in relation to social evolution – a case study from Northern Chile. Journal of Arid
Environments, 56(2), 193–211.
Echeverria, C., Coomes, D., Salas, J., Rey-Benayas, J. M., Lara, A., & Newton, A. (2006). Rapid deforestation and fragmentation of Chilean temperate forests.
Biological Conservation, 130(4), 481–494.
Ellis, E. C., & Ramankutty, N. (2008). Putting people in the map: anthropogenic biomes of the world. Frontiers in Ecology and the Environment, 6(8), 439–447.
ESRI. (2006). ArcGIS (version 9.2). Redlands: Environmental Systems Research Institute Inc.
Feddema, J. J., Oleson, K. W., Bonan, G. B., Mearns, L. O., Buja, L. E., Meehl, G. A., et al. (2005). The importance of land-cover change in simulating future
climates. Science, 310(5754), 1674–1678.
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(1), 19–35.
Foley, J. A., DeFries, R., Asner, G. P., Barford, C., Bonan, G., Carpenter, S. R., et al. (2005). Global consequences of land use. Science, 309(5734), 570–574.
Fuentes, E. R., Avile
´s, R., & Segura, A. (1989). Landscape change under indirect effects of human use: the savanna of Central Chile. Landscape Ecology, 2(2),
73–80.
Fuentes, E. R., Hoffmann, A. J., Poiani, A., & Alliende, M. C. (1986). Vegetation change in large clearings: patterns in the Chilean matorral. Oecologia, 68(3),
358–366.
Geri, F., Amici, V., & Rocchini, D. (2010). Human activity impact on the heterogeneity of a Mediterranean landscape. Applied Geography, 29. doi:10.1016/j.
apgeog.2009.10.006.
Gutie
´rrez, J. R., Holmgren, M., Manrique, R., & Squeo, F. A. (2007). Reduced herbivore pressure under rainy ENSO conditions could facilitate dryland
reforestation. Journal of Arid Environments, 68(2), 322–330.
Holmgren, M. (2002). Exotic herbivores as drivers of plant invasion and switch to ecosystem alternative states. Biological Invasions, 4(1), 25–33.
Holmgren, M., & Scheffer, M. (2001). El Nin
˜o as a window of opportunity for the restoration of degraded arid ecosystems. Ecosystems, 4(2), 151–159.
INFOR. (1970). Plano de las plantaciones forestales de la zona central de Chile, Provincias de Aconcagua, Vaparaiso, Santiago, O
´Higgins, Colchagua, Curico y Talca.
Santiago de Chile: Instituto Forestal & Instituto Geogra
´fico Militar.
Izquierdo, A. E., & Grau, H. R. (2009). Agriculture adjustment, land-use transition and protected areas in Northwestern Argentina. Journal of Environmental
Management, 90(2), 858–865.
Jenness, J., & Wynne, J. J. (2006). Kappa analysis extension for ArcView 3.x. (version 2.1). Flagstaff: Jenness Enterprises. Retrieved January 26, 2009 from:
http://www.jennessent.com/arcview/kappa_stats.htm.
Jimenez, H. E., & Armesto, J. J. (1992). Importance of the soil seed bank of disturbed sites in Chilean matorral in early secondary succession. Journal of
Vegetation Science, 3(5), 579–586.
Johnson, D. L., & Lewis, L. A. (2006). Land degradation: Creation and destruction. New York: Rowman and Littlefield.
Kusnetzoff, F. (1987). Urban and housing policies under Chile’s military dictatorship 1973–1985. Latin American Perspectives, 14(2), 157–186.
Lasanta, T., Gonza
´lez-Hidalgo, J. C., Vicente-Serrano, S. M., & Sferi, E. (2006). Using landscape ecology to evaluate an alternative management scenario in
abandoned Mediterranean mountain areas. Landscape and Urban Planning, 78(1–2), 101–114.
Lasanta-Martı
´nez, T., Vicente-Serrano, S. M., & Cuadrat-Prats, J. M. (2005). Mountain Mediterranean landscape evolution caused by the abandonment of
traditional primary activities: a study of the Spanish Central Pyrenees. Applied Geography, 25(1), 47–65.
Le Houerou, H. N. (2000). Restoration and rehabilitation of arid and semiarid Mediterranean ecosystems in North Africa and West Asia: a review. Arid Soil
Research and Rehabilitation, 14, 3–14.
Lo
´pez, S., & Sierra, R. (2010). Agricultural change in the Pastaza River Basin: a spatially explicit model of native Amazonian cultivation. Applied Geography, 29.
doi:10.1016/j.apgeog.2009.10.004.
Lu, D., Mausel, P., Brondizio, E., & Moran, E. (2004). Change detection techniques. International Journal of Remote Sensing, 25(12), 2365–2407.
Maletta, H. E. (2001). Recent Latin American developments in irrigation. Social Science Research Network. doi:10.2139/ssrn.347680.
Millennium Ecosystem Assessment. (2005). Ecosystems and human well-being. Current state and trends. Washington, D.C.: Island Press.
Montenegro, G., Ginocchio, R., Segura, A., Keely, J. E., & Go
´mez, M. (2004). Fire regimes and vegetation responses in two Mediterranean-climate regions.
Revista Chilena de Historia Natural, 77, 455–464.
Mouillot, F., Ratte, J.-P., Joffre, R., Mouillot, D., & Rambal, S. (2005). Long-term forest dynamic after land abandonment in a fire prone Mediterranean
landscape (central Corsica, France). Landscape Ecology, 20(1), 101–112.
Myers, N., Mittermeier, R. A., Mittermeier, C. G., da Fonseca, G. A. B., & Kent, J. (2000). Biodiversity hotspots for conservation priorities. Nature, 403, 853–858.
Oltremari, J. V., & Thelen, K. D. (2003). Planificacio
´ndea
´reas silvestres protegidas. Santiago, Chile: Comision Nacional Del Medio Ambiente & FAO. Editoral
Salesianos.
Ovalle, C., Avendan
˜o, J., Aronson, J., & Del Pozo, A. (1996). Land occupation patterns and vegetation structure in the anthropogenic savannas (espinales) of
Central Chile. Forest Ecology and Management, 86(1–3), 129–139.
Ovalle, C., Del Pozo, A., Casado, M. A., Acosta, B., & de Miguel, J. M. (2006). Consequences of landscape heterogeneity on grassland diversity and productivity
in the Espinal agroforestry system of Central Chile. Landscape Ecology, 21(4), 585–594.
PCI. (2000). PCI works (version 7.0). Richmond Hill: PCI Geomatics.
Pielke, R. A. (2005). Land use and climate change. Science, 310(5795), 1625–1626.
Pontius, R. G., Jr., Shusasand, E., & McEachern, M. (2004). Detecting important categorical land changes while accounting for persistence. Agriculture
Ecosystems and Environment, 101(2–3), 251–268.
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447446
Author's personal copy
Pueyo, J., & Beguerı
´a, S. (2007). Modeling the rate of secondary succession after farmland abandonment in a Mediterranean mountain area. Landscape and
Urban Planning, 83(4), 245–254.
Puyravaud, J. P. (2003). Standardizing the calculation of the annual rate of deforestation. Forest Ecology and Management, 177(1–3), 593–596.
Rey Benayas, J. M., Bullock, J., & Newton, A. C. (2008). Creating woodland islets to reconcile ecological restoration, conservation, and agricultural land use.
Frontiers in Ecology and the Environment, 6(6), 329–336.
Rey Benayas, J. M., Newton, A. C., Dı
´az, A., & Bullock, J. (2009). Enhancement of biodiversity and ecosystem services by ecological restoration: a meta-
analysis. Science, 325(5944), 1121–1124.
Reynolds, J. F., Maestre, F. T., Kemp, P. R., Stafford-Smith, D. M., & Lambin, E. (2007). Natural and human dimensions of land degradation in drylands: causes
and consequences. In J. G. Canadell, D. E. Pataki, & L. F. Pitelka (Eds.), Terrestrial ecosystems in a changing world (global change – The IGBP series) (pp. 247–
257). Berlin, Heidelberg: Springer.
Reynolds, J. F., & Stafford Smith, D. M. (2002). Do humans cause deserts? In J. F. Reynolds, & D. M. Stafford Smith (Eds.), Global desertification: Do humans
cause deserts? (pp. 1–22) Dahlem University Press.
Romero-Calcerrada, R., & Perry, G. L. W. (2004). The role of land abandonment in landscape dynamics in the SPA ‘Encinares del rı
´o Alberche y Cofio’, Central
Spain, 1984–1999. Landscape and Urban Planning, 66(4), 217–232.
Rundel, P. W. (1999). Disturbance in Mediterranean climate shrublands and woodlands. In L. R. Walker (Ed.), Ecosystems of disturbed ground (pp. 271–285).
New York: Elsevier.
Serra, P., Pons, X., & Saurı
´, D. (2008). Land-cover and land-use change in a Mediterranean landscape: a spatial analysis of driving forces integrating
biophysical and human factors. Applied Geography, 28(3), 189–209.
Shalaby, A., & Tateishi, R. (2007). Remote sensing and GIS for mapping and monitoring land cover and land-use changes in the Northwestern coastal zone of
Egypt. Applied Geography, 27(1), 28–41.
Shao, G., & Wu, J. (2008). On the accuracy of landscape pattern analysis using remote sensing data. Landscape Ecology, 23(5), 505–511.
Silva, E. (2004). The political economy of forest policy in Mexico and Chile. Singapore Journal of Tropical Geography, 25, 261–280.
Turner, B. L., II, Lambin, E. F.,& Reenberg, A. (2007). The emergence of land change science for global environmental change and sustainability. Proceedings of
the National Academy of Sciences of the United States of America, 104(52), 20666–20671.
Turner, B. L., Moss, R. H., & Skole, D. L. (1993). Relating land use and global land-cover change: A proposal for an IGBP-HDP core project. Report from the IGBP-
HDP Working Group on Land-Use/Land-Cover Change. Joint publication of the International Geosphere-Biosphere Programme (Report No. 24) and the
Human Dimensions of Global Environmental Change Programme (Report No. 5). Stockholm: Royal Swedish Academy of Sciences.
Valde
´s, A., & Foster, W. (2005). Externalidades de la Agricultura Chilena. Santiago de Chile: Edicio
´nes Universidad Cato
´lica de Chile.
Vallejo, R., Aronson, J., Pausas, J. G., & Cortina, J. (2006). Restoration of Mediterranean woodlands. In J. van Andel, & J. Aronson (Eds.), Restoration ecology: The
new frontier (pp. 193–207). Malden: Blackwell Publishing.
Vitousek, P. M. (1994). Beyond global warming: ecology and global change. Ecology, 75(7), 1861–1876.
Vogiatzakis, I. N., Mannion, A. M., & Griffiths, G. H. (2006). Mediterranean ecosystems: problems and tools for conservation. Progress in Physical Geography,
30(2), 175–200.
Wu, Q., Li, H., Wang, R., Paulussen, J., Hec, Y., Wang, M., Wang, B., & Wang, Z. (2006). Monitoring and predicting land use change in Beijing using remote
sensing and GIS. Landscape and Urban Planning, 78(4), 322–333.
Yuan, F., Sawaya, K. E., Loeffelholz, B. C., & Bauer, M. E. (2005). Land cover classification and change analysis of the twin cities (Minnesota) metropolitan area
by multitemporal landsat remote sensing. Remote Sensing of Environment, 98(2–3), 317–328.
J.J. Schulz et al. / Applied Geography 30 (2010) 436–447 447
... Therefore, mapping West Kordofan Region land cover is urgent to public and scientific society, ranging from societal needs to sustainable land cover management. Remote sensing has long been distinguished as an effective tool for large-scale land cover monitoring (Rogan et al., 2003, Hansen and Loveland, 2012, Schulz et al., 2010, Foody, 2001, Rawat and Kumar, 2015 [19,14,21,10,18] , and they are useful for land cover mapping. It was, therefore, used efficiently in a number of land cover classification, mapping, and change detection studies using different satellite images with various resolutions (Baker et al., 2006, Friedl et al., 2002, Hansen et al., 2000, Hansen et al., 2013 [3,11,13,15] . ...
... Therefore, mapping West Kordofan Region land cover is urgent to public and scientific society, ranging from societal needs to sustainable land cover management. Remote sensing has long been distinguished as an effective tool for large-scale land cover monitoring (Rogan et al., 2003, Hansen and Loveland, 2012, Schulz et al., 2010, Foody, 2001, Rawat and Kumar, 2015 [19,14,21,10,18] , and they are useful for land cover mapping. It was, therefore, used efficiently in a number of land cover classification, mapping, and change detection studies using different satellite images with various resolutions (Baker et al., 2006, Friedl et al., 2002, Hansen et al., 2000, Hansen et al., 2013 [3,11,13,15] . ...
Article
Full-text available
Mapping land cover is fundamental way to support environmental survey and monitoring of ecosystem transition status. This study aimed to map the major land cover categories in order to shed more light on natural environmental processes and human induced in West Kordofan Region. We classified land cover categories by using supervised classification based on Maximum Likelihood Classifier. Then, in 2015 we found that, sand is one of the major land cover categories with 8374.56 km 2 ; others land cover includes water, vegetation, agriculture and bare lands with percentage area of 0.09 %, 20.4 %, 17.06 % and 55.09 % respectively. We also found that Landsat ETM+ is a powerful tool for land cover mapping, and we obtained a classification map with an overall accuracy of 85.7 % and Kappa statistic at 0.80. The resultant represents constant reality in discrete land cover categories, showing where and how land deterioration is likely to occur. This study can contribute to future policies and planning in the region.
... Using a maximum likelihood classification of a single Landsat scenes at three points in time (1975,1990,2000), Echeverria et al. [27] show that forest ecosystems are being replaced by industrial tree plantations. Other studies work on the same dataset or also on single Landsat scenes [26,[28][29][30][31][32][33][34]. Heilmayr et al. [35] study a much larger region using 42 Landsat scenes at three time points (1985-1987, 1999-2001, 2009-2011). ...
... The results of individual case studies [26,[28][29][30][31][32][33][34], have thus been scaled up in both temporal [37] and spatial terms [35,39,108]. The results are consistent in their main statements about the main processes that took place. ...
Article
Full-text available
In southern Chile, an establishment of a plantation-based forest industry occurred early in the industrial era. Forest companies claim that plantations were established on eroded lands. However, the plantation industry is under suspicion to have expanded its activities by clearing near-natural forests since the early 1970s. This paper uses a methodologically complex classification approach from own previously published research to elucidate land use dynamics in southern Chile. It uses spatial data (extended morphological profiles) in addition to spectral data from historical Landsat imagery, which are fusioned by kernel composition and then classified in a multiple classifier system (based on support, import and relevance vector machines). In a large study area (~67,000 km2), land use change is investigated in a narrow time frame (five-year steps from 1975 to 2010) in a two way (prospective and retrospective) analysis. The results are discussed synoptically with other results on Chile. Two conclusions can be drawn for the coastal range. Near-natural forests have always been felled primarily in favor of the plantation industry. Vice versa, industrial plantations have always been primarily established on sites, that were formerly forest covered. This refutes the claim that Chilean plantations were established primarily to restore eroded lands; also known as badlands. The article further shows that Chile is not an isolated case of deforestation by afforestation, which has occurred in other countries alike. Based on the findings, it raises the question of the extent to which the Chilean example could be
... Species Distribution Models (SDM) are a very useful tool in ecological studies (Guisanet al., 2013), some authors consider it a discipline part of macroecology and biogeography, which serves to publicize the distribution areas of species (Soberónet al., 2017), also have become a very dynamic and active line of research today, although, they are involved in conceptual problems that remain ambiguous (Soberón and Nakamura, 2009). DEM focuses on understanding the distribution of species around the transformation of ecosystems (Schulz et al., 2010), processes of change, construction, expansion, and fragmentation (Chen et al., 2011), in this ecosystemic evolution, there is a constant ecological dynamic, in which both the biota and the biotope are affected, which is why the habitat of the species changes. This is where DEMs allow estimating the relationship between species records and environmental characteristics (Franklin, 2010), represented in cartographic models of probability in space. ...
Article
Full-text available
Species distribution models (SDM), in many disciplines, have become an important tool in the environmental management of the territory, this includes several lines of knowledge, such as health, epidemiology, biodiversity, conservation, and forestry studies. This article is oriented to describe the importance, procedures, analysis, applications, and controversies that have arisen around this useful tool. This article focuses on highlighting research applications in forest management and conservation, as well as its applications and possible challenges for the future of forestry systems.
... ~3,5 °C (RCP 8.5) y una disminución de la precipitación de ~3% (RCP 2.6) a ~30% (RCP 8.5) para el año 2100(Bozkurt et al., 2018). Por lo tanto, se prevé un escenario complejo para los recursos vegetales e hídricos, y a su vez para la utilización directa e indirecta de estos recursos por la población y las comunidades locales, especialmente de la vegetación andina de los Andes central (> 2.000 m s.n.m.)(Payne et al., 2017;Cavieres et al., 2021).Junto al cambio climático, los ecosistemas de esta región se encuentran presionados por la sustitución de ecosistemas nativos por cultivos agrícolas y plantaciones forestales, expansión urbana, tala no sustentable e incendios forestales, provocando la fragmentación de hábitat y pérdida de biodiversidad(Schulz et al., 2010;Hernández et al., 2016). A su vez, estos ecosistemas son la base para el desarrollo de la ganadería trashumante, que corresponde al pastoreo de tipo intensivo en continuo movimiento donde el ganado permanece aislado en la zona montañosa y deambula para buscar las mejores zonas de alimentación(Razeto, 2007). ...
Chapter
Full-text available
Anglo American opera la mina Los Bronces, yacimiento con más de 150 años de historia, en la alta cordillera de la Región Metropolitana, en un ecosistema mediterráneo de alta sensibilidad. Por ello, asumimos como parte de nuestra gestión la importancia de avanzar rápidamente hacia una visión ecosistémica e integral del territorio, del conjunto de elementos presentes en él y de cómo nuestra actividad interactúa con este entorno. A partir de esta mirada, estamos implementando diversas medidas para reducir en forma progresiva nuestra huella ambiental en el marco de un plan con metas de corto, mediano y largo plazo, junto con asumir también la necesidad de generar estrategias que permitan proteger y conservar la biodiversidad del territorio. En el nuevo escenario global surge la necesidad de generar y compartir más conocimiento sobre nuestro entorno con una mirada de desarrollo sostenible. Durante años hemos estudiado –junto a universidades, científicos y expertos– la zona de montaña de Lo Barnechea y sus principales cuencas, para asegurar su sustentabilidad en el largo plazo. Esto nos ha permitido contar con información significativa sobre la situación de los glaciares, de los recursos hídricos y la flora y fauna que rodea a Los Bronces. Los resultados obtenidos durante el desarrollo de los diferentes trabajos realizados representan un importante cúmulo de conocimientos del más alto nivel, validados por la comunidad científica, y son un enorme aporte a la ciencia nacional, así como a la gestión territorial. Esperamos que los datos, proyecciones y propuestas descritas en esta publicación se conviertan en un insumo útil para la discusión de políticas públicas, estrategias de conservación y actividades productivas que conversen de mejor manera con la situación actual y futura de este valioso ecosistema. Sabemos que existen muchos desafíos sociales y ambientales en la zona que requieren de la participación activa de diversos actores locales y, dentro de ese panorama, este libro se incorpora como una nueva capa de evidencia que nos permite comprender cómo el funcionamiento de este sistema cambiará en el futuro y cómo podemos tomar parte en su conservación. Adicionalmente, esperamos que su lectura les resulte provechosa y los reencante con estos ecosistemas tanto como a nosotros.
... This pattern follows the global trend, where large areas have been degraded to multiple patches that are smaller and disconnected and put the survival of the species at risk (Fahrig, 2003;Wu et al., 2021). These trends have also been reported in other Latin American countries, such as Chile (Hernández et al., 2016;Schulz et al., 2010) and Brazil (Salazar et al., 2021). The accelerated rate of forest fragmentation in the NPA region forced the SB to retreat from there, avoiding contact within the community, increasing the overlap zones between human economic activities and SB (Rojas et al., 2019). ...
Article
Full-text available
Spectacled bears (SB) (Tremarctos ornatus) are the only bear species native to South America. This particular bear is the single species of its genus, and it is listed as vulnerable according to the IUCN red list. A critical SB conservation habitat is in the rural territories of the Peruvian Amazon, where anthropogenic land-use changes and landscape fragmentation threaten SB habitats. The following questions arise in this context: How much has land-use changed? How to design the establishment of ecological corridors (ECs) to support the conservation of SB?. We investigated the temporal land use and land cover changes for last 30 years (1990-2020) for a better projection of the ECs and to quantify the temporal landscape metrics. Furthermore, we integrated cloud computing, machine learning models with cost-effective techniques to delineate the ECs for SB within the rural territories. Ensemble Random Forest model associated with Google Earth Engine (GEE) was used to develop four land use and land cover (LULC) maps (for the years 1990, 2000, 2010 and 2020). The least cost path (LCP) model based on Dijkstra's shortest path algorithm was assembled based on six variables (altitude; slope; distance to roads; distance to population centers; land use map; inventory map of SB). Then, we calculated the ECs based on the multidirectional origin-destination points, we found that forest patches increased by 57% Global Ecology and Conservation 36 (2022) e02126 2 between 1990 and 2020. Results showed statistically significant agreement (R 2 = 0.47; p < 0.05) between cost/ha* and percentage of forest cover. We observed that the higher the forest cover, the better the connectivity and the lower the cost of mobilization in the ECs. Our study outcomes validated through the images obtained from trap cameras that confirms that delineated routs for SB movements. The proposed model can be adopted for other parts of the global forest including other species of interest. To formulate a sustainable conservation action plan, we provided five recommendations that will support conservation practices, design cost-effective ECs for policy makers.
... As suggested by Bayar and Karabacak (2017), changes in land use also affect the functioning and biodiversity of the ecosystem Sala et al., 2000), soil degradation (Tolba et al., 1992), and global, regional, and local climatic changes (Chase et al., 2000;Houghton et al., 1999). Along with the advancements in space technology, monitoring of land use based on satellite images and incorporating the same into land use studies (Peiman, 2011;Schulz et al., 2010;Tovar et al., 2013;Vittek et al., 2014) added a different dimension to the land use studies. There are a number of studies in Turkey, especially since the 1950s, on land utilization, land uses, and classification of land (Arınç, 2003;Atalay, 1989;Bayar, 2004;Elmastaş, 2008;Erol, 1959;Erol, 1977;Gözenç, 1974;Özçağlar, 1994;Özçağlar et al., 2006;Tunçdilek, 1985;Yiğitbaşıoğlu, 1993). ...
... This pattern follows the global trend, where large areas have been degraded to multiple patches that are smaller and disconnected and put the survival of the species at risk (Fahrig, 2003;Wu et al., 2021). These trends have also been reported in other Latin American countries, such as Chile (Hernández et al., 2016;Schulz et al., 2010) and Brazil (Salazar et al., 2021). The accelerated rate of forest fragmentation in the NPA region forced the SB to retreat from there, avoiding contact within the community, increasing the overlap zones between human economic activities and SB (Rojas et al., 2019). ...
Preprint
Spectacled bears (SB) (Tremarctos ornatus) are the only bear species native to South America. This particular bear is the single species of its genus, and it is listed as vulnerable according to the IUCN red list. A critical SB conservation habitat is in the rural territories of the Peruvian Amazon, where anthropogenic land-use changes and landscape fragmentation threaten SB habitats. The following questions arise in this context: How much has land-use changed? How to design the establishment of ecological corridors (ECs) to support the conservation of SB?. We investigated the temporal land use and land cover changes for last 30 years (1990-2020) for a better projection of the ECs and to quantify the temporal landscape metrics. Furthermore, we integrated cloud computing, machine learning models with cost-effective techniques to delineate the ECs for SB within the rural territories. Ensemble Random Forest model associated with Google Earth Engine (GEE) was used to develop four land use and land cover (LULC) maps (for the years 1990, 2000, 2010 and 2020). The least cost path (LCP) model based on Dijkstra’s shortest path algorithm was assembled based on six variables (altitude; slope; distance to roads; distance to population centers; land use map; inventory map of SB). Then, we calculated the ECs based on the multidirectional origin-destination points, we found that forest patches increased by 57% between 1990 and 2020. Results showed statistically significant agreement (R² = 0.47; p<0.05) between cost/ha* and percentage of forest cover. We observed that the higher the forest cover, the better the connectivity and the lower the cost of mobilization in the ECs. Our study outcomes validated through the images obtained from trap cameras that confirms that delineated routs for SB movements. The proposed model can be adopted for other parts of the global forest including other species of interest. To formulate a sustainable conservation action plan, we provided five recommendations that will support conservation practices, design cost-effective ECs for policy makers.
... In this regard, remotely sensed satellite data together with GIS techniques offers promising tool for accurate assessment and monitoring of land use through observation of land cover (Karl and Maurer, 2010;Rozenstein and Karnieli, 2011;Liping et al., 2018;Hamud et al., 2019;Sánchez-Espinosa et al., 2019;Schulz et al., 2021;Shegwe et al., 2021). Several previous studies reported the use of Landsat Thematic Mapper (TM) satellite data having 30 m resolution for precise land use classification (Manandhar et al., 2009;Schulz et al., 2010;Cai et al., 2019;Xu et al., 2020;Kumar et al., 2021;Hussain et al., 2022;Sarif and Gupta, 2022). Landsat satellite imagery is commonly used for land use classification on regional scale due to their relatively lower cost, longer history, and acquire measurements in all major portions of the solar electromagnetic spectrum (visible, nearinfrared, and shortwave-infrared) (Rozenstein and Karnieli, 2011). ...
Article
Full-text available
Rapid urbanization is among the dominant causes of land use changes all over the world. Therefore, detection of land use changes of a particular area is a matter of great concern. In this scenario, geo-spatial techniques offer promising tool for land use classification of an area over variable time. In this regard, present study was carried out for land use classification of district Sialkot. For this purpose, Landsat satellite images (Landsat 5 TM, Landsat 8 OLI) for the year 1990, 2000 and 2016 of district Sialkot were obtained through United State Geological Survey (USGS). Pre-processing of the obtained satellite imagery i.e., layer stacking, image enhancement etc. was performed using Erdas imagine 9.0 and Arc Map 10.1 software. Supervised classification of satellite images was carried out and the total area of district Sialkot was classified in to four different land use classes i.e., agriculture land, bare soil, build up area and water bodies. According to the results, the total agriculture area in 1990 was 2109.60 km 2 which was decreased in 2000 (2027.07 km 2) and increased in 2016 (2354.65 km 2). Bare soil area (809.70 km 2) in 1990 was found decreased and reached up to 391.45 km 2 in 2016. Build up area was also increased in district Sialkot and reached at 176.59 km 2 in 2016. So based on results, it can be concluded that land use classes such as agriculture and built-up area were increased while the area covered by bare soil exhibited decline and water bodies exhibited a positive trend in all the years. Moreover, results of present study also suggested that geo-spatial techniques have considerable potential for land use classification of a particular area. Additionally, the results of present study might be useful for developing and implementing valuable management strategies for resource utilization on sustainable basis.
... Sclerophyll Mediterranean shrubland is one of the most extensive vegetation types in Chile, covering 1,631,441 hectares (11.1% of the national territory) between the regions of Coquimbo and Bio Bio (INFOR, 2020). This ecosystem is of great importance in nutrient cycles, carbon sequestration, soil erosion control, biodiversity and endemic species conservation, and water regulation (Schulz et al., 2010;Ruiz-Peinado et al., 2017). Despite this great importance, it is beset by anthropogenic fires, over-grazing from livestock, agricultural expansion, and logging for firewood, which have occurred historically from the Euro-Chilean settlement, and as a result have reduced shrubland cover in central Chile (Schulz et al., 2011). ...
Article
Full-text available
Aim of the study: The aim of this study is to provide information on species-specific basic wood density (g cm-3) and moisture content (%) in Mediterranean shrublands. Area of study: The study covers two sites of the sclerophyllous shrubland in central Chile, Cortaderal (34°35’S 71°29’W) and Miraflores (34°08’S 70°37’W), characterized by different climatic and topographic conditions. Material and methods: The sampling area covers 4,000 m2 over four plots at two sites. Shrub species were identified and size-related attributes such as height and crown size measured. A total of 322 shrubs were sampled at 0.3 m aboveground to determine basic wood density and moisture content. Species-specific differences and similarities were analyzed by multiple pairwise comparisons (post-hoc tests) and by ordination and hierarchical clustering. Main results: We found high variation across species in wood density (0.46-0.77 g cm-3) and moisture content (41.6-113.1%), with many significant differences among species in wood density and among sites in moisture content. Because intraspecific variability could not be explained by shrub size and pronounced differences in wood density (0.49-0.64 g cm-3) also occurred between species of the same genus (e.g., Baccharis linearis and Baccharis macraei), our results suggested that phylogenetic affinity may be less important than adaptation to local conditions. Research highlights: The values presented here were variable according to the type of species and environmental conditions, necessitating the determination of basic wood density (BWD) and moisture content at site – and species-specific level. The provided BWD estimates allow converting green volume to aboveground biomass in shrubland areas and are an essential source of information for estimating the carbon stocks.
Article
Full-text available
The Mediterranean-type Ecosystems of Central Chile is one of the most threatened regions in South America by global change, particularly evidenced by the historical megadrought that has occurred in central Chile since 2010. The sclerophyllous forest stands out, whose history and relationship with drought conditions has been little studied. Cryptocarya alba and Beilschmiedia miersii (Lauraceae), two large endemic trees, represent an opportunity to analyze the incidence of intense droughts in the growth of sclerophyllous forests by analyzing their tree rings. Here, we considered > 400 trees from nineteen populations of C. alba and B. miersii growing across a latitudinal gradient (32°–35° S). To study the influence of local and large-scale climatic variability on tree growth, we first grouped the sites by species and explored the relationships between tree-growth patterns of C. alba and B. miersii with temperature, precipitation, and climate water deficit (CWD). Second, we performed Principal Component Analysis to detect common modes of variability and to explore relationships between growth patterns and their relationship to Palmer Drought Severity Index (PDSI), ENSO and SAM indices. We detected a breaking point as of 2002 at regional level, where a persistent and pronounced decrease in tree growth occurred, mainly influenced by the increase in CWD and the decrease in winter-spring rainfall. In addition, a positive (negative) relationship was showed between PC1 growth-PDSI and PC1 growth-ENSO (growth-SAM), that is, growth increases (decreases) in the same direction as PDSI and ENSO (SAM). Despite the fact that sclerophyllous populations are highly resistant to drought events, we suggest that the sclerophyllous populations studied here experienced a generalized growth decline, and possibly the natural dynamics of their forests have been altered, mainly due to the accumulating effects of the unprecedented drought since 2010. Graphical abstract
Article
Full-text available
Humans have fundamentally altered global patterns of biodiversity and ecosystem processes. Surprisingly, existing systems for representing these global patterns, including biome classifications, either ignore humans altogether or simplify human influence into, at most, four categories. Here, we present the first characterization of terrestrial biomes based on global patterns of sustained, direct human interaction with ecosystems. Eighteen "anthropogenic biomes" were identified through empirical analysis of global population, land use, and land cover. More than 75% of Earth's ice-free land showed evidence of alteration as a result of human residence and land use, with less than a quarter remaining as wildlands, supporting just 11% of terrestrial net primary production. Anthropogenic biomes offer a new way forward by acknowledging human influence on global ecosystems and moving us toward models and investigations of the terrestrial biosphere that integrate human and ecological systems.
Chapter
Full-text available
In mediterranean-climate regions (MCRs), historical and geographical as well as ecological approaches are needed to elucidate processes and patterns occurring at that rarely defined level of complexity called “landscape”. In the case of Chile, a glance at the past five centuries of history is particularly crucial to the understanding of the various impacts of landscape degradation. Furthermore, to aid in our attempt to combine both ecological and human geographical considerations, we will borrow the “three waves” paradigm of sociologist Alvin Toffler (1980). Toffler was, of course, dealing with all of human history, and at the full planetary scale. Here we will be zooming in on the so-called secano interior, or “interior drylands”, of the subhumid region of the Chilean MCR.
Article
Full-text available
Wildfires resulting from thunderstorms are common in some Mediterranean-climate regions, such as southern California, and have played an important role in the ecology and evolution of the flora. Mediterranean-climate regions are major centers for human population and thus anthropogenic impacts on fire regimes may have important consequences on these plant formations. However, changes in fire regimes may have different impacts on Mediterranean type-ecosystems depending on the capability of plants to respond to such perturbations. Therefore, we compare here fire regimes and vegetation responses of two Mediterranean-climate regions which differ in wildfire regimes and history of human occupation, the central zone of Chile (matorral) and the southern area of California in United States (chaparral). In Chile almost all fires result from anthropogenic activities, whereas lightning fires resulting from thunderstorms are frequent in California. In both regions fires are more frequent in summer, due to high accumulation of dry plant biomass for ignition. Humans have markedly increased fires frequency both in the matorral and chaparral, but extent of burned areas has remained unaltered, probably due to better fire suppression actions and a decline in the built-up of dry plant fuel associated to increased landscape fragmentation with less flammable agricultural and urban developments. As expected, post-fire plant regeneration responses differs between the matorral and chaparral due to differences in the importance of wildfires as a natural evolutionary force in the system. Plants from the chaparral show a broader range of post-fire regeneration responses than the matorral, from basal resprouting, to lignotuber resprouting, and to fire-stimulated germination and flowering with fire-specific clues such as heat shock, chemicals from smoke or charred wood. Plants from the matorral have some resprouting capabilities after fire, but these probably evolved from other environmental pressures, such as severe and long summer droughts, herbivory, and volcanism. Although both Mediterranean-type ecosystems have shown to be resilient to anthropogenic fires, increasing fire frequency may be an important factor that needs to be considered as it may result in strong negative effects on plant successional trends and on plant diversity.
Article
Conversion of land to grow crops, raise animals, obtain timber, and build cities is one of the foundations of human civilization. While land use provides these essential ecosystem goods, it alters a range of other ecosystem functions, such as the provisioning of freshwater, regulation of climate and biogeochemical cycles, and maintenance of soil fertility. It also alters habitat for biological diversity. Balancing the inherent trade-offs between satisfying immediate human needs and maintaining other ecosystem functions requires quantitative knowledge about ecosystem responses to land use. These responses vary according to the type of land-use change and the ecological setting, and have local, short-term as well as global, long-term effects. Land-use decisions ultimately weigh the need to satisfy human demands and the unintended ecosystem responses based on societal values, but ecological knowledge can provide a basis for assessing the trade-offs.
Article
The most important processes shaping communities are likely to leave observable imprints on their spatial structure. The net outcome of positive and negative interactions among plants has been proposed to vary inversely with environmental harshness. In the Mediterranean-type matorral of central Chile, slope aspect strongly influences mesoclimatic conditions: equatorial-facing slope are xeric and are dominated by spiny shrubs and cacti, while evergreen sclerophyllous trees dominate polar-facing slopes due to their more mesic conditions. Here we use spatial pattern analyses to infer on the interactions that shape community structure in this matorral. We sampled the vegetation on two polar- and equatorial-facing slopes and assessed the dominant interaction at each slope through: (1) a species co-occurrence index (C-score) for the entire community; (2) the frequencies of negative and positive associations between species pairs; and (3) the species richness at each slope. Our results indicated that negative spatial associations dominated plant assemblages on all slopes, but the importance of negative associations decreased from mesic (∼30%) to xeric slopes (∼16%). Species richness was higher on xeric than on mesic slopes. Although competition seems to be the main biotic force structuring communities on both mesic and xeric slopes, we suggest that the decrease in importance of negative interactions from mesic to xeric habitats could produce an increase in diversity.