Rapid deforestation and fragmentation of Chilean Temperate Forests Cristian Echeverriaa,b,*,1, David Coomesa, Javier Salasc, Jose �� Mar��a �� Rey-Benayasd, Antonio Larab, Adrian Newtone a Department of Plant Sciences, University of Cambridge, Cambridge CB2 3EA, United Kingdom b Instituto de Silvicultura, Universidad Austral de Chile, Casilla 567, Valdivia, Chile cDepartamento de Geograf�� ��a, Universidad de Alcala, �� C/Colegios, 2, 28801 Alcala �� de Henares, Spain d Departamento de Ecolog�� ��a, Edificio de Ciencias, Universidad de Alcala, �� 28871 Alcala �� de Henares, Spain eSchool of Conservation Sciences, Bournemouth University, Talbot Campus, Poole, Dorset BH12 5BB, United Kingdom A R T I C L E I N F O Article history: Received 11 September 2004 Received in revised form 30 December 2005 Accepted 12 January 2006 Available online 10 March 2006 Keywords: Chile Forest fragmentation Deforestation Landscape indices Nothofagus Temperate forests A B S T R A C T The temperate forests of Chile are classified a biological ������hotspot������ as a result of their high species diversity and high endemism. However, they are being rapidly destroyed, with sig- nificant negative impacts on biodiversity. Three land-cover maps were derived from satel- lite imagery acquired over 25 years (1975, 1990 and 2000), and were used to assess the patterns of deforestation and forest fragmentation in the coastal range of south-central Chile. Between 1975 and 2000, there was a reduction in natural forest area of 67% in the study area, which is equivalent to an annual forest loss rate of 4.5% per year using a com- pound-interest-rate formula. Forest fragmentation was associated with a decrease in forest patch size, which was associated with a rapid increase in the density of small patches (100 ha), and a decrease in area of interior forest and in connectivity among patches. Since the 1970s, native forest loss was largely caused by an expansion of commercial plantations, which was associated with substantial changes in the spatial configuration of the native forests. By 2000, most native forest fragments were surrounded by highly connected exo- tic-species plantations. The assessment of forest loss and fragmentation provides a basis for future research on the impacts of forest fragmentation on the different component of biodiversity. Conservation strategies and land use planning of the study area should con- sider the spatial configuration pattern of native forest fragments and how this pattern changes over time and space at landscape level. �� 2006 Elsevier Ltd. All rights reserved. 1. Introduction Habitat fragmentation and forest loss have been recognized as a major threat to ecosystems worldwide (Armenteras et al., 2003 Dale and Pearson, 1997 Iida and Nakashizuka, 1995 Noss, 2001). These two processes may have negative ef- fects on biodiversity, by increasing isolation of habitats (Debinski and Holt, 2000), endangering species, and modify- ing species��� population dynamics (Watson et al., 2004). Frag- mentation may also have negative effects on species 0006-3207/$ - see front matter �� 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocon.2006.01.017 * Corresponding author: Tel.: +44 1223 764736. E-mail addresses: ce227@cam.ac.uk, cechever@uach.cl (C. Echeverria). 1 Tel.: +56 63 293221 fax: +56 63 293418/221230. B I O L O G I C A L C O N S E R V A T I O N 1 3 0 ( 2 0 0 6 ) 4 8 1 ��� 4 9 4 available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/biocon

richness by reducing the probability of successful dispersal and establishment (Gigord et al., 1999) as well as by reducing the capacity of a patch of habitat to sustain a resident popu- lation (Iida and Nakashizuka, 1995). For example, fragmenta- tion of Maulino temperate forest in central Chile has affected the abundance of bird richness (Vergara and Simonetti, 2004) and regeneration of shade-tolerant species (Bustamante and Castor, 1998), and has also favoured the invasion of alien spe- cies (Bustamante et al., 2003). The ecological consequences of fragmentation can differ depending on the pattern or spatial configuration imposed on a landscape and how this varies both temporally and spatially (Armenteras et al., 2003 Ite and Adams, 1998). Some studies have shown that the spatial configuration of the landscape and community structure may significantly affect species richness at different scales (Steiner and Kohler, �� 2003). Other authors emphasise the need to incor- porate the spatial configuration and connectivity attributes at a landscape level in order to protect the ecological integrity of species assemblages (Herrmann et al., 2005 Piessens et al., 2005). The temporal evaluation of forest change based on satel- lite imagery linked to fragmentation analysis is becoming a valuable set of techniques for assessing the degree of threat to ecosystems (Armenteras et al., 2003 Franklin, 2001 Imber- non and Branthomme, 2001 Luque, 2000 Sader et al., 2001). A number of deforestation studies have been conducted in trop- ical forests (Imbernon and Branthomme, 2001 Sader et al., 2001 Skole and Tucker, 1993 Steininger et al., 2001 Turner and Corlett, 1996) and, in particular, in the Amazon, which is now considered as the most studied region in the world by some researchers (Jorge and Garcia, 1997 Laurance, 1999 Laurance et al., 2000 Pedlowski et al., 1997 Ranta et al., 1998 Sierra, 2000). Conversely, few studies of deforestation and fragmentation have been reported for temperate forests (Staus et al., 2002), particularly in the southern hemisphere. Chile has the largest temperate forest area in South Amer- ica and more than half of the total area of temperate forests in the southern hemisphere (Donoso, 1993). Most of these for- ests are distributed along the Coastal and the Andean Range of Chile from 35�� to 56�� S and extend to a total of 13.4 million ha in the country (Conaf et al., 1999a). The temperate forest of Chile has been classified a biodiversity hotspot for conserva- tion (Myers et al., 2000) and has also been included among the most threatened eco-regions in the world in the Global 200 initiative launched by WWF and the World Bank (Diner- stein et al., 1995). In these forests, a 34% of the plant genera are endemic (90% monotypic) (Armesto et al., 1997). However, Chile���s temperate forests are being harvested to supply the increasing global demand for wood and paper products. A substantial amount of forest has also been lost due to the conversion of native forests to pasturelands, human-set fires, high grading (selective felling) and other logging practices (Lara et al., 2000). Although some of the ecological consequences of forest fragmentation have been studied in Chile (Bustamante and Grez, 1995 Donoso et al., 2003 Vergara and Simonetti, 2004 Willson et al., 1994), integrated spatial and temporal analyses have not been conducted. Although some attempts have been made in Chile to estimate the rate of deforestation (Lara et al., 1989 Olivares, 2000) or to assess land cover change (Conaf et al., 1999b Sandoval, 2001), these have been undertaken at local scales, over short time intervals (no more than 10 years), or using different types of data to compare forest cover over time, which confers some methodological limitations. Long- term analyses of the spatial patterns of deforestation and fragmentation of temperate forest ecosystems at the land- scape scale have not yet been reported either in Chile or else- where in the southern hemisphere. The purpose of this study is to contribute to the under- standing of the patterns of deforestation and fragmentation in the temperate forests of Chile at the landscape level. In par- ticular, we examined the patterns of land-cover change and the changes in the spatial configuration in the Maulino tem- perate forest over time and space by using satellite scenes ac- quired at different time intervals. In this study, we hypothesised that there has been a substantial loss of native forest as a result of an increase in area of exotic-species plan- tations. Also, we anticipated that this forest loss was related to fragmentation of Maulino forests due to changes in the spatial configuration in terms of size, shape and degree of iso- lation of forest patches. This work is the first step to under- stand the potential ecological effects of fragmentation and the proximate drivers and causes of deforestation, which will be addressed in other studies. 2. Study area The Rio Maule-Cobquecura study area covers approximately 578,164 ha of land located in the Coastal Range of the Maule and Bio-Bio regions in south-central Chile (Fig. 1). The studied landscape partially includes the municipalities of Constit- ucion, �� San Javier, Quirihue and Cobquecua, and all the area of the municipalities Empedrado, Chanco, Pelluhue, and Cauquenes. The area is characterised by rainfall concentrated during the winter that leads to dry summers from September to April with little cloud cover and high luminosity. The nat- ural forest is mainly dominated by secondary forest of Nothof- agus species (N. obliqua and N. glauca) (Fagaceae) and Fig. 1 ��� Location of study area in the coastal range, between Rio Maule and Cobquecura. 482 B I O L O G I C A L C O N S E R V A T I O N 1 3 0 ( 2 0 0 6 ) 4 8 1 ��� 4 9 4