ARTICLE IN PRESS Biological Conservation xxx (2009) xxx–xxx

Contents lists available at ScienceDirect

Biological Conservation journal homepage: www.elsevier.com/locate/biocon

Identifying and mapping biodiversity processes for conservation planning in islands: A case study in Réunion Island (Western Indian Ocean) Erwann Lagabrielle a,b,*, Mathieu Rouget c,d, Karine Payet a,d,e, Nokuthula Wistebaar a,d, Laurent Durieux f, Stéphane Baret f,g,h, Amanda Lombard b, Dominique Strasberg a a

UMR PVBMT, Université de la Réunion, Cirad, 7 chemin de l’Irat, Ligne paradis, 97410 Saint-Pierre, France Botany Department, Nelson Mandela Metropolitan University, Saasveld Campus, Private Bag X6531, George 6530, South Africa c Conservation Science Group, Zoology Department, University of Cambridge, Downing Street, CB2 3EJ, United Kingdom d Biodiversity Planning Unit, South African National Biodiversity Institute, Private Bag X101, Pretoria 0001, South Africa e Department of Conservation Ecology & Entomology, Faculty of AgriSciences, Stellenbosch University, Private Bag x1, Matieland, 7602, South Africa f Institute for Research and Development (IRD), Unité Espace S140, IRD Brésil, CP 7091 – Lago Sul, 71619 – 970 Brasilia (DF), Brazil g Parc national de La Réunion, 112 rue Sainte Marie, 97400 Saint-Denis, France h National Botanical Conservatory of Mascarin, 2 rue du Père Georges, Les Colimaçons, 97436 Saint-Leu, France b

a r t i c l e

i n f o

Article history: Received 2 January 2008 Received in revised form 1 December 2008 Accepted 15 February 2009 Available online xxxx Keywords: Integrated conservation planning Biological processes Island biogeography Corridors Land use modelling Socio-economic cost

a b s t r a c t Over the last century, island biodiversity has become one of the most threatened in the world. Although many island conservation plans address biodiversity requirements at the species level, few plans address the spatial requirements of the biodiversity processes that underpin the persistence of these species. Using systematic conservation planning principles, we map the spatial components of biodiversity processes (SCBPs) and use these to design broad-scale conservation corridors for Réunion Island. Our method is based upon a literature review, expert knowledge, spatially explicit base data, conservation planning software, and spatial modelling. We combine a target-driven algorithm with least-cost path analyses to delineate optimal corridors for capturing key biodiversity processes while simultaneously considering biodiversity pattern targets, conservation opportunities, and future threats. We identify five SCBPs: the oceanic–terrestrial interface; riverine corridors; macrohabitat interfaces; the boundaries of isolated topographic units; and lowland–upland gradients. A large proportion of the SCBPs (81.3%) is currently untransformed, whereas 3% is irreversibly transformed by urbanisation and 15.7% is transformed but restorable. However, SCBPs are almost fully disrupted by urbanisation in the lowlands, thereby compromising functional corridors along full altitudinal gradients. This study is a contribution toward the reconciliation of conservation versus development objectives on Réunion Island but we believe that the delineation method is sufficiently general to be applied to other islands. Our results highlight the need for integrating marine, coastal and terrestrial conservation planning as a matter of urgency, given the rapid transformation of coastal areas on islands. Ó 2009 Elsevier Ltd. All rights reserved.

1. Introduction 1.1. The status of island biodiversity Biodiversity in oceanic islands is particularly rich in endemic species, and contributes disproportionately high per unit area to global biodiversity (Stattersfield and Capper, 2000). During the last century, island biodiversity has become one of the most restricted * Corresponding author. Address: Botany Department, Nelson Mandela Metropolitan University, Private Bag X6531, George 6530, South Africa. Tel.: +27 (0) 44 343 1856; fax: +27 (0) 86 614 0137. E-mail addresses: [email protected] (E. Lagabrielle), rouget@ sanbi.org (M. Rouget), [email protected] (K. Payet), [email protected] (N. Wistebaar), [email protected] (L. Durieux), [email protected] (S. Baret), [email protected] (A. Lombard).

and threatened in the world (Mueller-Dombois and Loope, 1990). For instance, more than 60% of documented vertebrate extinctions have occurred on islands (Case et al., 1992; Diamond, 1989). Such an unprecedented rate of extinction has generally been attributed to three major reasons: 1. The small size and isolation of islands reduces spatial options for in situ persistence of biodiversity patterns and processes (Whittaker et al., 2001). Consequently, islands are structurally more vulnerable, especially to global climate change (Pelling and Uitto, 2001). 2. Ecosystem deterioration induced by anthropogenic pressures such as land use dynamics, fire and over-harvesting has recently become more intense and less controlled than in continental areas (Lane, 2006; Cudihhy and Stone, 1990).

0006-3207/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocon.2009.02.022

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS 2

E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

3. Historically isolated biodiversity patterns are intrinsically more vulnerable, particularly to biological invasions, because they evolved without large herbivores or taxonomically diverse competitors (Komdeur and Pels, 2005; Fritts and Rodda, 1998; Mueller-Dombois and Loope, 1990). There is thus an urgency to conserve biodiversity patterns and processes in island systems. More than 70% of the 34 hotspots listed by Conservation International (Mittermeier et al., 2005) are islands where biodiversity resources are a key factor in shaping sustainable development strategies. Yet, very few islands have undertaken systematic conservation planning which seeks to identify spatial options for representing and maintaining all the biodiversity within a region (Margules and Pressey, 2000; Balmford, 2003). Conservation plans initiated in island systems have often focused on pattern representation only, ignoring key biodiversity processes (Moritz, 2002; Cowling and Pressey, 2003). 1.2. The status of biodiversity on Réunion Island Réunion (2512 km2) is a recent tropical volcanic island (2–3 millions years BP) located to the east of Madagascar (780 km), close to Mauritius (210 km) in the Western Indian Ocean (Fig. 1). Its steep relief reaches 3069 m in the centre and 2631 m in the southeast (an active volcano). Mean annual temperatures decrease from 24 °C in the lowlands to 12 °C at ca 2000 m. Mean annual precipitation ranges from 3 m on the eastern windward coast, up to 8 m in the mountains and down to 1 m along the south western coast. Vegetation is most clearly structured along gradients of altitude and rainfall (Cadet, 1980). Based on previous studies, expert knowledge and aerial photography, Strasberg et al. (2005) mapped 20 native broad habitat units belonging to four macrohabitat units (coastal-lowland, submountain, mountain and sub-alpine habitats). As with other islands, biodiversity in Réunion is facing escalating threats that have led to the extinction of 30 of 45 vertebrate

species (Mourer-Chauvire et al., 1999; Cheke, 1987). Since European occupation of the island (in the year 1665), 73% of the native vegetation cover has been transformed by agriculture (36%), urbanisation or other land use (12%) and invasive species (25%) (Lagabrielle, 2007). Urbanisation pressure is extremely high in both cultivated and natural areas. More than 80% of the total population (7,80,000 inhabitants) lives on the coastal fringe of the island where most of the socio-economic activities are concentrated. The mean percentage of the annual increase in population was 1.9% from 1990 to 2000, and the population is predicted to reach 1,000,000 in 2030 (Actif and Lardoux, 2006). Lowland habitats are almost fully transformed, except on harsh slopes and ravines (Gigord et al., 1999). Clearing of native forest for cattle breeding in the uplands is also increasing. Habitat degradation by invasive alien species is currently the main threat to native biodiversity (Thébaud and Strasberg, 1991). In addition to 500 native vascular plants, more than 3500 plant species were introduced of which 62 are highly invasive (Baret et al., 2006; Lavergne, 1999; Macdonald et al., 1991), according to criteria defined by Richardson et al. (2000). A total of 42 vertebrate species is also naturalised at present but mainly in highly modified habitats (Simberloff, 1992). Since the recent creation of a National Park in 2007, conservation areas encompass more than 42.6% of the island but their spatial distribution remains largely biased toward the uplands. 1.3. Biodiversity patterns and processes While biodiversity patterns describe the spatial distribution of features (e.g., species, habitats) in a region, biodiversity processes are the dynamic interactions within and among ecosystems. These processes operate through both space (e.g., migration in the landscape), and time (e.g., speciation). Biodiversity processes therefore encompass the ecological and evolutionary processes that maintain, sustain and generate biodiversity within a region (Cowling

Fig. 1. Location of Réunion Island and the spatial distribution of urban and cultivated areas.

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

et al., 1999; Balmford et al., 1998). These processes can be biotic (e.g., seasonal migration of birds) or abiotic (e.g., nutrient cycles) whereas evolutionary processes rely on the genetic adaptation of biodiversity patterns to changing environments (Moritz, 2002). The disruption of key biodiversity processes in insular regions has led to catastrophic ‘avalanche extinction’ (Vitousek et al., 1995; Brook and Kikkawa, 1998). In particular, the extinction of species ensuring key ecological processes, such as pollination or seed dispersal, has a disastrous impact on insular ecosystems because few alternative solutions exist for ensuring the continuity of such processes (Olesen and Valido, 2003; Townsend, 1996). Biodiversity processes have been integrated into conservation planning protocols using various approaches. Often, a simple comparison is done between the requirements of a focal species and the spatial criteria of conservation areas (size, shape and connectivity) (Coppolillo et al., 2004). Rouget et al. (2003) stress that these criteria relate only partly to the demographic, genetic and evolutionary processes important for the persistence of biodiversity patterns. Different methods exist to assess the spatial requirements of biodiversity processes, including in situ short-term observations (e.g., bird movements in the landscape) and long-term surveys of individuals or populations (e.g., monitoring of bird migration) (Dmowski and Kozakiewicz, 1990), combined with meta-population modelling and spatial modelling (Fullera et al., 2006; Chefaoui et al., 2005). Environmental surrogates have been used to identify particular assemblages of current ecological factors and historical traits known (or assumed) to originate and maintain particular biodiversity processes (Ferrier, 2002). For example, broad habitat units, or BHUs (Cowling and Heijnis, 2001), known as good overall surrogates for biodiversity patterns (Lombard et al., 2003), can act as relevant surrogates for mapping biodiversity processes (Rouget et al., 2003). Areas deemed suitable for maintaining ecosystem processes are generally restricted to pristine or near-pristine vegetation, but they can also encompass extensive areas of cultural/suburban landscapes (Arendt, 2003). The latter case is where stewardship programmes between conservation planners and private landowners are important in contributing to the broad-scale connectedness of ecological systems. 1.4. Biodiversity patterns and processes on Réunion Island Réunion Island (2512 km2) is recognised as one of 34 global biodiversity hotspots (Mittermeier et al., 2005). As with other islands, several factors hinder the establishment of an adequate conservation area network for biodiversity processes, including the lack and the fragmentation of data and knowledge on these processes, the pressures from competing land uses (Lagabrielle, 2007) and the

absence of stakeholder commitment (Strasberg et al., 2005). Spatially explicit information on biodiversity processes is urgently required for developing and implementing a systematic conservation plan in Réunion Island. The method for identifying and mapping such processes on the Island was based on a literature review, expert knowledge and basic Geographic Information System (GIS) data. Below we discuss the five processes identified as important for Réunion Island (see Table 1). 1.4.1. The oceanic–terrestrial interface The oceanic–terrestrial interface is a key component of biodiversity processes in insular systems. On Réunion Island, this interface encompasses coastal habitats associated with cliffs, recent lava flows and sandy beaches, some of which are contiguous to coral reefs (Le Corre and Strasberg, 2002). The coastal fringe supports specific ecological processes that are vital for the persistence of terrestrial and marine biodiversity, such as the feeding of marine birds nesting inland (Jaquemet et al., 2007; Le Corre and Safford, 2001). This interface is also a central component of species foundation and evolutionary processes in oceanic islands. Cadet (1980) estimated that, before their naturalization, 8% of native plant genera in Réunion were transported by oceanic currents from external landmasses. This rate is estimated to be about 50% for phanerogam genera coming from remote regions such as Australia and Asia (Cadet, 1980; Rivals, 1952). River mouths along the coast are also key components for anadromous migration of aquatic biota. 1.4.2. Riverine corridors Réunion comprises a dense network of 750 rivers (only 13 are perennial) that flow parallel to the slopes. Many rivers are surrounded by cliffs (up to 300 m), which form, together with the vegetation, a specific landscape unit named ‘ravine’. Riverine corridors support exchanges between lowlands and uplands, particularly top–down nutrient flows and bird movements (Le Corre and Safford, 2001). Gigord et al. (1999) demonstrated that ravines act as refugees for indigenous species in transformed landscapes. 1.4.3. Macrohabitat interfaces Macrohabitat interfaces are defined as the border between contiguous primary order BHUs. Such interfaces (or ecotones) are assumed to drive the ecological diversification of plant and animal lineages (Cadet, 1980). In Réunion, macrohabitat interfaces are spatially distributed as nested rings, from lowland to sub-alpine macrohabitats (Strasberg et al., 2005) (Fig. 2). 1.4.4. Isolated topographic units boundaries In Réunion, isolated topographic units are presumed to support the allopatric diversification of taxa (Warren et al., 2006). We distinguished three large isolated topographic units: namely, the cir-

Table 1 Spatial components of biodiversity processes in Réunion Island. Spatial component

Process

Delineation method

Spatially fixed Oceanic–terrestrial interface Riverine corridors

Fauna movement between oceanic and terrestrial domain Settlement of new species Migration and exchange between lowland and upland

400 m wide buffer expanded inland along the coastline

Macrohabitat interfaces Isolated topographic units boundaries Spatially flexible Lowland–upland gradient

Ecological diversification of plant lineages Ecological diversification of plant lineages

Ecological diversification of plant lineages (radiative speciation) and fauna movement

3

Weighted buffer of 50, 100, 150 and 200 m wide along perennial rivers 50 m wide buffer along non-perennial rivers Buffer 200 m wide along primary macrohabitat interfaces Buffer 200 m wide along primary topographic boundaries

1 km wide stripe linking lowland areas to upland areas

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS 4

E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

Fig. 2. Current macrohabitat distribution in Réunion Island. White lines represent original macrohabitat ecotones. Data from Strasberg et al. (2005).

ques of Mafate, Cilaos and Salazie. Such topographic units were originated by structural movements (Gillot et al., 1994) and water erosion, more particularly during cyclonic events (Mairine and Bachèlery, 1997). 1.4.5. Lowland–upland gradients With altitudes ranging from 0 to 3070 m, lowland–upland gradients represent key drivers of radiative speciation and floral diversification in Réunion (Jacquemyn et al., 2005; Warren et al., 2006). This hypothesis is corroborated by genetic studies on the indigenous Psiadia genus (Besse et al., 2003). Lowland–upland gradients are important for indigenous fruit-feeding birds and insects, which follow asynchronous fructification and flowering of native plant species across a wide range of altitudes (Probst, personal communication). Such seasonal migrations of biota are also important for seed dispersal and pollination (Pailler and Micheneau, 2005). 1.5. Study objectives In this study we propose a protocol for the rapid mapping of the Island’s biodiversity processes described above. The overarching goal of our study is to inform land use planning decision-makers of the spatial requirements of biodiversity processes that underpin the ecological functioning of the Island’s biodiversity. More specifically, our objectives are: (1) To identify the biodiversity processes that sustain and generate biodiversity at a medium scale (1:100,000), which is the scale of regional land use planning in Réunion Island.

(2) To map the spatial components of these processes. (3) To assess their ecological status. (4) To delineate an optimal network of corridors able to contribute to safeguarding their persistence in a rapidly changing environment. 2. Methods 2.1. Base data 2.1.1. Broad habitat units BHUs act as broad-scale surrogates for biodiversity distribution as a whole and are particularly useful for conservation planning when data on biodiversity patterns or processes are poor or incomplete. For this study we used a BHU map comprised of 44 habitats, including pristine to transformed habitats, as well as a map of BHUs before human settlement (Lagabrielle et al., 2008; Strasberg et al., 2005). 2.1.2. Irreplaceability map Irreplaceability is a measure of the contribution of a given site to the achievement of conservation targets in a planning domain (Ferrier et al., 2000). Typically, targets are set for biodiversity patterns (e.g., 20% of the pre-transformation extent of all vegetation types, or 10 occurrences of a subset of species) as well as processes (e.g., the full extent of a riverine corridor) (see Rouget et al., 2003, 2006, and Lombard et al., 2007 for more examples). We used Marxan software (Ball and Possingham, 2000) to calculate irreplaceability values and to derive a map of irreplaceability for Réunion

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS 5

E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

Island. Irreplaceability values were calculated for planning units (hexagons, radius = 200 m) with targets of habitats ranging from 24% to 45% of their pre-transformation extent (Lagabrielle, 2007), representing a total of 676 km2. For instance, the target for the windward mountain rainforest was deemed to 116 km2, 35 km2 for the semi-dry forest BHUs and 7 km2 for the wetlands. 2.1.3. Transformation status Based on the BHU map, four transformation statuses were determined (Strasberg et al., 2005; Driver et al., 2003): extant; invaded; transformed restorable; and, irreversibly transformed (Table 2 and Fig. 3): ‘Extant’ relates to pristine habitats, including lightly invaded vegetation (Baret et al., 2006), ‘invaded’ contains moderately to highly invaded habitats and ‘transformed restorable’ relates to fully transformed but potentially restorable habitats (i.e. secondary vegetation, crops and forestry). Secondary vegetation refers to vegetated areas with no native species. Urban areas were considered ‘irreversibly transformed’. 2.1.4. Predictive models of threat occurrence The outcomes of three predictive models on the probable future distribution of urbanisation, agriculture and plant invasion were combined for this study (Table 3). Urbanisation probabilities were derived from non-linear regression analysis on 12 factors explaining urban sprawl observed from 1989 to 2002 (Thinon et al., 2007). The relationship between observed urban sprawl and population growth was then derived by a simple linear regression analysis. Urbanisation expansion was calibrated using population prediction for 2030. Potential areas for agriculture (sugar cane and pastures) were identified by experts (Chambre d’Agriculture de La Réunion, unpublished data). Potential extents of invasive plants were calculated by Baret et al. (2006) using Climatic Envelop Modelling. A map of invasion threat probability and density was derived by summing the potential extent of the 20 most invasive species. The occurrence probability of each threat varied from ‘not probable’ (score = 1) to ‘highly probable’ (score = 10). An index of threat was then derived by calculating a mean score among the three threats (urbanisation, agriculture and alien plants) in each spatial cell of the model (resolution = 25 m). For combining threat scores, the mean-score method was preferred to the highest-score method owing to the high intensity and density of threats in the planning domain (saturation effect). 2.1.5. GIS data on riverine systems, topography, land tenure and conservation areas GIS layers on rivers and the topography (digital elevation model, resolution = 25 m) were supplied by IGN BD Topo 1997. Data on

land tenure (public or private ownership) and conservation areas were provided by the Mission Parc National de La Réunion. 2.2. Delineating the spatial components of biodiversity processes (SCBPs) and conservation corridors 2.2.1. Stage 1: collating existing knowledge on SCBPs Numerous methods for identifying biodiversity processes can be found in the abundant literature on key biogeographic trends in tropical oceanic islands (Whittaker, 1998; Vitousek et al., 1995; Carlquist, 1974; Mac Arthur and Wilson, 1967). This literature, however, is largely biased toward charismatic vertebrate species, and not much is known about the spatial distribution of other key biodiversity processes. In order to gather expert information on key processes in Réunion Island, we interviewed five experts from the regional scientific committee of natural heritage (CSRPN). They were selected by the regional prefecture for their field experience and scientific background. Each interview was structured into four parts. First, we described our project objectives and the scale of the study (1:100,000). We asked the experts to describe their areas of expertise and to list biodiversity patterns that they considered conservation priorities. Second, we asked them to identify and describe biodiversity processes that maintain and generate these patterns. Third, we asked the experts to assess the sensitivity of biodiversity processes to anthropogenic threats (urbanisation, agriculture, alien invasive plants). Finally, we developed a conceptual scheme that represented the relationships between the patterns, the processes and the threats. Based on this scheme we developed specifications for mapping the spatial surrogates of biodiversity processes. 2.2.2. Stage 2: mapping the SCBPs Rouget et al. (2003) define SCBPs as: ‘‘the physical feature of a region with which particular ecological and evolutionary processes are associated”. SCBPs on Réunion Island were mapped as surface elements aligned along linear environmental interfaces or gradients. For all analyses, we used ArcGIS version 9.2 (ESRI, 2006) and its extension Spatial Analyst. Two types of SCBPs were distinguished: (i) spatially fixed SCBPs associated with clearly identified physical features and (ii) spatially flexible SCBPs where several options exist for their spatial allocation (Rouget et al., 2003). In further steps of the analyses, all SCBPs were considered of equal importance and were given an equal weight. We delineated the four fixed SCBPs previously described (i.e. oceanic–terrestrial interfaces; riverine corridors; macrohabitat interfaces; and isolated topographic units boundaries) and one flexible SCBP (i.e. lowland–upland gradients) (Table 1). The four fixed SCBPs were

Table 2 Categories of habitat transformation in Réunion Island (adapted from Strasberg et al., 2005). Transformation status Pristine Extant Intact Lightly invaded Invaded restorable Moderately invaded Highly invaded Transformed Transformed restorable Secondary vegetation Cultivated Irreversibly transformed a

Description

Reference

Area (% of total)a

Not invaded or presence of some alien plant individuals in an intact canopy and under storey (alien species 90%) but under storey invaded (10–90%)

Strasberg et al. (2005) and Baret et al. (2006)

26.9 7.7

Strasberg et al. (2005) and Baret et al. (2006)

19.3

Canopy and under storey invaded (native species cover between 50% and 90% in the canopy) Canopy and under storey invaded (native species cover between 10% and 50% in the canopy)

Strasberg et al. (2005) and Baret et al. (2006)

25.3 12.8

Strasberg et al. (2005) and Baret et al. (2006)

12.5

No native species Crops including forestry Urban areas

Lagabrielle (2007) Lagabrielle (2007) AGORAH (2002)

36.3 17.7 18.6 9.9

Transformation status of 1.6% of the area of the island remains unknown.

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS 6

E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

Fig. 3. Habitat transformation in Réunion Island (see Table 2 for sub-categories). Table 3 Description of models used for predicting threat occurrence probability. Model

Model type

Model elaboration

Temporal calibration

Urban expansion

Probabilistic

Agriculture expansion Invasive plant expansion

Multi-criteria analysis Probabilistic

Multiple non-linear regression analysis on 12 factors explaining observed urban sprawl from 1989 to 2002 Multi-criteria analysis driven by agriculture experts

Calibrated using population prediction for 2030 Implicitly calibrated to predict agriculture distribution in 2030 No calibration (potential distribution of all invasive species)

Sum of potential distribution of 20 of the most invasive species present in Réunion (climatic envelop modelling)

delineated using the map of BHUs before human settlement, and the transformation status of SCBPs was estimated using the current BHU map. For oceanic–terrestrial interfaces we used a buffer of 400 m stretching along the coastline to delineate the interface area. The oceanic side of the interface was considered as extant. For riverine corridors, we selected the perennial rivers (n = 13), their secondary order tributary streams and the first order nonperennial rivers (n = 61). Non-perennial rivers encompass seasonal, episodic and ephemeral rivers (sensu Roux et al., 2002). Riverine corridors were delineated using a 200 m wide buffer on each side of the perennial rivers and a 50 m wide buffer on each side of the selected non-perennial rivers. Each riverine corridor was divided into adjoining sections. The sections of the perennial rivers were subdivided into eight parallel compartments associated with a weighting factor ranked from one, at the corridor border, to four, at the core of the corridor (Fig. 4). This technical refinement made it possible to calculate different estimates of the impact of land transformation, depending on the distance from the river. The transformation status was estimated per segment, per section and per whole riverine corridor. Secondary vegetation was considered suitable for biodiversity processes in riverine corridors. Macrohabitat interfaces were delineated with a 100 m wide buffer on each side of the interface line to capture the evolutionary processes previously described.

For Isolated topographic unit boundaries, the conservation of the boundaries aims to ensure the unit’s isolation. We used a digital elevation model and a 200 m wide buffer to delineate the area of this SCBP. Spatial components of lowland–upland gradients were mapped in two stages. First, we identified their sources in the uplands and destination points in the lowlands. The source points (n = 3) were located on the three highest points of the island (Fig. 1). The destinations points (n = 24) were distributed at regular intervals along the coastline, preferably in untransformed areas, river mouths and protected areas (marine and terrestrial). We also placed destination points between summits. Second, the least-cost paths linking the source to the destination points were calculated across a cost matrix coded with costs associated with transformation status, protection status, land tenure, distance to border inside pristine habitat, and threat occurrence probability (Table 4). A 500 m wide buffer zone was then created along each least-cost (optimal) path. We delineated 25 lowland–upland gradients. 2.2.3. Stage 3: Quantifying the habitat transformation of SCBPs Each SCBP was divided into sections, which acted as discrete spatial units for assessing their habitat transformation (ecological status). Each section was 500 m long and the width ranged from 100 m to 1 km, depending on the SCBP (Table 1). Widths of sections were determined by expert knowledge. For each section we

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS 7

E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

Fig. 4. Method for estimating the transformation status of the perennial riverine corridors. Each riverine corridor is divided into adjacent sections (a) and each section is further divided into eight parallel compartments, each with a weighting factor (b). The transformation status per section is calculated by aggregating weighted percentage of area transformed per compartment. The same method is applied to the sections when calculating the transformation status of the whole riverine corridor.

Table 4 Spatial parameters used to design lowland–upland paths, and corridor expansion. The minimum and the maximum value of each cost component are indicated. Cost component

Associated variables

Values (min, max)

Transformation status Irreplaceabilitya

5 categories of habitat transformation Irreplaceability values calculated per planning unit with conservation targets ranking from 24% to 45% of original habitat extant Presence/absence of SCBP Presence/absence of protection Public/private owned area 6 classes of distance (outside – 0 m, 0–100 m, 100–200 m, 200–300m, 300– 400 m, >400 m) Probability of occurrence (see Table 3)

0–5000 0–4000

Fixed SCBPsa Protection Land tenure Distance to border inside pristine habitats Threatsb a b

0–1000 0–250 0–150 0–125 0–100

Cost component used only for designing corridors. Threats encompass urbanisation, agriculture, and invasive species.

calculated the percentage of transformed habitat (urban, cultivated, secondary vegetation. Other habitats were assumed suitable for processes. Transformed sections (i.e. area transformed >50%) were deemed restorable, except for those where urbanisation covered more than 50% of the area. 2.2.4. Stage 4: Optimising the design of conservation corridors Broad-scale conservation corridors aim to guarantee the persistence and integrity of biodiversity processes. The aim of step four was to delineate the best options for corridors that would encompass as much as possible of both the biodiversity patterns, and the five SCBPs described above, within a minimal area, and avoiding areas in the landscape where conservation options would be difficult to implement. Corridors were thus designed along altitudinal gradients, as buffers around the lowland–upland gradient SCBPs which formed the core areas. The lateral expansion of the lowland–upland SCBPs into the wider corridors was then controlled by both ecological and socio-economic constraints expressed in a cost matrix, using a method adapted from Rouget et al. (2006), where the source of the expansion is the lowland–upland SCBP it-

self. The cost matrix used was similar to the one used for designing the lowland–upland SCBPs (Table 4), but also included costs associated with irreplaceability values and the presence of fixed SCBPs in order to preferentially capture these areas within corridors. The spatial expansion was achieved when an arbitrary total cost threshold of 2 million was reached (corridors network A). This was compared with a configuration obtained for a threshold of 4 million (corridors network B), and allowed us to evaluate the ‘‘friction” of the landscape to the expansion algorithm. 3. Results 3.1. Fixed and flexible SCBPs SCBPs (including fixed and flexible) initially covered 905 km2 in Réunion Island, which represented 36% of the island area (Table 5, Fig. 5). Currently, a large proportion of SCBPs (81.3%) is still suitable for biodiversity processes, whereas 3% is irreversibly transformed by urbanisation and 15.7% is transformed but remains restorable, mainly in the lowlands. The overlap among fixed SCBPs

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS 8

E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

Table 5 Ecological status of fixed and flexible SCBPs. Extant sections are untransformed by urbanisation and agriculture (including forestry of exotic species). Restorable areas are transformed by agriculture and secondary vegetation. Lost sections are irreversibly transformed by urbanisation. In the case of riverine corridors, secondary vegetation was considered extant. SCBP

Extant (%)

Restorable (%)

Lost (%)

Length (km)

Surface (km2)

Spatially fixed Oceanic–terrestrial interface (n = 1) Riverine corridors (perennial) (n = 16) Riverine corridors (non-perennial) (n = 52) Macrohabitat interfaces (n = 14) Isolated topographic units boundaries (n = 9) Spatially flexible Lowland–upland gradient Fixed + flexible

78.6 12.0 94.4 95.9 80.6 100.0

17 58.4 2.2 2.8 18.6 0.0

4.4 29.6 3.4 1.4 0.9 0.0

– 256 355 570 1218 81

700 83 123 54 441 65

87.7 81.3

10.3 15.7

2.0 3.0

451 –

423 905

Fig. 5. Transformation status of spatially fixed components of biodiversity processes in Réunion Island.

is about 9.4% (66 km2) and increases to 24.1% between fixed and flexible SCBPs. Oceanic–terrestrial interfaces cover 83 km2, of which 58.4% is now transformed but restorable and 29.6% is irreversibly transformed. Consequently, the persistence of coastal biodiversity patterns and processes is highly compromised. Only 5.6% of the perennial riverine corridors and 4.2% of non-perennial corridors is considered irreversibly transformed (analysis per riverine section). Their relatively low level of transformation is owed to their steep morphology at higher altitudes, which prevents urbanisation and agriculture. However, the rate of land transformation increases to 16.6% for sections of perennial rivers located below 100 m. Of concern is the fact that, of 15

perennial rivers, only five remain undisrupted by irreversible habitat transformation. Macrohabitat interfaces were moderately transformed (18.6%), although interfaces between lowland macrohabitats were more highly transformed (24.7%) and almost fully transformed on the leeward side of the island. Fortunately, the spatial components sustaining processes associated with the isolated topographic units are still intact, mostly because they are located in a harsh environment that is unsuitable for human activities. The spatial components of biodiversity processes associated with the lowland–upland gradients covered 427 km2, which represented 46.5% of the total surface of these SCBPs. The area covered by the lowland–upland gradients is less transformed than fixed

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS 9

E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

Fig. 6. Percentage of area per categories of transformation status for processes (i) within the lowland–upland gradients SCBP (in grey) and (ii) for the whole planning domain (in black). This graph shows the gain of the least-path analysis that preferentially incorporated land suitable for processes. For instance, the lowland– upland gradient SCBP integrates 87.3% of Extant land (e.g., suitable for processes) while the whole landscape comprised only 70.8% of land belonging to this category.

SCBPs: 87.7% were estimated to be untransformed, whereas only 2% and 10.3%, respectively are estimated to be lost or restorable. Nonetheless, this assessment remains largely biased by the delineation method, which favoured the selection of untransformed areas. We used comparable evaluation units (500  500 m) to assess the quantitative gain provided by the least-cost path analysis. We estimated that 70.8% of the surface of Réunion Island was suitable for ecological processes, which is 19.3% less than lowland–upland SCBPs (Fig. 6). 3.2. Conservation corridors Conservation corridors covered 29.8% of the planning region (909 km2). They entirely encompass flexible SCBPs and 51.1% of the fixed SCBPs (Table 6 and Fig. 7). The lower contribution of the corridors to the conservation of fixed SCBPs is mainly because of the perpendicular distribution of macrohabitat interfaces. The corridors added 2.6% to the achievement of the conservation targets for the vegetation types (88.8% being already achieved by extant protected areas). Because it was impossible to avoid transformed areas when designing the corridors, 25.6% of the corridor area included transformed habitats (irreversibly transformed or restorable) whereas only 3.8% of the protected areas are transformed. As a consequence, threats to biodiversity are higher in the corridors (mean threat score = 3.6) than in the protected areas (mean threat score = 2.5), even if threatening agents differ (landuse in transformed areas, and plant invasions in protected areas).

A large proportion of corridors is currently unprotected (33.5%). This represents a priority for conservation in Réunion of which 79.7% is privately owned and 57.0% contains secondary vegetation or cultivated areas. Conservation options for privately owned areas include land care or stewardship alternatives. Corridor areas located outside protected areas are highly vulnerable to threats from land use change and invasive species (mean threat score = 5.9). Doubling the cost threshold that stops the corridor expansion process (from 2 million for corridors network A to 4 million for corridors network B, see Table 6) vastly increased the total area of the corridors (from 30% to over 40% of the planning region) without significantly increasing the representation of fixed SCBPs or biodiversity pattern targets. In fact, inside corridors network B the level of target achievement for the biodiversity patterns was barely increased (+0.3%). Thus, with only limited conservation gains, the corridors network B option includes a higher proportion of privately owned areas, unprotected areas and areas vulnerable to land use pressures and is not a sensible option. 4. Discussion 4.1. Biodiversity processes assessment We identified five spatial components supporting key biodiversity processes in Réunion Island: the oceanic–terrestrial interface, riverine corridors, macrohabitat interfaces, isolated topographic unit boundaries, and lowland–upland gradients. The spatial density of these SCBPs in Réunion Island is very high compared with continental regions (Rouget et al., 2003), and fortunately the SCBPs are in general extant (81.3%), although their persistence is highly threatened in the lowlands by urbanisation and agricultural expansion. Although this assessment is highly dependent on the delineation methods used for the SCBPs, and on the assumptions that these areas support key biodiversity processes, the rapid rate of ecological deterioration and high threats faced by native biodiversity on Réunion justify immediate conservation action, using the best data and methods currently available. 4.2. Conservation corridor design Our approach for designing conservation corridors was based on systematic conservation planning principles (Margules and Pressey, 2000) and differs from previous studies based on expert judgements or key biodiversity areas only (Conservation International, 2000; Knight et al., 2007). We managed to incorporate potentially diverging conservation goals by optimising corridor de-

Table 6 Adequacy of broad-scale corridors in terms of extant habitat, fixed and flexible processes, target achievement for vegetation types, protected and public areas and threat exposure. Indicators are evaluated for protected areas, whole planning region, whole conservation corridors (cumulated cost = 2 or 4 millions units) and unprotected areas of corridors. Whole planning region (2512 km2) Area (% of planning region) Extant and moderately invaded area (% of area) Highly invaded area (% of area) Transformed restorable area (% of area) Irreversibly transformed area or unknown status (% of area) Flexible SCBP (%) Fixed SCBP (%) Pattern target achievement (average%)a Protected areas (% of area) Public domain (% of area) Threat (mean score) a

Protected areas

Corridors network A (cumulated cost = 2 million units)

Corridors network B (cumulated cost = 4 million units)

Unprotected area of corridors network A

100.0 39.7

42.6 86.8

29.8 61.8

42.0 59.6

10.0 76

12.5 36.3

9.3 3.6

11.7 20.8

11.9 22.9

21.5 57.0

11.5

0.2

4.8

5.7

14.3

100.0 100.0 –

72.0 70.1 88.8

100.0 51.4 91.4

100.0 63.7 91.7

12.7 41.0 91.4

42.8 45.6 4.7

100.0 89.5 2.5

66.5 67.7 3.6

63.8 64.9 3.7

0.0 20.3 5.9

The percentage of target achievement for the biodiversity features includes the contribution of protected areas plus areas of corridors outside protected areas.

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS 10

E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

Fig. 7. Corridor configuration and hypothetical cost factors for conservation planning.

sign in order to encompass both biodiversity patterns and processes, and to avoid constraints (vulnerable, privately owned and unprotected areas). The method is sufficiently general to be applied to other insular regions. However, we recommend testing the adaptability of the method to islands of greater or smaller size, with different spatial land use patterns, and with substrates of volcanic, coralline, sedimentary or metamorphic origin, and with different climatic regimes. The final configuration of the conservation corridors depends on the relative costs associated with constraint factors. Nevertheless, in Réunion, given the level of transformation and the spatial organisation of the landscape, the delineation options for corridors are very limited. The resulting corridors capture the entire extent of the altitudinal gradients identified as flexible processes, and include a significant amount of the other (spatially fixed) processes (51.4%) (Table 6). Together with existing conservation areas, the corridors achieve a large proportion of the conservation targets for vegetation types (91.4%). Moreover, conservation corridors integrate well with publicly owned areas (67.7% of the area of corridors) and existing protected areas (66.5%). However, areas of high land use pressure could not be avoided in the lowlands as a consequence of higher amounts and rates of habitat transformation. Furthermore, we showed that expanding the total area covered by corridors (using a cost threshold of 4 million) did not necessarily improve benefits for biodiversity conservation, but did require more land unsuitable for biodiversity conservation. Further studies should carefully assess the impacts of cost weights in the cost matrix. Such an analysis would consist of developing corridor designs based only on biological criteria. Conservation targets, manage-

ment and implementation considerations such as land cost would be ignored. This baseline analysis would allow assessing the efficiency of integrating socio-economic costs within the least-path analysis for designing corridors. The next step in the analysis would then be to combine the corridor design process with an economic assessment in order to base the least-cost path analysis on ‘‘real” economic costs. Interestingly, the incorporation of corridors into conservation planning strategies raises an ethical question: how does one plan for conservation with partial and biased knowledge of ecosystem functioning? The dilemma between scientific rigor, uncertainty and the urgency of conservation actions can be addressed by adopting the adaptive management strategy (Lee, 2002). This approach aims to balance the requirements of management with the need to learn about the system being managed, which theoretically leads to better decisions. This strategy should be adopted in Réunion Island, and more generally in insular regions, as available data and knowledge on biodiversity processes are often insufficient. Consequently, the shape of SCBPs and corridors should be modified as new data and insights emerge. Nevertheless, in small islands, the planning options for corridor design are so limited that they should all be explored. 4.3. Implementing conservation corridors Designing conservation corridors is one of the stages of a Systematic Conservation Plan (Margules and Pressey, 2000). In order to be effective, this stage, as well as other stages of Systematic Conservation Planning, must be coupled with a stakeholder

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

involvement and an implementation strategy (Knight et al., 2006a,b; Pierce et al., 2005), to avoid the study falling through the ‘‘knowing–doing gap” (Knight et al., 2008). Indeed, many studies have shown that the active participation of stakeholders is vital for the success of conservation (Brown, 2003) and land use policies (Castella et al., 2005). In such a framework, the results of our study should not be perceived as a single solution, but rather as a departure point for framing conservation and other land use actions. For instance, information produced by our study can be mobilised by conservationists and other stakeholders when negotiating a development plan, at different spatial scales (from regional to local), both inside and outside protected areas. In addition, our analyses allow one to assign ecological values to, and suggest alternative futures for, agricultural areas. In Reunion Island, the conservation corridors we designed could guide the implementation of the Free Membership Zone (zone de libre adhésion) that surrounds the National Park. Municipalities are the members of this zone, which is ruled by a common management charter. Membership is renegotiated every 10 years. In return, municipalities must develop and implement sustainable land use plans, compatible with conservation objectives. More generally, the implementation of the conservation corridors in Reunion Island is a major challenge owing to the spatial organisation of the island in parallel belts (urban, agricultural and natural belts). The implementation of corridors would require the integration of management regimes, both administrative and sectoral, from the seashore to the summits. The implementation of corridor-friendly management measures within privately owned areas should be based on land care and stewardship programmes, financial incentives and capacity-building initiatives dedicated to private owners. Conservation corridors do, however, represent an opportunity for integrated and sustainable development of natural, rural and urban areas on the Island. Their implementation would thus act as a common geographic reference for the protection of landscapes that link the ocean to the uplands. Such corridors could provide linkages between the Terrestrial National Park and the National Marine Natural Reserve on the west coast (Fig. 7). 4.4. Managing and monitoring conservation corridors Activities and actions, to be developed within the conservation corridors, should be negotiated with, and managed by, with a wide group of stakeholders (Cowling, 2005). All current land use managers should be major stakeholders, but in addition to these, we suggest the formation of corridor management committees (CMC). These committees could potentially use the management models developed for integrated and participatory management of natural resources (Berkes and Folke, 1998). A CMC would thus be composed of elected officials and representatives of users including conservation managers, farmers and tourism professionals. The objectives of a CMC would be (i) to develop corridor management plans in line with ecological and social requirements; (ii) to identify and to propose mechanisms to integrate corridor management plans within current management structures; (iii) to monitor the implementation, management and the impacts of the plans; and (iv) to review and adapt the management plans by learning from the monitoring outcomes. The management of conservation corridors will require the development of monitoring programmes that assess the ecological functions and integrity of the corridors, and the impacts of management actions. Results of these monitoring programmes should then inform the development and revision of land use plans, both within the corridors and the free membership zone of the National Park. Finally, the CMC should aim to be a learning institution (Knight et al., 2006a), able to adapt the management of the corri-

11

dors by drawing upon a permanently updated knowledge base of the interactions among social and ecological systems, within and around the corridors. 5. Conclusion Oceanic islands are highly dynamic and vulnerable eco- and socio-systems where reconciling conservation and sustainable development is not a trivial task owing to the scarcity of planning options. We have proposed a systematic protocol for identifying and protecting biodiversity processes through a network of conservation corridors encompassing both terrestrial and coastal components, and have integrated socio-economic sustainability constraints into the design process. Islands require urgent conservation actions, and future research should focus on biological processes along terrestrial–coastal–marine interfaces and on methods for integrating these and other biodiversity processes into land and resource management plans. Acknowledgements This research was funded by the Regional council of Réunion Island, the European Development Fund FEDER (APIC-bio project) and the ANR BIOTAS. It involved the following institutions: CIRAD, University of Reunion Island, South African National Biodiversity Institute, Mission Parc National de la Reunion and the Nelson Mandela Metropolitan University. We thank Lucien Tron and Jean-Cyrille Notter (Mission Parc National de la Reunion), Vincent Boullet (Conservatoire Botanique National de Mascarin), Jean-Michel Probst (Association Nature et Patrimoine), Marc Salamolard (Société d’Etudes Ornithologiques de La Réunion) and Mathieu Le Corre (Université de La Réunion) for their excellent advice. We are grateful to two anonymous reviewers for their constructive comments on a previous version of this manuscript. References Actif, N., Lardoux, J.-M., 2006. Six scénarios pour répartir la population de 2030 in Dossier: Projections 2030. Economie de La Réunion 125, 12–15. AGORAH, 2002. Tache urbaine de juillet 2002. Délimitation manuelle à partir d0 une image spot 5 de juillet 2002. Arendt, R., 2003. Linked landscapes: creating greenway corridors through conservation subdivision design strategies in the northeastern and central United States. Landscape and Urban Planning 68 (2–3), 241–269. Ball, I.R., Possingham, H.P., 2000. MARXAN (V1.8.2): Marine Reserve Design Using Spatially Explicit Annealing: A Manual. Balmford, 2003. Conservation planning in the real world: South Africa is showing the way. Trends in Ecology and Evolution 18 (9), 435–438. Balmford, A., Mace, G., Ginsberg, J.R., 1998. The challenges to conservation in a changing world: putting processes on the map. In: Mace, G., Balmford, A., Ginsberg, J.R. (Eds.), Conservation in a Changing World. Cambridge University Press, Cambridge, UK, pp. 1–28. Baret, S., Rouget, M., Richardson, D.M., Lavergne, C., Egoh, D., Dupont, J., Strasberg, D., 2006. Current distribution and potential extent of the most invasive alien species on La Réunion (Indian Ocean, Mascarene Islands). Austral Ecology 31, 747–758. Berkes, F., Folke, C., 1998. Linking Social and Ecological Systems: Management Practices and Social Mechanisms for Building Resilience. Cambridge University Press, Cambridge, UK. Besse, P., Da Silva, D., Humeau, L., Govinden-Soulange, J., Gurib-Fakim, A., Kodja, H., 2003. A genetic diversity study of endangered Psiadia species endemic from Mauritius Island using PCR markers. Biochemical Systematics and Ecology 31, 1427–1445. Brook, B.W., Kikkawa, J., 1998. Examining threats faced by island birds: a population viability analysis on the Capricorn silvereye using long-term data. Journal of Applied Ecology 35 (4), 491–503. Brown, K., 2003. Tree challenges for a real people-centred conservation. Global Ecology & Biogeography 12, 89–92. Cadet, T., 1980. La végétation de l’île de La réunion, étude phytoécologique et phytosociologique. PhD Thesis, University of Aix Marseille. Carlquist, S., 1974. Island Biology. Columbia University Press, New York, USA. Case, T.J., Bolger, D.L., Richman, A.D., 1992. Reptilian extinctions: the last ten thousand years. In: Fielder, P.L., Jain, S.K. (Eds.), Conservation Biology: The Theory and Practice of Nature Conservation, Preservation, and Management. Chapman and Hall, New York, USA, pp. 91–125.

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS 12

E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx

Castella, J.-C., Trung, T.N., Boissau, S., 2005. Participatory simulation of land-use changes in the northern mountains of Vietnam: the combined use of an agentbased model, a role-playing game, and a geographic information system. Ecology and Society 10 (1), 27. . Chefaoui, R.M., Hortal, J., Lobo, J.M., 2005. Potential distribution modelling, niche characterization and conservation status assessment using GIS tools: a case study of Iberian Copris species. Biological Conservation 122 (2), 327–338. Cheke, A.S., 1987. An ecological history of the Mascarene Islands, with particular reference to extinctions and introduction of land vertebrates. In: Diamond, A.W. (Ed.), Studies of the Mascarene Islands Birds. Cambridge University Press, Cambridge, UK, pp. 5–89. Conservation International, 2000. Designing Sustainable Landscapes. Center for Applied Biodiversity Science, Washington, DC, USA. Coppolillo, P., Gomez, H., Maisels, F., Wallace, R., 2004. Selection criteria for suites of landscape species as a basis for site-based conservation. Biological Conservation 115 (3), 419–430. Cowling, R.M., 2005. Maintaining the research-implementation continuum in conservation. Society for Conservation Biology Newsletter 12, 4. . Cowling, R.M., Heijnis, C.E., 2001. The identification of broad habitat units as biodiversity entities for systematic conservation planning in the Cape Floristic Region. South African Journal of Botany 67, 15–38. Cowling, R.M., Pressey, R.L., 2003. Introduction to systematic conservation planning in the Cape Floristic Region. Biological Conservation 112 (1–2), 1–13. Cowling, R.M., Pressey, R.L., Lombard, A.T., Desmet, P.G., Ellis, A.G., 1999. From representation to persistence: requirements for a sustainable reserve system in the species-rich Mediterranean-climate deserts of South Africa. Diversity and Distributions 5, 51–71. Cudihhy, L.W., Stone, C.P., 1990. Alteration of native Hawaiian vegetation, Effects of Humans, Their Activities and Introductions. Cooperative National Park Resources Studies Unit, University of Hawaii at Manoa, Honolulu, USA. Diamond, J.M., 1989. The present, past and future of human caused extinctions. Philosophical Transactions of the Royal Society of London B 325, 469–477. Dmowski, K., Kozakiewicz, M., 1990. Influence of a shrub corridor on movements of passerine birds to a lake littoral zone. Landscape Ecology 4, 99–108. Driver, A., Cowling, R., Maze, K., 2003. Planning for Living Landscapes: Perspectives and Lessons from South Africa. Botanical Society of South Africa, Cape Town. ESRI, 2006. ArcGIS Spatial Analyst version 9.2. Environment Research System Inc., Redlands, CA, USA. Ferrier, S., 2002. Mapping spatial pattern in biodiversity for regional conservation planning: where to from here? Systematic Biology 51 (2), 331–363. Ferrier, S., Pressey, R.L., Barrett, T.W., 2000. A new predictor of the irreplaceability of areas for achieving a conservation goal, its application to real-world planning, and a research agenda for further refinement. Biological Conservation 93 (3), 303–325. Fritts, T.H., Rodda, G.H., 1998. The role of introduced species in the degradation of island ecosystems. Annual Reviews of Ecology and Systematics 29, 113–140. Fullera, T., Munguy´, M., Mayfielda, M., Sanchez-Corderob, V., Sarkara, S., 2006. Incorporating connectivity into conservation planning: a multi-criteria case study from central Mexico. Biological Conservation 133, 131–142. Gigord, L., Picot, F., Shykoff, J., 1999. Effects of habitat fragmentation on Dombeya acutangula (Sterculiaceae), a native tree on La Réunion (Indian Ocean). Biological Conservation 88, 43–51. Gillot, P.-Y., Lefèvre, J.-C., Nativel, P.-E., 1994. Model for the structural evolution of the volcanoes of Réunion Island. Earth and Planetary Science Letters 122 (3–4), 291–302. Jacquemyn, H., Micheneau, C., Roberts, D.L., Pailler, T., 2005. Elevational gradients of species diversity, breeding system and floral traits of orchid species on Réunion Island. Journal of Biogeography 32, 1751–1761. Jaquemet, S., Le Corre, M., Quartly, G.D., 2007. Ocean control of the breeding regime of the sooty tern in the southwest Indian ocean. Deep Sea Research Part I: Oceanographic Research Papers 54 (1), 130–142. Knight, A.T., Cowling, R.M., Campbell, B.M., 2006a. An operational model for implementing conservation action. Conservation Biology 20 (2), 408– 419. Knight, A.T., Driver, A., Cowling, R.M., Maze, K., Desmet, P.G., Lombard, A.T., Rouget, M., Botha, M.A., Boshoff, A.F., Castley, J.G., Goodman, P.S., Mackinnon, K., Pierce, S.M., Sims-Castley, R., Stewart, W.I., von Hase, A., 2006b. Designing systematic conservation assessments that promote effective implementation: best practice from South Africa. Conservation Biology 20 (3), 739–750. Knight, A.T., Smith, R.J., Cowling, R.M., Desmet, P.G., Faith, D.P., Ferrier, S., Gelderblom, C.M., Grantham, H., Lombard, A.T., Maze, K., Nel, J.L., Parrish, J.D., Pence, G.Q.K., Possingham, H.P., Reyers, B., Rouget, M., Roux, D., Wilson, K.A., 2007. Improving the key biodiversity areas approach for effective conservation planning. BioScience 57 (3), 256–261. Knight, A.T., Cowling, R.M., Rouget, M., Balmford, A., Lombard, A.T., Campbell, B.M., 2008. Knowing but not doing: selecting priority conservation areas and the research–implementation gap. Conservation Biology 22 (3), 610–617. Komdeur, J.A., Pels, M.D., 2005. Rescue of the Seychelles warbler on Cousin Island, Seychelles: the role of habitat restoration. Biological Conservation 124, 15–26. Lagabrielle, E., 2007. Conservation planning and territorial modelling in Réunion Island: stakes and methods. PhD Thesis, Saint-Denis, University of Réunion Island, France.

Lagabrielle, E., Le Bourgeois, T., Durieux, L., Robin, M., Strasberg, D., 2008. Planification systémique de la conservation de la biodiversité à l’île de La Réunion (Océan Indien). Revue Internationale de Géomatique 18 (3), 283–303. Lane, M.B., 2006. Towards integrated coastal management in Solomon Islands: Identifying strategic issues for governance reform. Ocean and Coastal Management 49 (7–8), 421–441. Lavergne, C., 1999. The invasive woody weed Ligustrum robustum subsp. walkeri threatens native forest on La Réunion. Biological Invasions 1 (4), 377–392. Le Corre, M., Safford, R.J., 2001. La Réunion and Iles Eparses. In: Fishpool, L.D.C., Evans, M.I. (Eds.), Important Bird Areas in Africa and Associated Islands. Priority Sites for Conservation. Birdlife International, Cambridge, UK, pp. 693–702. Le Corre, M., Strasberg, D., 2002. Inventaire et conservation des éléments indigènes de la flore et de l’avifaune sur le littoral réunionnais. The Journal of Nature 14 (1), 51–58. Lee, K.N., 2002. Appraising adaptive management. Conservation Ecology 3 (2). . Lombard, A.T., Cowling, R.M., Pressey, R.L., Rebelo, A.G., 2003. Effectiveness of land classes as surrogates for species in conservation planning for the Cape Floristic Region. Biological Conservation 112 (1–2), 45–62. Lombard, A.T., Reyers, B., Schonegevel, L.Y., Cooper, J., Smith-Adao, L.B., Nel, D.C., Froneman, P.W., Ansorge, I.J., Bester, M.N., Tosh, C.A., Strauss, T., Akkers, T., Gon, O., Leslie, R.W., Chown, S.L., 2007. Conserving pattern and process in the Southern ocean: designing a marine protected area for the prince Edward Islands. Antarctic Science 19 (1), 39–54. Mac Arthur, R.H., Wilson, E.O., 1967. Island Biogeography. Princeton University Press, Princeton, NJ, USA. Macdonald, I.A.W., Thébaud, C., Strahm, W.A., Strasberg, D., 1991. Effects of alien invasions on native vegetation remnants on La Réunion (Mascarenes Islands, Indian ocean). Environmental Conservation 18 (1), 51–61. Mairine, P., Bachèlery, P., 1997. Un grand épisode érosionnel dans l’histoire ancienne du Piton de la Fournaise (île de la Réunion). Comptes Rendus de l’Académie des Sciences – Series IIA – Earth and Planetary Science 325 (4), 243– 249. Margules, C.R., Pressey, R.L., 2000. Systematic conservation planning. Nature 405, 243–253. Mittermeier, R.A., da Fonseca, G.A.B., Hoffman, M., Pilgrim, J., Brooks, T., Gill, P.R., Mittermeier, C.G., Lamoreux, J., 2005. Hotspots Revisited: Earth’s Biologically Richest and Most Endangered Terrestrial Ecoregions. CEMEX, Conservation International. Moritz, C., 2002. Strategies to protect biological diversity and the evolutionary processes that sustain it. Systematic Biology 5, 238–254. Mourer-Chauvire, C., Bour, R., Ribes, S., Moutou, F., 1999. The avifauna of Réunion Island (Mascarene Islands) at the time of arrival of the first Europeans. Avian Paleontology at the closes of the 20th Century, Washington, DC, 4–7 June 1996, Smithsonian Contributions to Paleobiology. Mueller-Dombois, D., Loope, L.L., 1990. Some unique ecological aspects of oceanic island ecosystems. Botanical Research and Management in Galapagos. Monographs in Systematic Botany from the Missouri Botanical Garden, vol. 32, pp. 21–28. Olesen, J.M., Valido, A., 2003. Lizards as pollinators and seed dispersers: an island phenomenon. Trends in Ecology and Evolution 18 (4), 177–181. Pailler, T., Micheneau, C., 2005. Orchid–pollinator interactions on Réunion Island. In: 18th World Orchid Conference, Dijon, France. Pelling, M., Uitto, J.I., 2001. Small island developing states: natural disaster vulnerability and global change. Global Environmental Change Part B: Environmental Hazards 3 (2), 49–62. Pierce, S.M., Cowling, R.M., Knight, A.T., Lombard, A.T., Rouget, M., Wolf, T., 2005. Systematic conservation planning products for land-use planning: interpretation for implementation. Biological Conservation 125, 441–458. Richardson, D.M., Pysek, P., Rejmanek, M., Barbour, M.G., Panetta, F.D., West, C.J., 2000. Naturalization and invasion of alien plants: concepts and definitions. Diversity and Distribution 6, 93–107. Rivals, P., 1952. Etudes sur la végétation de l’île de La Réunion. PhD Thesis, University of Toulouse, France. Rouget, M., Cowling, R.M., Pressey, R.L., Richardson, D.M., 2003. Identifying spatial components of ecological and evolutionary processes for regional conservation planning in the Cape Floristic Region, South Africa. Diversity and Distribution 9, 191–210. Rouget, M., Cowling, R.M., Lombard, A.T., Knight, A.T., Kerley, G.I.H., 2006. Designing large-scale conservation corridors for pattern and process. Conservation Biology 20 (2), 549–561. Roux, D., de Moor, F., Cambray, J., Barber-James, H., 2002. Use of landscape-level river signatures in conservation planning: a South African case study. Conservation Ecology 6 (2). . Simberloff, D., 1992. Extinction, survival, and effects of birds introduced to the Mascarenes. Acta Oecologica 13 (6), 663–678. Stattersfield, A., Capper, D., 2000. Threatened Birds of the World. Birdlife International, Cambridge, UK. Strasberg, D., Rouget, M., Richardson, D.M., Baret, S., Dupont, J., Cowling, R.M., 2005. An assessment of habitat diversity, transformation and threats to biodiversity on Réunion Island (Mascarene Islands, Indian Ocean) as a basis for conservation planning. Biodiversity and Conservation 14 (12), 3015–3032. Thébaud, C., Strasberg, D., 1991. Conservation of plant species and their habitat on the island of La Réunion (Mascarene Islands, Indian Ocean). WWF International, Project 4543, Genève, Switzerland.

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022

ARTICLE IN PRESS E. Lagabrielle et al. / Biological Conservation xxx (2009) xxx–xxx Thinon, P., Martignac, C., Metzger, P., Cheylan, J.-P., 2007. Analyse géographique et modélisation des dynamiques d’urbanisation à La Réunion. Cybergeo 389. . Townsend, C.R., 1996. Invasion biology and ecological impacts of brown trout Salmo trutta in New Zealand. Biological Conservation 78 (1–2), 13–22. Vitousek, P.M., Loope, L., Adsersen, H.E., 1995. Islands: Biological Diversity and Ecosystem Function. Springer-Verlag, Berlin, Germany.

13

Warren, B.H., Bermingham, E., Prys-Jones, R.P., Thebaud, C., 2006. Immigration, species radiation and extinction in a highly diverse songbird lineage: whiteeyes on Indian Ocean islands. Molecular Ecology 15 (12), 3769–3786. Whittaker, R.J., 1998. Island Biogeography, Ecology, Evolution and Conservation. Oxford University Press, Oxford, UK. Whittaker, R.J., Willis, K.J., Field, R., 2001. Scale and species richness: toward a general, hierarchical theory of species diversity. Journal of Biogeography 28, 453–470.

Please cite this article in press as: Lagabrielle, E., et al. Identifying and mapping biodiversity processes for conservation planning ... Biol. Conserv. (2009), doi:10.1016/j.biocon.2009.02.022