Protected areas and forest fragmentation: sustainability index for prioritizing fragments for landscape restoration

Tatiana Possati Vieira De Matos; Veridiana Possati Vieira De Matos; Kaline De Mello; Roberta Averna Valente

doi:10.1080/24749508.2019.1696266

Protected areas and forest fragmentation: sustainability index for prioritizing fragments for landscape restoration

De Matos, Tatiana Possati Vieira; De Matos, Veridiana Possati Vieira; De Mello, Kaline; Valente, Roberta Averna 2021-01-02 00:00:00 GEOLOGY, ECOLOGY, AND LANDSCAPES 2021, VOL. 5, NO. 1, 19–31 INWASCON https://doi.org/10.1080/24749508.2019.1696266 RESEARCH ARTICLE Protected areas and forest fragmentation: sustainability index for prioritizing fragments for landscape restoration a a b Tatiana Possati Vieira De Matos , Veridiana Possati Vieira De Matos , Kaline De Mello and Roberta Averna Valente Programa de Pós-graduação em Diversidade Biológica e Conservação, Universidade Federal de São Carlos, campus Sorocaba – UFSCar, b c Sorocaba, Brazil; Department of Ecology, Institute of Biosciences, University of São Paulo. R. do Matão, São Paulo, Brazil; Departamento de Ciências Ambientais, Universidade Federal de São Carlos, campus Sorocaba – UFSCAR, Sorocaba, Brazil ABSTRACT ARTICLE HISTORY Received 21 May 2019 Tropical forests are threatened due to forest clearing and fragmentation which lead to Accepted 16 October 2019 a decrease in forest cover area and landscape connectivity, while also increasing edge eﬀects. These impacts aﬀect biodiversity and ecosystem service provision which are essential for KEYWORDS humanity’s well-being. Protected areas are created to minimize these eﬀects on biodiversity, Landscape connectivity; but many of them are threatened due to forest fragmentation in the surrounding areas. Thus, forest conservation; atlantic forest restoration is needed to ensure the suitability of protected areas in the landscape. forest; land use; buﬀer zone However, restoration planning needs to adopt an ecological landscape approach to ensure the recovery of the biodiversity and the ecological processes. This study developed a forest sustainability index (FSI) which represents the forest patch potential in facilitating landscape restoration in a protected area and its surroundings. A land-use/land-cover map was used to calculate landscape metrics at two levels: landscape and patch. Landscape metrics were subsequently selected to create the index. The tested landscape presents a great number of fragments with most of them being small in size and having irregular shapes. The focus areas for forest restoration are located close to forest fragments with higher FSI values, as these can facilitate natural restoration and guarantee the maintenance of the ecosystem processes. 1. Introduction such as exotic species introduction and other serious impacts on species diversity and composition, commu- Tropical forests have been replaced by land uses lead- nity structure and dynamics, and ecosystem function- ing to deforestation and forest fragmentation, which is ing (Carrara et al., 2015; Joly, Metzger, & Tabarelli, the main cause of biodiversity loss and aﬀects natural 2014; Santos et al., 2016; Slade et al., 2013). resource and ecosystem service provisioning including Furthermore, these impacts also aﬀect the ecosystem wood, drinking water, and soil protection. The ability to recover from disturbances, i.e. its resilience to Atlantic Forest in particular is a biodiversity hotspot anthropogenic changes (Bregman, Sekercioglu, & (Rezende et al., 2018) which is highly threatened since Tobias, 2014; Sekercioglu et al., 2015). it has been reduced to 28% of its original cover mainly In the current global scenario of environmental consisting of small fragments and edge-aﬀected or degradation, protected areas are the cornerstone of secondary vegetation disconnected from larger frag- eﬀorts to sustain biodiversity and natural ecosystem ments (Rezende et al., 2018; Ribeiro, Metzger, processes (Bingham et al., 2019). However, they are Martensen, Ponzoni, & Hirota, 2009). According to threatened due to the forest fragmentation in the sur- Ferraz et al. (2014), the Atlantic forest cover is cur- rounding areas. The protected area and its buﬀer zone rently a heterogeneous forest mosaic of diﬀerent ages must be well managed to provide habitat quality for and at many degrees of conservation and located in the species and to decrease neighborhood eﬀects diﬀerent landscape conditions within a complex (Moraes, Mello, & Toppa, 2017; Saura et al., 2018). matrix composed by agriculture and other uses. This For example, planted forests can contribute to species scenario occurs worldwide in most of the tropical conservation if they keep a diverse and complex landscapes with forest remnants (Chazdon, 2017; understory (Fonseca et al., 2009). Singh & Mishra, 2014). Therefore, forest restoration can play an important Forest fragmentation decreases the forest cover area role in mitigating the forest fragmentation impacts on and the landscape connectivity, thereby increasing the the natural ecosystems and their services (Geneletti, edge eﬀects which lead to physical and biotic alterations 2004; Leite, Tambosi, Romitelli, & Metzger, 2013). associated with the artiﬁcial boundaries of fragments Forest restoration must be directed to restore the CONTACT Tatiana Possati Vieira de Matos veri_gemeas_@hotmail.com Programa de Pós-graduação em Diversidade Biológica e Conservação, Universidade Federal de São Carlos, campus Sorocaba – UFSCAR, Rod. João Leme dos Santos, km 110, Sorocaba, Brasil © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group on behalf of the International Water, Air & Soil Conservation Society(INWASCON). This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 20 T. P. V. DE MATOS ET AL. landscape structure under the perspective of the land- restoration in a protected area and its surroundings, scape ecology in order to ensure biodiversity recovery and which is located in a complex agricultural matrix. in diﬀerent taxa and the ecological processes and ser- The speciﬁc objectives were: (1) to assess the landscape vices (Leite et al., 2013; Metzger & Brancalion, 2013; structure based on landscape metrics; (2) to select metrics Santos Junior, Marques, Lima, & Dos Anjos, 2016; to create the forest sustainability index; and (3) to apply Stanturf, Palik, & Dumroese, 2014; Vettorazzi & the index to identify priority forest fragments that Valente, 2016). Thus, forest restoration based on eco- enhance landscape restoration in the protected area and logical criteria can improve the forest structure and its surroundings. landscape connectivity, promoting the connection The creation of this index based on landscape between the protected areas and other forest metrics assists in diagnosing sites which still have fragments. a minimum of resilience with high restoration poten- Ecological criteria can be represented by the land- tial, thus favoring an increase in biodiversity and scape characteristics that inﬂuence the restoration pro- maintaining ecological processes and services. In this cess, including the species richness, establishment, way, the index will facilitate implementing restoration dispersal and interrelations (Banks-Leite, Ewers, projects, increasing environmental gains and decreas- Kapos, Martensen, & Metzger, 2011; Santos Junior ing costs, potentiating the natural regeneration. et al., 2016). Recent studies have shown that forest restoration plans in landscapes with minimum resili- ence, i.e. which still present connection between forest 2. Material and methods fragments, have more success and mitigate costs better 2.1 Study area than others on highly degraded landscapes (Bortoleto, Figueira, Dunning, Rodgers, & Silva, 2016; Molin, The study area is Capão Bonito National Forest (CBNF) Chazdon, Ferraz, & Brancalion, 2018;Strassburg et al., and its surrounding area, comprising a total area of 2019; Tambosi, Martensen, Ribeiro, & Metzger, 2014). 74,575 ha, wherein the CBNF represent 4,773.83 ha Thus, assessing the landscape pattern is the ﬁrst and the surrounding area comprises 69,790.65 ha. step to identify potential areas for forest restoration Considering that the CBMF Management Plan is in plans (Aronson & Sasha, 2013; De Jesus, Ferreira, elaboration because it will be supported by this work Aragão, Santos, & Rocha, 2015; Metzger & (among others), we considered a surrounding area of Brancalion, 2013; Moraes et al., 2017; Strassburg 10 km from the perimeter protected area as a primary et al., 2019; Valente & Vettorazzi, 2008). Landscape study The CBNF is located in the Alto de metrics can quantify and describe the landscape com- Paranapanema Watershed, São Paulo state, Southeast ponents (matrix, fragments, and corridors) and their Brazil (Figure 1), in the Atlantic Forest Biome. National spatial distribution, minimizing the uncertainties of forest is a category of sustainable use protected area in the landscape analysis (Convertino et al., 2013). Brazil, where the primary purpose is sustainable exploi- The combination of various landscape metrics can tation of the forest and its natural resources. be used to build important indexes which are currently The CBNF is currently covered by native vegetation applied for deﬁning priority areas for forest conserva- and stands of exotic forest species (originally commer- tion and restoration (Bortoleto et al., 2016; Mello, cial stands) which are related to the history of the area. Toppa, & Cardoso-Leite, 2016; Valente & Vettorazzi, They have been replaced by natural vegetation by the 2008; Vettorazzi & Valente, 2016). Mello et al. (2016) management of the protected area. Even with the and Bortoleto et al. (2016) developed indexes for presence of exotic species, the area was deﬁned as priority areas for forest conservation and restoration, a protected area by the federal government because it respectively, in the same area in Southern Brazil, while is an ecological tension area between two important Vettorazzi and Valente (2016) used several landscape Biomes: Cerrado and Atlantic Forest. Thus, the area criteria to identify priority areas for forest restoration presents forest remnants of savanna, mixed ombro- aiming toward water resource conservation. philous forest, and semi-deciduous seasonal forest. In Moreover, Santos et al. (2016) selected potential forest addition, there are Araucaria angustifolia plantations fragments for seed harvesting based on landscape (endangered species according to IUCN) which are metrics. However, potential forest fragments which already considered an important genetic bank for cur- contribute to the success of restoration actions in rent and future use. The exotic species such as Pinus a landscape are still needed to be identiﬁed, among elliottii in these same areas have supported scientiﬁc which not only aim for local restoration, but also experiments (Bechara, 2006; Veloso, 2012). consider a landscape restoration perspective (Molin CBNF was a large farm originally named “Itanguá”, et al., 2018; Vettorazzi & Valente, 2016). which in 1944 was categorized as Forest Park for wood In this context, the main objective of this study was to production. The plantation of Araucaria angustifolia develop a forest sustainability index that represents the has started in 1945. In the 1950s, the Pinus elliotti potential of a forest fragment to facilitate landscape plantations have expanded to reach 2,400 ha, and in GEOLOGY, ECOLOGY, AND LANDSCAPES 21 Figure 1. Capão Bonito National Forest and its surrounding 10 km-buﬀer location, in the São Paulo state, Southeast Brazil. 1953 the farm was denominated as Forest Park 1997). Most of the soil types are Latosols and Deep “Getúlio Vargas”. The genus Pinus was introduced in sandy-quartz soils (Peixoto & Theodorovicz, 2009). the areas to conduct experiments due to the large-scale reforestation policy adopted in Brazil at this time, 2.2 Landscape structure which used Pinus spp. to restore deforested areas. The area ﬁnally became a National Forest in 1968 A land-use/land-cover map was produced by super- and it started to be managed by the national agency vised digital classiﬁcation to assess the landscape of protected areas – Chico Mendes Institute for structure. For this, SPOT satellite images Biodiversity Conservation (Instituto Chico Mendes (2,5 m-spatial resolution; spectral bands: blue, green, De Conservação Da Biodiversidade [ICMBio], 2017). red and near-infrared; year: 2009) supplied by the Because National Forests area sustainable use Environment Secretariat of the State of São Paulo, protected areas, planted forests of exotic species Environmental Planning Coordination (SMA-CPLA) such as Pinus spp. is allowed, but it must follow were used. The images were re-projected in UTM some regulationssuchaskeeping native forest 23S – SIRGAS 2000, as it was for all data set. The along the river banks and around springs (Brasil, software ENVI (ENVI – Itt Visual Information 2012). Studies in the CBNF are of great relevance Solutions 2009) was used for the geometric and atmo- due to the rich understory of native vegetation spheric corrections of the spectral bands and for the within the forest plantation stands and the several image classiﬁcation by using Maximum Likelihood environments which oﬀer varied resources for ani- algorithm. The following land-use/land-cover types mal and plant species. were considered: The surrounding area of the CBNF is a complex matrix composed by various land uses, including agri- (a) Native forest: native forest fragments; culture in small and large scale, pasture, and commercial (b) Planted Pinus sp. forest (Pinus): Pinus sp. forest plantations. There is also a highway which sepa- stands with various levels of understory rates the protected area into two parts with diﬀerent sizes regeneration; (Figure 1). Its proximity to other large-protected areas (c) Planted Eucalyptus sp. forest (Eucalyptus): (Intervales State Park in the South, Carlos Botelho State Eucalyptus sp. stands managed in diﬀerent Park in the Southeast and Angatuba Ecological Station in periods; the northwest) provides importance for increasing land- (d) Planted Araucaria angustifolia forest scape connectivity through ecological restoration. (Araucaria): Araucaria angustifolia stands The region is under the inﬂuence of a Cwa climate with various levels of understory regeneration; (subtropical-dry winter) and annual precipitation is (e) Agriculture: soybean, wheat, corn and citrus above 22°C (Centro De Pesquisas Meteorológicas crops; E Climáticas Aplicadas À Agricultura [CEPAGRI], (f) Pasture: grassland destined to intensive and 2013). The average elevation is 700 m above the sea extensive cattle production systems, or without level, with slightly undulated relief (Ross & Moroz, cattle; 22 T. P. V. DE MATOS ET AL. (g) Urban areas: urban patches located in the sur- Thus, we developed the FSI to prioritize forest frag- rounding area of the CBNF; ments for landscape restoration (Eq. 1). (h) Road network: roads and highways in the study FSI ¼ IxAREA þ IxSHAPE þ IxENN (1) n n n area; (i) Rivers network: rivers, streams, and springs in Where, the study area. I: importance of the metric to the landscape restoration; AREA : normalized AREA value (scale 1 to River networks, road networks, and urban areas 7bytes); SHAPE : normalized SHAPE value (scale 1 to 7 were obtained by on-screen digitizing of the SPOT bytes); ENN : normalized ENN value (scale 1 to 7 bytes). images. These landscape metrics that compose the FSI were A confusion matrix and the Kappa coeﬃcient based on the following concepts: (Congalton & Green, 1998) were used to assess accu- racy and agreement, based on 80 ground control (a) AREA: this is one of the most important pieces points distributed stratiﬁed randomly along the study of information of the forest fragment which area. We obtained a global accuracy of 93.41% and an inﬂuences other characteristics of the forest accuracy of 99.14% for the native forest. The map was fragments such as perimeter and core area. considered very good according to Landis and Kock The greatest forest fragments in the landscape (1977) based on the kappa value (0.89). present the largest core area and they present Landscape metrics were used to describe the land- higher natural resources and have greater bio- scape structure (Online Resource 1) calculated for diversity than small forest patches in most cases landscape level (land-use/land-cover class) and for (Fahrig, 2003; Nichol, Abbas, & Fischer, 2017; forest patches level through the extension V-Late 2.0 Rocha-Santos et al., 2016). Metzger, Bernacci, in the software ArcGIS 10.1. We chose these metrics and Goldenberg (1997) pointed out that the based on their contents and their application to eco- minimum forest patch size for its ﬂoristic sta- logical processes aiming at their importance for deﬁn- bility in the semi-deciduous seasonal forest is ing priority areas for landscape restoration (Baral, 25 ha. Keenan, Sharma, Stork, & Kasel, 2014; Oakleaf et al., (b) SHAPE: the shape of a habitat patch can be 2017; Vettorazzi & Valente, 2016). Forest patches used to represent the edge eﬀect on the forest smaller than 1 ha were not considered in the analysis, fragment. Forest fragments with irregular according to (Valente and Vettorazzi (2008)). shapes suﬀer more inﬂuence of the edge eﬀect, A 30 m-edge eﬀect was used to calculate CORE and and thus they are under a greater inﬂuence of TCAI metrics, as proposed by Périco and Cemin the matrix. The edge-dominated Atlantic forest (2006), and Milan and Moro (2012) for semi- fragments move towards early-successional deciduous seasonal forest remnants. Boxplot graphics systems, compromising the forest fragment were performed in the software RCore Team 3.0.2 sustainability (Joly et al., 2014). However, (2013) to describe the landscape pattern. large forest fragments in highly fragmented landscapes tend to have more irregular shape than small patches (Mello et al., 2016). 2.3 Forest suitability index (c) ENN: this indicates the distance of the nearest Some tests were performed to build the forest suitabil- neighbor related to the degree of isolation of ity index (FSI) after obtaining data for the whole land- the forest fragments (McGarigal, 2015). The scape. We performed these tests and accessed previous connectivity among the forest fragments is prioritization maps using the landscape metrics in important to the forest restoration under addition to literature consultation to develop the a landscape-based approach (Molin et al., index. We selected uncorrelated metrics representing 2018). The landscape connectivity will improve characteristics of forest fragments which are impor- the species dispersion and recruitment capa- tant for biodiversity conservation. In addition to the city, which are related to the forest restoration forest patch size, its isolation level, connectivity, and success (Peña-Domene, Minor, & Howe, 2016). its quality inﬂuence the persistence of species in tro- Landscape connectivity analysis is a valuable pical agricultural landscapes (Banks-Leite, Ewers, criterion for prioritizing restoration opportu- Kapos, & Metzger, 2013; Uezu & Metzger, 2016). nities, helping to identify areas which hold the Next, we tested the combination of the following greatest potential for increasing connectivity metrics: AREA, SHAPE, CORE, and ENN. We calcu- and the natural integrity of the landscape lated the correlation among the metrics to avoid the (Rudnick et al., 2012). use of similar metrics leading to wrong results and interpretations. The metrics combination which best After selecting the metrics set, we then performed represented our goal was: AREA, SHAPE, and ENN. tests to deﬁne the importance for each metric. When GEOLOGY, ECOLOGY, AND LANDSCAPES 23 the same value was attributed for SHAPE and ENN, and protected area. We elaborated a map of the forest closetothe valuefor AREA (AREA40%,SHAPE 30%, fragments classiﬁed according to the FSI values. and ENN 30%), we observed that larger and more con- nected forest fragments showed lower FSI than small and 3. Results irregular fragments (Figure 2). Thus, the index did not represent the objective. Thus, we performed tests to The land-use/land-cover map (Figure 3) presented obtain values which represent forest fragments that do a global classiﬁcation accuracy of 99.14%, which nothavealargeedge eﬀect and can function as source means that the map has a high agreement with the areas for landscape restoration over time. ground reality. This map indicated that the landscape We considered a relative value (total of 100%) of comprised by the CBNF and its surrounding area is importance (I) for each landscape metric according to predominantly agricultural (Table 2). their importance for the objective. Thus, a value of Agricultural areas such as corn, soy, and wheat 60% was assigned for AREA, 10% for SHAPE and 30% crops represent 40% (CA = 40) of the total landscape for ENN. (Table 2), mainly out of the protected area (Figure 2). The metrics values were normalized (1–7bytes- It showed the greatest MPS value (Table 2) among the scale) to avoid the use of diﬀerent units or scales to land use/land cover types. The largest agricultural calculate FSI, as shown in Table 1. patch has 1832 ha (Figure 4(a)), which is also the greatest patch in the landscape. Another characteristic that is a consequence of its predominance in the land- 2.4 Forest suitability index applied to the capão scape is the low ENNM value (45 m – Table 2), the bonito national forest same value of native forest, showing that both are very We calculated the FSI for forest fragments in the scattered over the landscape, but agriculture is predo- CBNF and its surrounding area to identify the impor- minant (Figure 3). tant fragments in the landscape for restoration actions Native forest was the second major land cover/land in order to improve the landscape connectivity for this use class in the study area, representing 24.45% of the Figure 2. Tests for the deﬁnition of importance percentage (I) for each metric (AREA 40%, SHAPE 30% and ENN 30%). 24 T. P. V. DE MATOS ET AL. Table 1. Normalized values for the landscape metrics. (11.06%) and Pinus sp. (6.71%) plantations. Shape Normalized scale However, the most important land use/land cover Area (ha) (dimensionless) ENN (m) (bytes) classes in the protected areas are Araucaria and < 5 >7 >36 1 Pinus plantations, especially in the largest area. The 5 – 25 6 −730 – 36 2 25 – 45 5 –624 – 30 3 CBNF is divided by a state road into two parts of 45 – 65 4 –518 – 24 4 diﬀerent sizes (Figure 3). The smallest portion pre- 65 – 85 3 –412 – 18 5 85 – 115 2 –36 – 12 6 sents some agricultural areas. >115 1 –2<6 7 On the other hand, the Eucalyptus plantations are mainly located in its surroundings in the east of the CBNF (Figure 3), and they belong to a paper landscape (Table 2). However, the remaining forest and pulp company. It represents the second land patches are interspersed with patches cleared for agri- use/land cover class with the highest NP value culture, planted forest or other uses, comprising (730) behind the native forest (Table 2). It showed a complex mosaic. It showed the greatest NP (941 the lowest MPS value (11.26 ha); however, the patches) and MSI (3.17) values among the land use/ patch sizes range from 1 ha to 677 ha (Figure 3). land cover classes, indicating the habitat fragmenta- In contrast, Araucaria and Pinus presented a MPS tion. It also presented a lower TCAI value than planted value of 28 and 29 ha, and the largest patches forests (Table 2) as a consequence of the forest frag- showed values of 323 and 224 ha, respectively mentation. MPS was 19 ha, however, its patch sizes (Figure 4(a)). varied from 1 ha to 1800 ha (Figure 4(a)), where the The three types of planted forest presented similar largest forest fragments are in the east of the study area TCAI (74.36%; 72.07% and 76.13%) and MSI (1.60; (Figure 3). Although it presented the same ENNM 1.90 and 1.90) values, i.e. they are similar regarding value of agriculture, the ENN values varied from their shapes and edge proportion. Planted forest 2.5 m to 604.78 m, which was the lowest range showed less complex shapes than the other land among the land use/land cover classes considering cover/land use classes (MSI < 2 – Table 2, Figure 3), this landscape metric (Figure 4(d)). which can be a result of the plantation design. They It is important to highlight that the native forest is presented a greater ENNM value than native forest not representative in the protected area. Its largest and agriculture of 45 m (Araucaria = 146.11 m; remnants are in the surrounding area, close to the Eucalyptus = 99.38 m and Pinus = 70.21 m). This Eucalyptus crops. Many forest fragments are divided shows that planted forests are concentrated in some into smaller patches by roads (Figure 3). The road portions of the study area, occurring in low density in network represents 2.55% of the landscape. the north and south (Figure 3). Thus, the greatest The third-largest land use/land cover class is distance of an Araucaria patch from another similar planted forest, accounting for 20.62% distributed patch was 2,900 m, while it was about 3,300 m for into Araucaria angustifolia (2.85%), Eucalyptus sp. Eucalyptus and Pinus (Table 2). Figure 3. Land use/land cover of the Capão Bonito National Forest (CBNF) and its surrounding 10 km-buﬀer in the State of São Paulo, Southeast Brazil. GEOLOGY, ECOLOGY, AND LANDSCAPES 25 Table 2. Landscape metrics for class level. Landscapemetrics Land use/land cover class CA (ha) CA (%) NP MPS (ha) MSI TCAI (%) ENNM (m) Agriculture 29,259.68 40.00 496 58.99 2.44 – 45.36 Araucaria 2116.87 2.85 76 27.85 1.90 76.13 146.11 Eucalyptus 8218.76 11.06 730 11.26 1.60 74.36 99.38 Pinus 4991.14 6.71 172 29.02 1.90 72.07 70.21 Pasture 8005.59 10.77 287 27.89 2.36 – 98.83 Nativeforest 18,173.32 24.45 941 19.31 3.17 54.49 44.68 CA: Total area; NP: Number of patches; MPS: Mean patch size; MSI: Mean shape index; TCAI: Total core area index; ENNM: Mean distance from the nearest neighboring patch. Figure 4. Boxplot of the variation of the landscape metrics: (a) patch size (PS), (b) shape index (SHAPE), (c) core area index (CA) and (d) distance from the nearest neighboring patch (ENN). Pasturelands represented 10.77% of the study The FSI application showed that important fragments area (Table 2) and they are concentrated in the for landscape restoration are large patches connected by south of the CBNF surroundings (Figure 3). They other forest patches in the landscape. As the forest cover are distributed into 287 patches with a MPS value is mostly comprised by forest patches with complex around 28 ha, close to the Araucaria and Pinus shapes, this metric received an importance value of values. The MSI value (2.36) was close to the agri- 10%. Thus, this metric has less impact than size and culture value, ranging from 1.29 to 6.15 (Figure 3), connectivity, but can still diﬀerentiate fragments with which represents more irregular shapes than similar importance only based on the ﬁrst two metrics planted forest. The ENNM value was similar to (Online Resource 1). The results also indicated that the the planted forest distances (99 m), showing that priority forest fragments for landscape restoration are the pasturelands are not well spread within the located in the CBNF surrounding area (Figure 5). landscape as agriculture and native forest. The forest fragments with the greatest FSI values are Urban areas represented less than 1% of the land- the largest patches (103 to 1176 ha) located in the scape. They are located in the northwest of the area, southeast, northeast, and northwest of the study area comprised by two little tows (Buri and Taquarivaí), (Figure 5). They represent strategic source areas in the which together have a population of 23,000. Urban landscape for forest restoration actions in their sur- areas, roads, and rivers represent, respectively, 0.93%, roundings. They are connected through other forest 2.55% and 1.31% of the total area. patches which showed lower FSI values, showing that 26 T. P. V. DE MATOS ET AL. Figure 5. Priority forest fragments for landscape restoration, according to the Forest Sustainability Index (FSI) in the Capão Bonito National Forest (CBNF) and its surrounding 10 km buﬀer in the State of São Paulo, Southeast Brazil. they are part of a complex landscape sustainability location, low slopes and not far from the coastal zone system even though they have less importance for land- (Mello et al., 2016; Moraes et al., 2017). scape restoration. The southern region above the pro- The landscape structure analysis showed that forest tected area has the forest fragments with the lowest FSI only represents 24.5% of the study area and it is highly values. However, they are important for the landscape fragmented, corroborated by the low core area and restoration regarding the local context of agriculture mean patch size values, and the high number of and pasture predominance. They presented small size patches (Table 2). Similar results were found by sev- (<5 ha), high ENN (up to 600 m) and irregular shapes eral studies conducted in the tropics, where natural (values ranging from 1.52 to 9.89). areas were replaced by agricultural crops (Carranza, Hoyos, Frate, Acosta, & Cabido, 2015; Moraes et al., 2017; Newman, McLaren, & Wilson, 2014; Ponte et al., 2017; Valente & Vettorazzi, 2005). The forest frag- 4. Discussion ments in our study are scattered within the agricul- The protected area surroundings are predominantly tural matrix (Figure 3), which can mainly aﬀect animal agricultural, presenting the greatest patches and very species that cannot transpose the open matrix; hence, scattered over the landscape (Figure 3, Table 2). Thus, the dispersal of some plant species is aﬀected it creates an agricultural matrix for the native species (Honorato, Crouzeilles, Ferreira, & Grelle, 2015). which exist in the CBNF and in the other forest frag- Most of the native forest fragments are in the sur- ments in the surrounding area. It is a common pattern rounding CBNF area (Figure 3), suﬀering pressure for many protected areas in Brazil and other tropical from the commercial planted forest stands and the regions (Garcia, Sawakuchi, Ferreira, & Ballester, expansion of agriculture and pasturelands. The high 2017; Ponte, Roch, Leinenkugel, Dech, & Kuenzer, shape metric values (MSI = 3.17 – Table 2, Figure 4) 2017), especially for those located in the region indicated that most of the forest fragments have elon- between Atlantic Forest and Cerrado, where agricul- gated and irregular shapes, which means they are tural expansion has been facilitated due to the strategic mainly composed by riparian corridors, which is GEOLOGY, ECOLOGY, AND LANDSCAPES 27 a common pattern in Atlantic Forest landscapes Pasture covered only 11% of the study area, but it (Mello et al., 2016; Moraes et al., 2017). Forest patches represented an important amount of land use in the with a complex shape such as riparian zones generate south where the forest fragments are small and limited more area under the edge eﬀect, showing a reduction to riparian corridors (Figure 3). According to in species richness and increased levels of species Latawiec, Strassburg, Brancalion, Rodrigues, and dominance (Joly et al., 2014; Matos et al., 2017). Gardner (2015), low-productivity pasturelands with In addition to the agriculture, pasture lands and few cattle are a great opportunity for forest restoration commercial stands of planted forest, the roads are an in tropical agricultural landscapes. The increase in important vector of habitat fragmentation in the study pasture productivity could provide new areas of land area, as shown in Figure 3. In addition, the state road for the restored forest. The forest restoration in our that divides the protected area into two parts is a threat study is important in the south area due to its high to the animal dispersion. The roads lead to deforesta- level of forest fragmentation, which compromises the tion, a decrease in patch size and an increase in the edge animals’ movement through the landscape. eﬀect, and also causes problems with roadkill and urban The FSI was a strong framework to determine expansion (Bager & Fontoura, 2013; Barber, Cochrane, priority forest fragments for landscape restoration Souza, & Laurance, 2014; Mello, Toppa, & Abessa, and can be replicated in other fragmented landscapes. 2011). Thus, it is a serious problem for many protected The mosaic comprised by agriculture, pasturelands, areas which are bordered or cut by roads (Barber et al., and planted forest allows restoration implementation, 2014; Diniz & Brito, 2015; Marcantonio, Rocchini, Geri, unlike in urban areas. According to Fagan, DeFries, Bacaro, & Amici, 2013;Moraesetal., 2017). Sesnie, Arroyo-Mora, and Chazdon (2016), the refor- The presence of planted forest (21%) is an important estation of priority areas in tropical agricultural land- characteristic of the study area, which is the predomi- scapes can improve connectivity by 2% with only 1% nant land use/land cover in the CBNF (Figure 3)due to gain in forest cover. its history of wood exploitation/extraction. The problem The focus areas for forest restoration are located caused by the expansion of exotic eucalyptus and pine close to the forest fragments with higher FSI values species over natural forest leading to habitat loss and (Figure 5), since they can facilitate natural restoration fragmentation has been reported in many tropical and guarantee the maintenance of ecosystem pro- regions (Ponte et al., 2017;Zamorano-Elgueta, cesses, as well as dispersion and pollination, which Benayas, Cayuela, Hantson, & Armenteras, 2015). are important for biodiversity restoration success. Although Eucalyptus sp. and Pinus sp. plantations The index prioritized forest fragments with the largest have been observed to be correlated to biodiversity size, close to others which are less irregular. These loss (Jara, Hylander, & Nemomissa, 2017), they can forest fragments are important because they can sup- provide regeneration sites for many native tree spe- port greater biological populations, high biodiversity cies in agricultural landscapes in the tropics (Silva, and diversity of habitats (Almeida, Gomes, & Queiroz, Poggiani, Sebbenn, & Mori, 2011), and may function 2011; Mello et al., 2016). Thus, they facilitate land- as environmental ﬁlters for native species coloniza- scape restoration because they still have a biological tion (Mendonça-Lima, Duarte, & Hartz, 2014). For structure and are connected to the landscape by other example, Zamorano-Elgueta et al. (2015) cited that forest fragments, as was indicated by Vettorazzi and natural forest regeneration in abandoned areas of Valente (2016). exotic tree plantations partly compensated forest In addition, the restoration of surrounding areas of loss in Chile. In fact, the understory of the planted forest fragments with high FSI value can improve their forests in the CBNF (Pinus sp. and Araucaria angu- size, shape, and connectivity. Thus, it is important to stifolia) present regeneration of native species of increase their potential for sheltering biodiversity and ombrophilous and semi-deciduous seasonal forests. being a dispersion source in the landscape because it According to Fonseca et al. (2009), tree monocul- can help to improve the ecological value of other forest tures (as Eucalyptus sp.) can greatly contribute to fragments (Valente & Vettorazzi, 2005). conserve biodiversity when maintaining a complex In this context, the maintenance of small fragments and diverse understory. with low FSI values is important to create habitat However, some native species can be inhibited by patches which are connected in the landscape, since Eucalyptus sp. or Pinus sp. roots (Zhang & Fu, 2009). they allow connection among forest fragments with According to Mendonça-Lima et al. (2014), native high importance (i.e. high FSI) (Mello et al., 2016). In pine (Araucaria angustifolia) has a higher diversity of simulating the removal of forest patches smaller than dispersal traits than exotic species (Pinus elliottii). This 50 ha in the Atlantic Forest cover in Brazil, Ribeiro shows the importance of eﬀective management of the et al. (2009) found an increase in the isolation of major protected area to allow native forest regeneration in forest patches and a reduction of their connectivity. planted forest stands, considering the existence of both The MPS in our study was equal to 19 ha, showing that native and exotic pine species. the landscape restoration depends on small fragments. 28 T. P. V. DE MATOS ET AL. Small forest fragments are important to promote the aiming at conserving protected areas, which in this structural and functional connectivity of larger frag- case is the CBNF. ments (Rubio & Saura, 2012). Forest fragments with Based on FSI, landscape planning can also support low FSI values contribute to restoring the landscape scenarios of biodiversity conservation actions, including and they can be prioritized in a second stage. restoration in the surrounding areas of the priority frag- Therefore, it is necessary to increase forest cover ments; ecological corridor design; and improvement in and habitat patch connectivity to reach the ecological the understory in the planted forest stands, aiming at processes restoration of the forest fragments in this improving the eﬀectiveness of the protected area. complex landscape mosaic. According to Ferraz et al. Future studies should be performed to verify the (2014), it is necessary to conserve old-growth forest eﬀectiveness of the proposed index using ﬁeld data of fragments and increase forest quality, as well as species richness and composition in the selected frag- increase forest cover. To do so, a combination of ments and in the forest restoration areas. The index actions can be implemented in the study area, includ- can be applied in other restoration planning schemes ing restoring the forest of surrounding areas of prior- worldwide, with diﬀerent types of protected areas and ity fragments (high values of FSI), implementing agricultural landscapes. ecological corridors and improving the understory The importance of the index is to identify priority diversity and regeneration of native species in the areas for implementing restoration actions to improve planted forest stands. This understory with native the restoration success in areas that present urgency of species in planted forests can provide important habi- vegetation recovery in order to better achieve restora- tat resources for wildlife species such as many birds tion targets. As the suitability of the protected areas and mammals in the Atlantic Forest (Brockerhoﬀ, depends on the surrounding landscape conservation, Jactel, Parrotta, & Ferraz, 2013). Improving the move- this index can help to set the priority areas for restora- ment of these animals will also increase plant species tion actions in a cost-eﬀective approach, increasing the dispersal in the landscape. chances of success of restoration projects. Thus, forest restoration planning based on an eco- logical landscape approach (Leite et al., 2013; Metzger Disclosure statement & Brancalion, 2013; Stanturf et al., 2014; Strassburg et al., 2019; Vettorazzi & Valente, 2016) can enable No potential conﬂict of interest was reported by the authors. reestablishing the landscape structure and ecological function of the focus area of restoration in the long Funding term. This work was supported by the FAPESP Fundação de Amparo à Pesquisa do Estado de São Paulo under Grant [number 13/16303-2]. 5. Conclusion The CBNF is a protected area of sustainable use which Data availability statement has a history of wood exploitation/extraction which nowadays is covered by planted forests of native The datasets used and/or analyzed during the current study are (Araucaria angustifolia) and exotic species (Pinus sp.). available from the corresponding author on reasonable Its surrounding area is predominantly occupied by request. agriculture, but we also observed a signiﬁcant area of planted Eucalyptus sp. forest. The native forest is scat- Geolocation information tered throughout the landscape and has a serious level of habitat fragmentation. Thus, there are a great num- Brazil ber of patches, with most of them being small in size and of irregular shape, suﬀering the inﬂuence of agri- References culture, pasturelands, planted forest stands, and roads. In this context, we proposed the Forest Suitability Almeida, F. S., Gomes, D. S., & Queiroz, J. M. (2011). Estratégicas para a conservação da biodiversidade Index (FSI) based on a landscape ecology approach to biológica em ﬂorestas fragmentadas. Ambiência, 7(2), prioritize forest patches that potentially support the 367–382. landscape restoration. Therefore, forest patches asso- Aronson, J., & Sasha, A. (2013). Ecosystem restoration is ciated with the highest FSI values have the best con- now a global priority: Time to roll up our sleeves. ditions to support an increase in biodiversity and Restoration Ecology, 21(3), 293–296. species dispersal throughout the landscape, thereby Bager, A., & Fontoura, V. (2013). Evaluation of the eﬀec- tiveness of a wildlife roadkill mitigation system in wet- facilitating natural restoration. Thus, we concluded land habitat. Ecological Engineering, 53,31–38. that FSI enables planning landscape restoration to Banks-Leite,C., Ewers,R.,Kapos,M.,&Metzger,J.P.(2013). increase functional connectivity and habitat quality, The confounded eﬀects of habitat disturbance at the local, GEOLOGY, ECOLOGY, AND LANDSCAPES 29 patch and landscape scale on understorey birds of the Convertino, M., Baker, K. M., Vogel, J. T., Lu, C., Suedel, B., Atlantic Forest: Implications for the development of & Linkov, I. (2013). Multi-criteria decision analysis to landscape-based indicators. Ecological Indicators, 31,82–88. select metrics for design and monitoring of sustainable Banks-Leite, C., Ewers, R. M., Kapos, V., Martensen, A. C., ecosystem restorations. Ecological Indicators, 26,76–86. & Metzger, J. P. (2011). Comparing species and measures De Jesus, E. N., Ferreira, R. A., Aragão, A. G., Santos, T. I. S., of landscape structure as indicators of conservation & Rocha, S. L. (2015). Structure of forest fragments of the importance. Journal of Applied Ecology, 48, 706–714. river basin Poxim-SE, as subsidy to ecological restoration. Baral, H., Keenan, R. J., Sharma, S. K., Stork, N. E., & Revista Árvore, 39(3), 467–474. Kasel, S. (2014). Spatial assessment and mapping of bio- Diniz, M., & Brito, D. (2015). Protected areas eﬀectiveness in diversity and conservation priorities in a heavily modiﬁed maintaining viable giant anteater (Myrmecophagatridactyla) and fragmented production landscape in north-central populations in an agricultural frontier. Brazilian Journal of Victoria, Australia. Ecological Indicators, 36, 552–562. Nature Conservation, 13, 143–151. Barber, C. P., Cochrane, M. A., Souza, C., Jr., & ENVI. Exelis Visual Information Solutions. Boulder: ITT Laurance, W. (2014). Roads, deforestation, and the miti- Visual Information Solutions. (2009). ENVI Version gating eﬀect of protected areas in the Amazon. Biological 4.7 August, 2009 Edition Copyright © ITT Visual Conservation, 177, 203–209. Information Solutions All Rights Reserved. Bechara, F. C. (2006). Unidades demonstrativas de Fagan, M. E., DeFries, R. S., Sesnie, S. E., Arroyo-Mora, P., & restauração ecológica através de técnicas nucleadoras: Chazdon, R. L. (2016). Targeted reforestation could reverse Floresta estacional semidecidual, Cerrado e Restinga declines in connectivity for understory birds in a tropical (Thesis (ph.D.)). Brasil: ESALQUSP, 248. habitat corridor. Ecological Applications, 26(5), 1456–1474. Bingham, H. C., Bignoli, D. J., Lewis, E., MacSharry, B., Fahrig, L. (2003). Eﬀects of habitat fragmentation on Burgess, N. D., Visconti, P., . . . Kingston, N. (2019). biodiversity. Annual Reviews in Ecology, Evolution and Sixty years of tracking conservation progress using the Systematics, 34, 487–515. world database on protected areas. Nature Ecology & Ferraz, S. F. B., Ferraz, K. M. P. M. B., Cassiano, C. C., Evolution, 3, 737–743. Brancalion, P. H., Da Luz, D. T. A., Azevedo, T. N., & Bortoleto, L. A., Figueira, C. J. M., Dunning, J. B., Jr, Metzger, J. P. (2014). How good are tropical forest Rodgers, J., & Silva, A. M. (2016). Suitability index for patches for ecosystem services provisioning? Landscape restoration in landscapes: An alternative proposal for Ecology, 29, 187–200. restoration projects. Ecological Indicators, 60, 724–735. Fonseca, C. R., Ganade, G., Baldissera, R., Becker, C. G., BRASIL. Lei federal nº. 12.651. de 25 de maio de 2012, Boelter, C. R., Brescovit, A. D., & Vieira, E. M. (2009). dispõe sobre a proteção da vegetação nativa; altera as Towards an ecologically-sustainable forestry in the Leis nos 6.938, de 31 de agosto de 1981, 9.393, de 19 de Atlantic Forest. Biological Conservation, Liverpool, 141, dezembro de 1996, e 11.428, de 22 de dezembro de 2006; 1209–1219. revoga as Leis nos 4.771, de 15 de setembro de 1965, Garcia, A. S., Sawakuchi, H. O., Ferreira, M. E., & e 7.754, de 14 de abril de 1989, e a Medida Provisória Ballester, M. V. R. (2017). Landscape changes in no 2.166-67, de 24 de agosto. a neotropical forest-savanna ecotone zone in central Bregman, T. P., Sekercioglu, C. H., & Tobias, J. A. (2014). Brazil: The role of protected areas in the maintenance of Global patterns and predictors of bird species responses to native vegetation. Journal of Environmental Management, forest fragmentation: Implication for ecosystem function 187,16–23. and conservation. Biological Conservation, 169,372–383. Geneletti, D. (2004). A GIS-based decision support system Brockerhoﬀ, E. G., Jactel, H., Parrotta, J. A., & Ferraz, S. F. B. to identify nature conservation priorities in an alpine (2013). Role of eucalypt and other planted forests in valley. Land Use Policy, 21(2), 149–160. biodiversity conservation and the provision of Honorato, R., Crouzeilles, R., Ferreira, M. S., & biodiversity-related ecosystem services. Forest Ecology Grelle, C. E. V. (2015). The eﬀects of habitat availability and Management, 301,43–50. and quality on small mammals abundance in the Carranza,M.L., Hoyos, L.,Frate,L., Acosta,A.T.R.,& Brazilian Atlantic Forest. Brazilia Journal of Nature Cabido, M. (2015). Measuring forest fragmentation Conservation, 13, 133–138. using multitemporal forest cover maps: Forest loss Instituto Chico Mendes De Conservação Da Biodiversidade and spatial pattern analysis in the Gran Chaco, (ICMBio). (2017). Plano de Manejo da FLONA de Capão Central Argentina. Landscape and Urban Planning, Bonito: Versão preliminar. Capão Bonito: ICMBIO 2017. 143,238–247. PLANO DE MANEJO DA FLORESTA NACIONAL DE Carrara, E., Arroyo-Rodríguez, V., Vega-Rivera, J. H., CAPÃO BONITO.São Paulo. Volume I. 230p Volume Schondube, J. E., de Freitas, S. M., & Fahrig, L. (2015). II.212p. Impact of landscape composition and conﬁguration on Jara, T., Hylander, K., & Nemomissa, S. (2017). Tree diver- forest specialist and generalist bird species in the frag- sity across diﬀerent tropical agricultural land use types. mented Lacandona rainforest, Mexico. Agriculture, Ecosystems and Environment, 240,92–100. BiologicalConservation, 184, 117–126. Joly, C. A., Metzger, J. P., & Tabarelli, M. (2014). Centro De Pesquisas Meteorológicas E Climáticas Aplicadas Experiences from the Brazilian Atlantic Forest: À Agricultura (CEPAGRI). (2013). Condições Ecological ﬁndings and conservation initiatives. New meteorológicas.Disponívelem http://www.cepagri.uni Phytologist, 204, 459–473. camp.br Landis, J. R., & Koch, G. G. (1977). The measurement of Chazdon, R. L. (2017). Tropical forest regeneration. observer agreement for categorical data. Biometrics, 33 Reference module in life sciences. Encyclopedia of (1), 159–174. Biodiversity, 2013, 277–286. Latawiec, A. E., Strassburg, B. B. N., Brancalion, P. H. S., Congalton, R. G., & Green, K. (1998). Assessing the accuracy Rodrigues, R. R., & Gardner, T. (2015). Creating space for of remotely sensed data: Principles and practices. large-scale restoration in tropical agricultural landscapes. New York: Lewis Publishers. Frontiers in Ecology Environment, 13(4), 211–218. 30 T. P. V. DE MATOS ET AL. Leite, M. S., Tambosi, L. R., Romitelli, I., & Metzger, J. P. Peixoto, C. A. B., & Theodorovicz, A. (2009). (2013). Landscape ecology perspective in restoration pro- Geodiversidade SP. BRASIL, C.-S. G. D. Brasília, jects for biodiversity conservation: A review. 2010 Peixoto, Carlos Augusto Brasil. Retrieved from Natureza&Conservação, 1(2), 108–118. http://www.cprm.gov.b r/publique/media/ Marcantonio, M., Rocchini, D., Geri, F., Bacaro, G., & Geodiversidade_SP.pdf Amici, V. (2013). Biodiversity, roads, & landscape frag- Peña-Domene, M., Minor, E. S., & Howe, H. F. (2016). mentation: Two Mediterranean cases. Applied Geography, Restored connectivity facilitates recruitment by an ende- 42,63–72. mic large-seeded tree in a fragmented tropical landscape. Matos, F. A. R., Magnago, L. F. S., Gastauer, M., Ecology, 97(9), 2511–2517. Carreiras, J. M. B., Simonelli, M., Meira-Neto, J. A. A., Périco, E., & Cemin, G. (2006). Caracterização da paisagem & Edwards, D. P. (2017). Eﬀects of landscape conﬁgura- do município de Arvorezinha, RS, com ênfase na tion and composition on phylogenetic diversity of trees in dinâmica dos fragmentos ﬂorestais, por meio de sistemas a highly fragmented tropical forest. Journal of Ecology, de informações geográﬁcas (SIGs). ScientiaForestalis, 70 105, 265–276. (1), 9–21. McGarigal, K., SA Cushman, and E Ene. 2012. FRAGSTATS Ponte, E., Roch, M., Leinenkugel, P., Dech, S., & Kuenzer, C. v4: Spatial Pattern Analysis Program for Categorical and (2017). Paraguay´s Atlantic Forest cover loss – Satellite- Continuous Maps. Computer software program pro- based change detection and fragmentation analysis duced by the authors at the University of Massachusetts, between 2003 and 2013. AppliedGeography, 79,37–49. Amherst. Available at the following web site: http://www. R Core Team. R. (2013). A language and environment for umass.edu/landeco/research/fragstats/fragstats.html. statistical computing. Vienna: R Foundation for Statistical Mello, K., Toppa, R. H., & Abessa, D. M. (2011). Dynamics Computing. of the landscape of the city of Cubatão: Growth among Rezende, C. L., Scarano, F. R., Assad, E. D., Joly, C. A., shores, industry facilities and the mountain range Serra Metzger, J. P., Strassburg, B. B. N., . . . do Mar. O Mundo Da Saúde, 35,42–46. Mittermeier, R. A. (2018). From hotspot to hopespot: Mello, K., Toppa, R. H., & Cardoso-Leite, E. (2016). Priority An opportunity for the Brazilian Atlantic Forest. areas for forest conservation in an urban landscape at the Perspectives in Ecology and Conservation, 16(4), 208–214. transition between Atlantic forest and Cerrado. CERNE Ribeiro, C. R., Metzger, J. P., Martensen, A. C., [Online], 22(3), ISSN 0104-7760, 277–288. Ponzoni, F. J., & Hirota, M. M. (2009). The Brazilian Mendonça-Lima, A., Duarte, L. S., & Hartz, S. M. (2014). Atlantic Forest: How much is left, and how is the remain- Comparing diversity and dispersal traits of tree commu- ing forest distributed? Implications for conservation. nities in plantations and native forests in Southern Brazil. BiologicalConservation, 142, 1141–1153. Brazilian Journal of Nature Conservation, 12(1), 24–29. Rocha-Santos, L.; Pessoa, M. S.; Cassano, C. R.; Talora, D. Metzger, J. P., Bernacci, L. C., & Goldenberg, R. (1997). C.; Orihuela, R. L.L.; Mariano-Neto, E.; Morante-Filho, J. Pattern of tree species diversity in riparian forest frag- C.; Faria, D.; Cazetta, E. .2016. The shrinkage of a forest: ments of diﬀerent widths (SE Brazil). Plant Ecology, 133 Landscape-scale deforestation leading to overall changes (2), 135–152. in local forest structure. Biological Conservation, v. 196, Metzger, J. P., & Brancalion, H. S. (2013). Challenges and p. 1-9. opportunities in applying a landscape ecology perspective Ross, J. L. S., & Moroz, I. (1997). Mapa geomorfológico in ecological restoration: A powerful approach to shape do estado de São Paulo - Escala 1:500.000 (pp. 64). São neolandscapes. Natureza&Conservação, 11(2), 103–107. Paulo: Departamento de Geograﬁa-FFLCH-USP/IPT/ Milan, E., & Moro, R. S. (2012). Padrões de fragmentação FAPESP. ﬂorestal natural no parque estadual de Vila Velha, Ponta Rubio, L., & Saura, S. (2012). Assessing the importance of Grossa (PR). Ambiência, 8(5), 685–697. individual habitat patches as irreplaceable connectivity Molin, P. G., Chazdon, R., Ferraz, S. F. B., & providers: And analysis of simulated and real landscape Brancalion, P. H. S. (2018). A landscape approach for data. Ecological Complexity, 11,28–37. cost-eﬀective large-scale forest restoration. Journal of Rudnick, D., Beier, P., Cushman, S., Dieﬀenbach, F., Applied Ecology, 55, 2767–2778. Epps, C. W., Gerber, L., & Trombulak, S. C. (2012). The Moraes, M. C. P., Mello, K., & Toppa, R. H. (2017). role of landscape connectivity in planning and implement- Protected areas and agricultural expansion: Biodiversity ing conservation and restoration Priorities. Issues in conservation versus economic growth in the Southeast of ecology Report No. 16. Ecological Society of America. Brazil. Journal of Environmental Management, 188, Washington, DC. 73–84. Santos Junior, P. C. A., Marques, F. C., Lima, M. R., & Dos Newman, M. E., McLaren, K. P., & Wilson, B. S. (2014). Anjos, L. (2016). The importance of restoration areas to Assessing deforestation and fragmentation in a tropical conserve Bird species in a highly fragmented Atlantic moist forest over 68 years; the impact of roads and legal forest landscape. Brazilian Journal of Nature protection in the Cockpit Country, Jamaica. Forest Conservation, 14,1–7. Ecology and Management, 315, 138–152. Santos, R. S., Ribeiro, C. A. A. S., Peluzio, T. M., Nichol, J. E., Abbas, S., & Fischer, G. A. (2017). Spatial Peluzio, J. B. E., Queiroz, V. T., Branco, E. R., & patterns of degraded tropical forest and biodiversity Almeida, S. L. H. (2016). Geotechnology and landscape restoration over 70-years of succession. Global Ecology ecology applied to the selection of potential forest frag- and Conservation, 11, 134–145. ments for seed harvesting. Journal of Environmental Oakleaf, J. R., Matsumoto, M., Kennedy, C. M., Management, 183, 1050–1063. Baumgarten, L., Miteva, D., Sochi, K., & Kiesecker, J. Saura, S., Bertzky, B., Bastin, L., Battistella, L., Mandrici, A., (2017). LegalGEO: Conservation tool to guide the siting & Dubois, G. (2018). Protected area connectivity: of legal reserves under the Brazilian Forest Code. Applied Shortfalls in global targets and country-level priorities. Geography, 86,53–65. Biological Conservation, 219,53–67. GEOLOGY, ECOLOGY, AND LANDSCAPES 31 Sekercioglu, C. H., Loarie, S., Oviedo-Brenes, F., Uezu,A.,&Metzger,J.P.(2016). Time-lag in Mendenhall,C.,Daily, G.R.,& Ehrlich,P.(2015). Tropical responses of birds to Atlantic Forest fragmentation: countryside riparian corridors provide critical habitat and Restoration opportunity and urgency. PloS One, 11 connectivity for seed-dispersing forest birds in a fragmented (1), e0147909. landscape. Journal of Ornithology, 56(S1), 343–353. Valente, R. O. A., & Vettorazzi, C. A. (2005). Comparação Silva, P.H.M.,Poggiani,F.,Sebbenn,A. M.,&Mori,E.S. entre métodos de avaliação multicriterial, em ambiente (2011). Can Eucalyptus invade native forest fragments close SIG, para a conservação e a preservação ﬂorestal. Scientia to commercial stands? Forest Ecology and Management, 261, Forestalis, 69,51–61. 2075–2080. Valente, R. O. A., & Vettorazzi, C. A. (2008). Deﬁnition of Singh, S., & Mishra, A. (2014). Spatiotemporal analysis of priority areas for forest conservation through the ordered the eﬀects of forest covers on stream water quality in weighted averaging method. Forest Ecology and Western Ghats of peninsular India. Journal of Management, 256(6), 1408–1417. Hydrology, 519, 214–224. Veloso, H.P.(2012). Sistema Fitogeográﬁco. In Instituto Slade, E. M., Merckx, T., Riutta, T., Bebber, D. P., Redhead, D., Brasileiro De Geograﬁa E Estatística. Manual Técnico Riordan, P., & Macdonald, D. W. (2013). Life-history traits da Vegetação Brasileira (2., pp. 9–38). Rio de Janeiro: and landscape characteristics predict macro-moth responses IBGE. to forest fragmentation. Ecology, 94(7), 1519–1530. Vettorazzi, C. A., & Valente, R. A. (2016). Priority areas for Stanturf, J. A., Palik, B. J., & Dumroese, R. K. (2014). forest restoration aiming at the conservation of water Contemporary forest restoration: A review emphasizing resources. EcologicalEngineering, 94, 255–267. function. Forest Ecology and Management, 331, 292–323. Zamorano-Elgueta, C., Benayas, J. M. R., Cayuela, L., Strassburg, B. B. N., Beyer, H. L., Crouzeilles, R., Hantson, S., & Armenteras, D. (2015). Native forest Iribarrem, A., Barros, F., Siqueira, M. F., . . . Uriarte, M. replacement by exotic plantations in southern Chile (2019). Strategic approaches to restoring ecosystems can (1985–2011) and partial compensation by natural triple conservation gains and halve costs. Nature Ecology regeneration. Forest Ecology and Management, 345, & Evolution, 3,62–70. 10–20. Tambosi, L., Martensen, A., Ribeiro, M., & Metzger, J. Zhang, C., & Fu, S. (2009). Allelopathic eﬀects of Eucalyptus (2014). A framework to optimize biodiversity restoration and the establishment of mixed stands of Eucalyptus and eﬀorts based on habitat amount and landscape native species. Forest Ecology and Management, 258(7), connectivity. RestorationEcology, 22, 169–177. 1391–1396. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Geology Ecology and Landscapes Taylor & Francis http://www.deepdyve.com/lp/taylor-francis/protected-areas-and-forest-fragmentation-sustainability-index-for-r1ZeAlHpbm

Loading next page...

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher: Taylor & Francis
Copyright: © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group on behalf of the International Water, Air & Soil Conservation Society(INWASCON).
ISSN: 2474-9508
DOI: 10.1080/24749508.2019.1696266
Publisher site: See Article on Publisher Site

Abstract

GEOLOGY, ECOLOGY, AND LANDSCAPES 2021, VOL. 5, NO. 1, 19–31 INWASCON https://doi.org/10.1080/24749508.2019.1696266 RESEARCH ARTICLE Protected areas and forest fragmentation: sustainability index for prioritizing fragments for landscape restoration a a b Tatiana Possati Vieira De Matos , Veridiana Possati Vieira De Matos , Kaline De Mello and Roberta Averna Valente Programa de Pós-graduação em Diversidade Biológica e Conservação, Universidade Federal de São Carlos, campus Sorocaba – UFSCar, b c Sorocaba, Brazil; Department of Ecology, Institute of Biosciences, University of São Paulo. R. do Matão, São Paulo, Brazil; Departamento de Ciências Ambientais, Universidade Federal de São Carlos, campus Sorocaba – UFSCAR, Sorocaba, Brazil ABSTRACT ARTICLE HISTORY Received 21 May 2019 Tropical forests are threatened due to forest clearing and fragmentation which lead to Accepted 16 October 2019 a decrease in forest cover area and landscape connectivity, while also increasing edge eﬀects. These impacts aﬀect biodiversity and ecosystem service provision which are essential for KEYWORDS humanity’s well-being. Protected areas are created to minimize these eﬀects on biodiversity, Landscape connectivity; but many of them are threatened due to forest fragmentation in the surrounding areas. Thus, forest conservation; atlantic forest restoration is needed to ensure the suitability of protected areas in the landscape. forest; land use; buﬀer zone However, restoration planning needs to adopt an ecological landscape approach to ensure the recovery of the biodiversity and the ecological processes. This study developed a forest sustainability index (FSI) which represents the forest patch potential in facilitating landscape restoration in a protected area and its surroundings. A land-use/land-cover map was used to calculate landscape metrics at two levels: landscape and patch. Landscape metrics were subsequently selected to create the index. The tested landscape presents a great number of fragments with most of them being small in size and having irregular shapes. The focus areas for forest restoration are located close to forest fragments with higher FSI values, as these can facilitate natural restoration and guarantee the maintenance of the ecosystem processes. 1. Introduction such as exotic species introduction and other serious impacts on species diversity and composition, commu- Tropical forests have been replaced by land uses lead- nity structure and dynamics, and ecosystem function- ing to deforestation and forest fragmentation, which is ing (Carrara et al., 2015; Joly, Metzger, & Tabarelli, the main cause of biodiversity loss and aﬀects natural 2014; Santos et al., 2016; Slade et al., 2013). resource and ecosystem service provisioning including Furthermore, these impacts also aﬀect the ecosystem wood, drinking water, and soil protection. The ability to recover from disturbances, i.e. its resilience to Atlantic Forest in particular is a biodiversity hotspot anthropogenic changes (Bregman, Sekercioglu, & (Rezende et al., 2018) which is highly threatened since Tobias, 2014; Sekercioglu et al., 2015). it has been reduced to 28% of its original cover mainly In the current global scenario of environmental consisting of small fragments and edge-aﬀected or degradation, protected areas are the cornerstone of secondary vegetation disconnected from larger frag- eﬀorts to sustain biodiversity and natural ecosystem ments (Rezende et al., 2018; Ribeiro, Metzger, processes (Bingham et al., 2019). However, they are Martensen, Ponzoni, & Hirota, 2009). According to threatened due to the forest fragmentation in the sur- Ferraz et al. (2014), the Atlantic forest cover is cur- rounding areas. The protected area and its buﬀer zone rently a heterogeneous forest mosaic of diﬀerent ages must be well managed to provide habitat quality for and at many degrees of conservation and located in the species and to decrease neighborhood eﬀects diﬀerent landscape conditions within a complex (Moraes, Mello, & Toppa, 2017; Saura et al., 2018). matrix composed by agriculture and other uses. This For example, planted forests can contribute to species scenario occurs worldwide in most of the tropical conservation if they keep a diverse and complex landscapes with forest remnants (Chazdon, 2017; understory (Fonseca et al., 2009). Singh & Mishra, 2014). Therefore, forest restoration can play an important Forest fragmentation decreases the forest cover area role in mitigating the forest fragmentation impacts on and the landscape connectivity, thereby increasing the the natural ecosystems and their services (Geneletti, edge eﬀects which lead to physical and biotic alterations 2004; Leite, Tambosi, Romitelli, & Metzger, 2013). associated with the artiﬁcial boundaries of fragments Forest restoration must be directed to restore the CONTACT Tatiana Possati Vieira de Matos veri_gemeas_@hotmail.com Programa de Pós-graduação em Diversidade Biológica e Conservação, Universidade Federal de São Carlos, campus Sorocaba – UFSCAR, Rod. João Leme dos Santos, km 110, Sorocaba, Brasil © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group on behalf of the International Water, Air & Soil Conservation Society(INWASCON). This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 20 T. P. V. DE MATOS ET AL. landscape structure under the perspective of the land- restoration in a protected area and its surroundings, scape ecology in order to ensure biodiversity recovery and which is located in a complex agricultural matrix. in diﬀerent taxa and the ecological processes and ser- The speciﬁc objectives were: (1) to assess the landscape vices (Leite et al., 2013; Metzger & Brancalion, 2013; structure based on landscape metrics; (2) to select metrics Santos Junior, Marques, Lima, & Dos Anjos, 2016; to create the forest sustainability index; and (3) to apply Stanturf, Palik, & Dumroese, 2014; Vettorazzi & the index to identify priority forest fragments that Valente, 2016). Thus, forest restoration based on eco- enhance landscape restoration in the protected area and logical criteria can improve the forest structure and its surroundings. landscape connectivity, promoting the connection The creation of this index based on landscape between the protected areas and other forest metrics assists in diagnosing sites which still have fragments. a minimum of resilience with high restoration poten- Ecological criteria can be represented by the land- tial, thus favoring an increase in biodiversity and scape characteristics that inﬂuence the restoration pro- maintaining ecological processes and services. In this cess, including the species richness, establishment, way, the index will facilitate implementing restoration dispersal and interrelations (Banks-Leite, Ewers, projects, increasing environmental gains and decreas- Kapos, Martensen, & Metzger, 2011; Santos Junior ing costs, potentiating the natural regeneration. et al., 2016). Recent studies have shown that forest restoration plans in landscapes with minimum resili- ence, i.e. which still present connection between forest 2. Material and methods fragments, have more success and mitigate costs better 2.1 Study area than others on highly degraded landscapes (Bortoleto, Figueira, Dunning, Rodgers, & Silva, 2016; Molin, The study area is Capão Bonito National Forest (CBNF) Chazdon, Ferraz, & Brancalion, 2018;Strassburg et al., and its surrounding area, comprising a total area of 2019; Tambosi, Martensen, Ribeiro, & Metzger, 2014). 74,575 ha, wherein the CBNF represent 4,773.83 ha Thus, assessing the landscape pattern is the ﬁrst and the surrounding area comprises 69,790.65 ha. step to identify potential areas for forest restoration Considering that the CBMF Management Plan is in plans (Aronson & Sasha, 2013; De Jesus, Ferreira, elaboration because it will be supported by this work Aragão, Santos, & Rocha, 2015; Metzger & (among others), we considered a surrounding area of Brancalion, 2013; Moraes et al., 2017; Strassburg 10 km from the perimeter protected area as a primary et al., 2019; Valente & Vettorazzi, 2008). Landscape study The CBNF is located in the Alto de metrics can quantify and describe the landscape com- Paranapanema Watershed, São Paulo state, Southeast ponents (matrix, fragments, and corridors) and their Brazil (Figure 1), in the Atlantic Forest Biome. National spatial distribution, minimizing the uncertainties of forest is a category of sustainable use protected area in the landscape analysis (Convertino et al., 2013). Brazil, where the primary purpose is sustainable exploi- The combination of various landscape metrics can tation of the forest and its natural resources. be used to build important indexes which are currently The CBNF is currently covered by native vegetation applied for deﬁning priority areas for forest conserva- and stands of exotic forest species (originally commer- tion and restoration (Bortoleto et al., 2016; Mello, cial stands) which are related to the history of the area. Toppa, & Cardoso-Leite, 2016; Valente & Vettorazzi, They have been replaced by natural vegetation by the 2008; Vettorazzi & Valente, 2016). Mello et al. (2016) management of the protected area. Even with the and Bortoleto et al. (2016) developed indexes for presence of exotic species, the area was deﬁned as priority areas for forest conservation and restoration, a protected area by the federal government because it respectively, in the same area in Southern Brazil, while is an ecological tension area between two important Vettorazzi and Valente (2016) used several landscape Biomes: Cerrado and Atlantic Forest. Thus, the area criteria to identify priority areas for forest restoration presents forest remnants of savanna, mixed ombro- aiming toward water resource conservation. philous forest, and semi-deciduous seasonal forest. In Moreover, Santos et al. (2016) selected potential forest addition, there are Araucaria angustifolia plantations fragments for seed harvesting based on landscape (endangered species according to IUCN) which are metrics. However, potential forest fragments which already considered an important genetic bank for cur- contribute to the success of restoration actions in rent and future use. The exotic species such as Pinus a landscape are still needed to be identiﬁed, among elliottii in these same areas have supported scientiﬁc which not only aim for local restoration, but also experiments (Bechara, 2006; Veloso, 2012). consider a landscape restoration perspective (Molin CBNF was a large farm originally named “Itanguá”, et al., 2018; Vettorazzi & Valente, 2016). which in 1944 was categorized as Forest Park for wood In this context, the main objective of this study was to production. The plantation of Araucaria angustifolia develop a forest sustainability index that represents the has started in 1945. In the 1950s, the Pinus elliotti potential of a forest fragment to facilitate landscape plantations have expanded to reach 2,400 ha, and in GEOLOGY, ECOLOGY, AND LANDSCAPES 21 Figure 1. Capão Bonito National Forest and its surrounding 10 km-buﬀer location, in the São Paulo state, Southeast Brazil. 1953 the farm was denominated as Forest Park 1997). Most of the soil types are Latosols and Deep “Getúlio Vargas”. The genus Pinus was introduced in sandy-quartz soils (Peixoto & Theodorovicz, 2009). the areas to conduct experiments due to the large-scale reforestation policy adopted in Brazil at this time, 2.2 Landscape structure which used Pinus spp. to restore deforested areas. The area ﬁnally became a National Forest in 1968 A land-use/land-cover map was produced by super- and it started to be managed by the national agency vised digital classiﬁcation to assess the landscape of protected areas – Chico Mendes Institute for structure. For this, SPOT satellite images Biodiversity Conservation (Instituto Chico Mendes (2,5 m-spatial resolution; spectral bands: blue, green, De Conservação Da Biodiversidade [ICMBio], 2017). red and near-infrared; year: 2009) supplied by the Because National Forests area sustainable use Environment Secretariat of the State of São Paulo, protected areas, planted forests of exotic species Environmental Planning Coordination (SMA-CPLA) such as Pinus spp. is allowed, but it must follow were used. The images were re-projected in UTM some regulationssuchaskeeping native forest 23S – SIRGAS 2000, as it was for all data set. The along the river banks and around springs (Brasil, software ENVI (ENVI – Itt Visual Information 2012). Studies in the CBNF are of great relevance Solutions 2009) was used for the geometric and atmo- due to the rich understory of native vegetation spheric corrections of the spectral bands and for the within the forest plantation stands and the several image classiﬁcation by using Maximum Likelihood environments which oﬀer varied resources for ani- algorithm. The following land-use/land-cover types mal and plant species. were considered: The surrounding area of the CBNF is a complex matrix composed by various land uses, including agri- (a) Native forest: native forest fragments; culture in small and large scale, pasture, and commercial (b) Planted Pinus sp. forest (Pinus): Pinus sp. forest plantations. There is also a highway which sepa- stands with various levels of understory rates the protected area into two parts with diﬀerent sizes regeneration; (Figure 1). Its proximity to other large-protected areas (c) Planted Eucalyptus sp. forest (Eucalyptus): (Intervales State Park in the South, Carlos Botelho State Eucalyptus sp. stands managed in diﬀerent Park in the Southeast and Angatuba Ecological Station in periods; the northwest) provides importance for increasing land- (d) Planted Araucaria angustifolia forest scape connectivity through ecological restoration. (Araucaria): Araucaria angustifolia stands The region is under the inﬂuence of a Cwa climate with various levels of understory regeneration; (subtropical-dry winter) and annual precipitation is (e) Agriculture: soybean, wheat, corn and citrus above 22°C (Centro De Pesquisas Meteorológicas crops; E Climáticas Aplicadas À Agricultura [CEPAGRI], (f) Pasture: grassland destined to intensive and 2013). The average elevation is 700 m above the sea extensive cattle production systems, or without level, with slightly undulated relief (Ross & Moroz, cattle; 22 T. P. V. DE MATOS ET AL. (g) Urban areas: urban patches located in the sur- Thus, we developed the FSI to prioritize forest frag- rounding area of the CBNF; ments for landscape restoration (Eq. 1). (h) Road network: roads and highways in the study FSI ¼ IxAREA þ IxSHAPE þ IxENN (1) n n n area; (i) Rivers network: rivers, streams, and springs in Where, the study area. I: importance of the metric to the landscape restoration; AREA : normalized AREA value (scale 1 to River networks, road networks, and urban areas 7bytes); SHAPE : normalized SHAPE value (scale 1 to 7 were obtained by on-screen digitizing of the SPOT bytes); ENN : normalized ENN value (scale 1 to 7 bytes). images. These landscape metrics that compose the FSI were A confusion matrix and the Kappa coeﬃcient based on the following concepts: (Congalton & Green, 1998) were used to assess accu- racy and agreement, based on 80 ground control (a) AREA: this is one of the most important pieces points distributed stratiﬁed randomly along the study of information of the forest fragment which area. We obtained a global accuracy of 93.41% and an inﬂuences other characteristics of the forest accuracy of 99.14% for the native forest. The map was fragments such as perimeter and core area. considered very good according to Landis and Kock The greatest forest fragments in the landscape (1977) based on the kappa value (0.89). present the largest core area and they present Landscape metrics were used to describe the land- higher natural resources and have greater bio- scape structure (Online Resource 1) calculated for diversity than small forest patches in most cases landscape level (land-use/land-cover class) and for (Fahrig, 2003; Nichol, Abbas, & Fischer, 2017; forest patches level through the extension V-Late 2.0 Rocha-Santos et al., 2016). Metzger, Bernacci, in the software ArcGIS 10.1. We chose these metrics and Goldenberg (1997) pointed out that the based on their contents and their application to eco- minimum forest patch size for its ﬂoristic sta- logical processes aiming at their importance for deﬁn- bility in the semi-deciduous seasonal forest is ing priority areas for landscape restoration (Baral, 25 ha. Keenan, Sharma, Stork, & Kasel, 2014; Oakleaf et al., (b) SHAPE: the shape of a habitat patch can be 2017; Vettorazzi & Valente, 2016). Forest patches used to represent the edge eﬀect on the forest smaller than 1 ha were not considered in the analysis, fragment. Forest fragments with irregular according to (Valente and Vettorazzi (2008)). shapes suﬀer more inﬂuence of the edge eﬀect, A 30 m-edge eﬀect was used to calculate CORE and and thus they are under a greater inﬂuence of TCAI metrics, as proposed by Périco and Cemin the matrix. The edge-dominated Atlantic forest (2006), and Milan and Moro (2012) for semi- fragments move towards early-successional deciduous seasonal forest remnants. Boxplot graphics systems, compromising the forest fragment were performed in the software RCore Team 3.0.2 sustainability (Joly et al., 2014). However, (2013) to describe the landscape pattern. large forest fragments in highly fragmented landscapes tend to have more irregular shape than small patches (Mello et al., 2016). 2.3 Forest suitability index (c) ENN: this indicates the distance of the nearest Some tests were performed to build the forest suitabil- neighbor related to the degree of isolation of ity index (FSI) after obtaining data for the whole land- the forest fragments (McGarigal, 2015). The scape. We performed these tests and accessed previous connectivity among the forest fragments is prioritization maps using the landscape metrics in important to the forest restoration under addition to literature consultation to develop the a landscape-based approach (Molin et al., index. We selected uncorrelated metrics representing 2018). The landscape connectivity will improve characteristics of forest fragments which are impor- the species dispersion and recruitment capa- tant for biodiversity conservation. In addition to the city, which are related to the forest restoration forest patch size, its isolation level, connectivity, and success (Peña-Domene, Minor, & Howe, 2016). its quality inﬂuence the persistence of species in tro- Landscape connectivity analysis is a valuable pical agricultural landscapes (Banks-Leite, Ewers, criterion for prioritizing restoration opportu- Kapos, & Metzger, 2013; Uezu & Metzger, 2016). nities, helping to identify areas which hold the Next, we tested the combination of the following greatest potential for increasing connectivity metrics: AREA, SHAPE, CORE, and ENN. We calcu- and the natural integrity of the landscape lated the correlation among the metrics to avoid the (Rudnick et al., 2012). use of similar metrics leading to wrong results and interpretations. The metrics combination which best After selecting the metrics set, we then performed represented our goal was: AREA, SHAPE, and ENN. tests to deﬁne the importance for each metric. When GEOLOGY, ECOLOGY, AND LANDSCAPES 23 the same value was attributed for SHAPE and ENN, and protected area. We elaborated a map of the forest closetothe valuefor AREA (AREA40%,SHAPE 30%, fragments classiﬁed according to the FSI values. and ENN 30%), we observed that larger and more con- nected forest fragments showed lower FSI than small and 3. Results irregular fragments (Figure 2). Thus, the index did not represent the objective. Thus, we performed tests to The land-use/land-cover map (Figure 3) presented obtain values which represent forest fragments that do a global classiﬁcation accuracy of 99.14%, which nothavealargeedge eﬀect and can function as source means that the map has a high agreement with the areas for landscape restoration over time. ground reality. This map indicated that the landscape We considered a relative value (total of 100%) of comprised by the CBNF and its surrounding area is importance (I) for each landscape metric according to predominantly agricultural (Table 2). their importance for the objective. Thus, a value of Agricultural areas such as corn, soy, and wheat 60% was assigned for AREA, 10% for SHAPE and 30% crops represent 40% (CA = 40) of the total landscape for ENN. (Table 2), mainly out of the protected area (Figure 2). The metrics values were normalized (1–7bytes- It showed the greatest MPS value (Table 2) among the scale) to avoid the use of diﬀerent units or scales to land use/land cover types. The largest agricultural calculate FSI, as shown in Table 1. patch has 1832 ha (Figure 4(a)), which is also the greatest patch in the landscape. Another characteristic that is a consequence of its predominance in the land- 2.4 Forest suitability index applied to the capão scape is the low ENNM value (45 m – Table 2), the bonito national forest same value of native forest, showing that both are very We calculated the FSI for forest fragments in the scattered over the landscape, but agriculture is predo- CBNF and its surrounding area to identify the impor- minant (Figure 3). tant fragments in the landscape for restoration actions Native forest was the second major land cover/land in order to improve the landscape connectivity for this use class in the study area, representing 24.45% of the Figure 2. Tests for the deﬁnition of importance percentage (I) for each metric (AREA 40%, SHAPE 30% and ENN 30%). 24 T. P. V. DE MATOS ET AL. Table 1. Normalized values for the landscape metrics. (11.06%) and Pinus sp. (6.71%) plantations. Shape Normalized scale However, the most important land use/land cover Area (ha) (dimensionless) ENN (m) (bytes) classes in the protected areas are Araucaria and < 5 >7 >36 1 Pinus plantations, especially in the largest area. The 5 – 25 6 −730 – 36 2 25 – 45 5 –624 – 30 3 CBNF is divided by a state road into two parts of 45 – 65 4 –518 – 24 4 diﬀerent sizes (Figure 3). The smallest portion pre- 65 – 85 3 –412 – 18 5 85 – 115 2 –36 – 12 6 sents some agricultural areas. >115 1 –2<6 7 On the other hand, the Eucalyptus plantations are mainly located in its surroundings in the east of the CBNF (Figure 3), and they belong to a paper landscape (Table 2). However, the remaining forest and pulp company. It represents the second land patches are interspersed with patches cleared for agri- use/land cover class with the highest NP value culture, planted forest or other uses, comprising (730) behind the native forest (Table 2). It showed a complex mosaic. It showed the greatest NP (941 the lowest MPS value (11.26 ha); however, the patches) and MSI (3.17) values among the land use/ patch sizes range from 1 ha to 677 ha (Figure 3). land cover classes, indicating the habitat fragmenta- In contrast, Araucaria and Pinus presented a MPS tion. It also presented a lower TCAI value than planted value of 28 and 29 ha, and the largest patches forests (Table 2) as a consequence of the forest frag- showed values of 323 and 224 ha, respectively mentation. MPS was 19 ha, however, its patch sizes (Figure 4(a)). varied from 1 ha to 1800 ha (Figure 4(a)), where the The three types of planted forest presented similar largest forest fragments are in the east of the study area TCAI (74.36%; 72.07% and 76.13%) and MSI (1.60; (Figure 3). Although it presented the same ENNM 1.90 and 1.90) values, i.e. they are similar regarding value of agriculture, the ENN values varied from their shapes and edge proportion. Planted forest 2.5 m to 604.78 m, which was the lowest range showed less complex shapes than the other land among the land use/land cover classes considering cover/land use classes (MSI < 2 – Table 2, Figure 3), this landscape metric (Figure 4(d)). which can be a result of the plantation design. They It is important to highlight that the native forest is presented a greater ENNM value than native forest not representative in the protected area. Its largest and agriculture of 45 m (Araucaria = 146.11 m; remnants are in the surrounding area, close to the Eucalyptus = 99.38 m and Pinus = 70.21 m). This Eucalyptus crops. Many forest fragments are divided shows that planted forests are concentrated in some into smaller patches by roads (Figure 3). The road portions of the study area, occurring in low density in network represents 2.55% of the landscape. the north and south (Figure 3). Thus, the greatest The third-largest land use/land cover class is distance of an Araucaria patch from another similar planted forest, accounting for 20.62% distributed patch was 2,900 m, while it was about 3,300 m for into Araucaria angustifolia (2.85%), Eucalyptus sp. Eucalyptus and Pinus (Table 2). Figure 3. Land use/land cover of the Capão Bonito National Forest (CBNF) and its surrounding 10 km-buﬀer in the State of São Paulo, Southeast Brazil. GEOLOGY, ECOLOGY, AND LANDSCAPES 25 Table 2. Landscape metrics for class level. Landscapemetrics Land use/land cover class CA (ha) CA (%) NP MPS (ha) MSI TCAI (%) ENNM (m) Agriculture 29,259.68 40.00 496 58.99 2.44 – 45.36 Araucaria 2116.87 2.85 76 27.85 1.90 76.13 146.11 Eucalyptus 8218.76 11.06 730 11.26 1.60 74.36 99.38 Pinus 4991.14 6.71 172 29.02 1.90 72.07 70.21 Pasture 8005.59 10.77 287 27.89 2.36 – 98.83 Nativeforest 18,173.32 24.45 941 19.31 3.17 54.49 44.68 CA: Total area; NP: Number of patches; MPS: Mean patch size; MSI: Mean shape index; TCAI: Total core area index; ENNM: Mean distance from the nearest neighboring patch. Figure 4. Boxplot of the variation of the landscape metrics: (a) patch size (PS), (b) shape index (SHAPE), (c) core area index (CA) and (d) distance from the nearest neighboring patch (ENN). Pasturelands represented 10.77% of the study The FSI application showed that important fragments area (Table 2) and they are concentrated in the for landscape restoration are large patches connected by south of the CBNF surroundings (Figure 3). They other forest patches in the landscape. As the forest cover are distributed into 287 patches with a MPS value is mostly comprised by forest patches with complex around 28 ha, close to the Araucaria and Pinus shapes, this metric received an importance value of values. The MSI value (2.36) was close to the agri- 10%. Thus, this metric has less impact than size and culture value, ranging from 1.29 to 6.15 (Figure 3), connectivity, but can still diﬀerentiate fragments with which represents more irregular shapes than similar importance only based on the ﬁrst two metrics planted forest. The ENNM value was similar to (Online Resource 1). The results also indicated that the the planted forest distances (99 m), showing that priority forest fragments for landscape restoration are the pasturelands are not well spread within the located in the CBNF surrounding area (Figure 5). landscape as agriculture and native forest. The forest fragments with the greatest FSI values are Urban areas represented less than 1% of the land- the largest patches (103 to 1176 ha) located in the scape. They are located in the northwest of the area, southeast, northeast, and northwest of the study area comprised by two little tows (Buri and Taquarivaí), (Figure 5). They represent strategic source areas in the which together have a population of 23,000. Urban landscape for forest restoration actions in their sur- areas, roads, and rivers represent, respectively, 0.93%, roundings. They are connected through other forest 2.55% and 1.31% of the total area. patches which showed lower FSI values, showing that 26 T. P. V. DE MATOS ET AL. Figure 5. Priority forest fragments for landscape restoration, according to the Forest Sustainability Index (FSI) in the Capão Bonito National Forest (CBNF) and its surrounding 10 km buﬀer in the State of São Paulo, Southeast Brazil. they are part of a complex landscape sustainability location, low slopes and not far from the coastal zone system even though they have less importance for land- (Mello et al., 2016; Moraes et al., 2017). scape restoration. The southern region above the pro- The landscape structure analysis showed that forest tected area has the forest fragments with the lowest FSI only represents 24.5% of the study area and it is highly values. However, they are important for the landscape fragmented, corroborated by the low core area and restoration regarding the local context of agriculture mean patch size values, and the high number of and pasture predominance. They presented small size patches (Table 2). Similar results were found by sev- (<5 ha), high ENN (up to 600 m) and irregular shapes eral studies conducted in the tropics, where natural (values ranging from 1.52 to 9.89). areas were replaced by agricultural crops (Carranza, Hoyos, Frate, Acosta, & Cabido, 2015; Moraes et al., 2017; Newman, McLaren, & Wilson, 2014; Ponte et al., 2017; Valente & Vettorazzi, 2005). The forest frag- 4. Discussion ments in our study are scattered within the agricul- The protected area surroundings are predominantly tural matrix (Figure 3), which can mainly aﬀect animal agricultural, presenting the greatest patches and very species that cannot transpose the open matrix; hence, scattered over the landscape (Figure 3, Table 2). Thus, the dispersal of some plant species is aﬀected it creates an agricultural matrix for the native species (Honorato, Crouzeilles, Ferreira, & Grelle, 2015). which exist in the CBNF and in the other forest frag- Most of the native forest fragments are in the sur- ments in the surrounding area. It is a common pattern rounding CBNF area (Figure 3), suﬀering pressure for many protected areas in Brazil and other tropical from the commercial planted forest stands and the regions (Garcia, Sawakuchi, Ferreira, & Ballester, expansion of agriculture and pasturelands. The high 2017; Ponte, Roch, Leinenkugel, Dech, & Kuenzer, shape metric values (MSI = 3.17 – Table 2, Figure 4) 2017), especially for those located in the region indicated that most of the forest fragments have elon- between Atlantic Forest and Cerrado, where agricul- gated and irregular shapes, which means they are tural expansion has been facilitated due to the strategic mainly composed by riparian corridors, which is GEOLOGY, ECOLOGY, AND LANDSCAPES 27 a common pattern in Atlantic Forest landscapes Pasture covered only 11% of the study area, but it (Mello et al., 2016; Moraes et al., 2017). Forest patches represented an important amount of land use in the with a complex shape such as riparian zones generate south where the forest fragments are small and limited more area under the edge eﬀect, showing a reduction to riparian corridors (Figure 3). According to in species richness and increased levels of species Latawiec, Strassburg, Brancalion, Rodrigues, and dominance (Joly et al., 2014; Matos et al., 2017). Gardner (2015), low-productivity pasturelands with In addition to the agriculture, pasture lands and few cattle are a great opportunity for forest restoration commercial stands of planted forest, the roads are an in tropical agricultural landscapes. The increase in important vector of habitat fragmentation in the study pasture productivity could provide new areas of land area, as shown in Figure 3. In addition, the state road for the restored forest. The forest restoration in our that divides the protected area into two parts is a threat study is important in the south area due to its high to the animal dispersion. The roads lead to deforesta- level of forest fragmentation, which compromises the tion, a decrease in patch size and an increase in the edge animals’ movement through the landscape. eﬀect, and also causes problems with roadkill and urban The FSI was a strong framework to determine expansion (Bager & Fontoura, 2013; Barber, Cochrane, priority forest fragments for landscape restoration Souza, & Laurance, 2014; Mello, Toppa, & Abessa, and can be replicated in other fragmented landscapes. 2011). Thus, it is a serious problem for many protected The mosaic comprised by agriculture, pasturelands, areas which are bordered or cut by roads (Barber et al., and planted forest allows restoration implementation, 2014; Diniz & Brito, 2015; Marcantonio, Rocchini, Geri, unlike in urban areas. According to Fagan, DeFries, Bacaro, & Amici, 2013;Moraesetal., 2017). Sesnie, Arroyo-Mora, and Chazdon (2016), the refor- The presence of planted forest (21%) is an important estation of priority areas in tropical agricultural land- characteristic of the study area, which is the predomi- scapes can improve connectivity by 2% with only 1% nant land use/land cover in the CBNF (Figure 3)due to gain in forest cover. its history of wood exploitation/extraction. The problem The focus areas for forest restoration are located caused by the expansion of exotic eucalyptus and pine close to the forest fragments with higher FSI values species over natural forest leading to habitat loss and (Figure 5), since they can facilitate natural restoration fragmentation has been reported in many tropical and guarantee the maintenance of ecosystem pro- regions (Ponte et al., 2017;Zamorano-Elgueta, cesses, as well as dispersion and pollination, which Benayas, Cayuela, Hantson, & Armenteras, 2015). are important for biodiversity restoration success. Although Eucalyptus sp. and Pinus sp. plantations The index prioritized forest fragments with the largest have been observed to be correlated to biodiversity size, close to others which are less irregular. These loss (Jara, Hylander, & Nemomissa, 2017), they can forest fragments are important because they can sup- provide regeneration sites for many native tree spe- port greater biological populations, high biodiversity cies in agricultural landscapes in the tropics (Silva, and diversity of habitats (Almeida, Gomes, & Queiroz, Poggiani, Sebbenn, & Mori, 2011), and may function 2011; Mello et al., 2016). Thus, they facilitate land- as environmental ﬁlters for native species coloniza- scape restoration because they still have a biological tion (Mendonça-Lima, Duarte, & Hartz, 2014). For structure and are connected to the landscape by other example, Zamorano-Elgueta et al. (2015) cited that forest fragments, as was indicated by Vettorazzi and natural forest regeneration in abandoned areas of Valente (2016). exotic tree plantations partly compensated forest In addition, the restoration of surrounding areas of loss in Chile. In fact, the understory of the planted forest fragments with high FSI value can improve their forests in the CBNF (Pinus sp. and Araucaria angu- size, shape, and connectivity. Thus, it is important to stifolia) present regeneration of native species of increase their potential for sheltering biodiversity and ombrophilous and semi-deciduous seasonal forests. being a dispersion source in the landscape because it According to Fonseca et al. (2009), tree monocul- can help to improve the ecological value of other forest tures (as Eucalyptus sp.) can greatly contribute to fragments (Valente & Vettorazzi, 2005). conserve biodiversity when maintaining a complex In this context, the maintenance of small fragments and diverse understory. with low FSI values is important to create habitat However, some native species can be inhibited by patches which are connected in the landscape, since Eucalyptus sp. or Pinus sp. roots (Zhang & Fu, 2009). they allow connection among forest fragments with According to Mendonça-Lima et al. (2014), native high importance (i.e. high FSI) (Mello et al., 2016). In pine (Araucaria angustifolia) has a higher diversity of simulating the removal of forest patches smaller than dispersal traits than exotic species (Pinus elliottii). This 50 ha in the Atlantic Forest cover in Brazil, Ribeiro shows the importance of eﬀective management of the et al. (2009) found an increase in the isolation of major protected area to allow native forest regeneration in forest patches and a reduction of their connectivity. planted forest stands, considering the existence of both The MPS in our study was equal to 19 ha, showing that native and exotic pine species. the landscape restoration depends on small fragments. 28 T. P. V. DE MATOS ET AL. Small forest fragments are important to promote the aiming at conserving protected areas, which in this structural and functional connectivity of larger frag- case is the CBNF. ments (Rubio & Saura, 2012). Forest fragments with Based on FSI, landscape planning can also support low FSI values contribute to restoring the landscape scenarios of biodiversity conservation actions, including and they can be prioritized in a second stage. restoration in the surrounding areas of the priority frag- Therefore, it is necessary to increase forest cover ments; ecological corridor design; and improvement in and habitat patch connectivity to reach the ecological the understory in the planted forest stands, aiming at processes restoration of the forest fragments in this improving the eﬀectiveness of the protected area. complex landscape mosaic. According to Ferraz et al. Future studies should be performed to verify the (2014), it is necessary to conserve old-growth forest eﬀectiveness of the proposed index using ﬁeld data of fragments and increase forest quality, as well as species richness and composition in the selected frag- increase forest cover. To do so, a combination of ments and in the forest restoration areas. The index actions can be implemented in the study area, includ- can be applied in other restoration planning schemes ing restoring the forest of surrounding areas of prior- worldwide, with diﬀerent types of protected areas and ity fragments (high values of FSI), implementing agricultural landscapes. ecological corridors and improving the understory The importance of the index is to identify priority diversity and regeneration of native species in the areas for implementing restoration actions to improve planted forest stands. This understory with native the restoration success in areas that present urgency of species in planted forests can provide important habi- vegetation recovery in order to better achieve restora- tat resources for wildlife species such as many birds tion targets. As the suitability of the protected areas and mammals in the Atlantic Forest (Brockerhoﬀ, depends on the surrounding landscape conservation, Jactel, Parrotta, & Ferraz, 2013). Improving the move- this index can help to set the priority areas for restora- ment of these animals will also increase plant species tion actions in a cost-eﬀective approach, increasing the dispersal in the landscape. chances of success of restoration projects. Thus, forest restoration planning based on an eco- logical landscape approach (Leite et al., 2013; Metzger Disclosure statement & Brancalion, 2013; Stanturf et al., 2014; Strassburg et al., 2019; Vettorazzi & Valente, 2016) can enable No potential conﬂict of interest was reported by the authors. reestablishing the landscape structure and ecological function of the focus area of restoration in the long Funding term. This work was supported by the FAPESP Fundação de Amparo à Pesquisa do Estado de São Paulo under Grant [number 13/16303-2]. 5. Conclusion The CBNF is a protected area of sustainable use which Data availability statement has a history of wood exploitation/extraction which nowadays is covered by planted forests of native The datasets used and/or analyzed during the current study are (Araucaria angustifolia) and exotic species (Pinus sp.). available from the corresponding author on reasonable Its surrounding area is predominantly occupied by request. agriculture, but we also observed a signiﬁcant area of planted Eucalyptus sp. forest. The native forest is scat- Geolocation information tered throughout the landscape and has a serious level of habitat fragmentation. Thus, there are a great num- Brazil ber of patches, with most of them being small in size and of irregular shape, suﬀering the inﬂuence of agri- References culture, pasturelands, planted forest stands, and roads. In this context, we proposed the Forest Suitability Almeida, F. S., Gomes, D. S., & Queiroz, J. M. (2011). Estratégicas para a conservação da biodiversidade Index (FSI) based on a landscape ecology approach to biológica em ﬂorestas fragmentadas. Ambiência, 7(2), prioritize forest patches that potentially support the 367–382. landscape restoration. Therefore, forest patches asso- Aronson, J., & Sasha, A. (2013). Ecosystem restoration is ciated with the highest FSI values have the best con- now a global priority: Time to roll up our sleeves. ditions to support an increase in biodiversity and Restoration Ecology, 21(3), 293–296. species dispersal throughout the landscape, thereby Bager, A., & Fontoura, V. (2013). Evaluation of the eﬀec- tiveness of a wildlife roadkill mitigation system in wet- facilitating natural restoration. Thus, we concluded land habitat. Ecological Engineering, 53,31–38. that FSI enables planning landscape restoration to Banks-Leite,C., Ewers,R.,Kapos,M.,&Metzger,J.P.(2013). increase functional connectivity and habitat quality, The confounded eﬀects of habitat disturbance at the local, GEOLOGY, ECOLOGY, AND LANDSCAPES 29 patch and landscape scale on understorey birds of the Convertino, M., Baker, K. M., Vogel, J. T., Lu, C., Suedel, B., Atlantic Forest: Implications for the development of & Linkov, I. (2013). Multi-criteria decision analysis to landscape-based indicators. Ecological Indicators, 31,82–88. select metrics for design and monitoring of sustainable Banks-Leite, C., Ewers, R. M., Kapos, V., Martensen, A. C., ecosystem restorations. Ecological Indicators, 26,76–86. & Metzger, J. P. (2011). Comparing species and measures De Jesus, E. N., Ferreira, R. A., Aragão, A. G., Santos, T. I. S., of landscape structure as indicators of conservation & Rocha, S. L. (2015). Structure of forest fragments of the importance. Journal of Applied Ecology, 48, 706–714. river basin Poxim-SE, as subsidy to ecological restoration. Baral, H., Keenan, R. J., Sharma, S. K., Stork, N. E., & Revista Árvore, 39(3), 467–474. Kasel, S. (2014). Spatial assessment and mapping of bio- Diniz, M., & Brito, D. (2015). Protected areas eﬀectiveness in diversity and conservation priorities in a heavily modiﬁed maintaining viable giant anteater (Myrmecophagatridactyla) and fragmented production landscape in north-central populations in an agricultural frontier. Brazilian Journal of Victoria, Australia. Ecological Indicators, 36, 552–562. Nature Conservation, 13, 143–151. Barber, C. P., Cochrane, M. A., Souza, C., Jr., & ENVI. Exelis Visual Information Solutions. Boulder: ITT Laurance, W. (2014). Roads, deforestation, and the miti- Visual Information Solutions. (2009). ENVI Version gating eﬀect of protected areas in the Amazon. Biological 4.7 August, 2009 Edition Copyright © ITT Visual Conservation, 177, 203–209. Information Solutions All Rights Reserved. Bechara, F. C. (2006). Unidades demonstrativas de Fagan, M. E., DeFries, R. S., Sesnie, S. E., Arroyo-Mora, P., & restauração ecológica através de técnicas nucleadoras: Chazdon, R. L. (2016). Targeted reforestation could reverse Floresta estacional semidecidual, Cerrado e Restinga declines in connectivity for understory birds in a tropical (Thesis (ph.D.)). Brasil: ESALQUSP, 248. habitat corridor. Ecological Applications, 26(5), 1456–1474. Bingham, H. C., Bignoli, D. J., Lewis, E., MacSharry, B., Fahrig, L. (2003). Eﬀects of habitat fragmentation on Burgess, N. D., Visconti, P., . . . Kingston, N. (2019). biodiversity. Annual Reviews in Ecology, Evolution and Sixty years of tracking conservation progress using the Systematics, 34, 487–515. world database on protected areas. Nature Ecology & Ferraz, S. F. B., Ferraz, K. M. P. M. B., Cassiano, C. C., Evolution, 3, 737–743. Brancalion, P. H., Da Luz, D. T. A., Azevedo, T. N., & Bortoleto, L. A., Figueira, C. J. M., Dunning, J. B., Jr, Metzger, J. P. (2014). How good are tropical forest Rodgers, J., & Silva, A. M. (2016). Suitability index for patches for ecosystem services provisioning? Landscape restoration in landscapes: An alternative proposal for Ecology, 29, 187–200. restoration projects. Ecological Indicators, 60, 724–735. Fonseca, C. R., Ganade, G., Baldissera, R., Becker, C. G., BRASIL. Lei federal nº. 12.651. de 25 de maio de 2012, Boelter, C. R., Brescovit, A. D., & Vieira, E. M. (2009). dispõe sobre a proteção da vegetação nativa; altera as Towards an ecologically-sustainable forestry in the Leis nos 6.938, de 31 de agosto de 1981, 9.393, de 19 de Atlantic Forest. Biological Conservation, Liverpool, 141, dezembro de 1996, e 11.428, de 22 de dezembro de 2006; 1209–1219. revoga as Leis nos 4.771, de 15 de setembro de 1965, Garcia, A. S., Sawakuchi, H. O., Ferreira, M. E., & e 7.754, de 14 de abril de 1989, e a Medida Provisória Ballester, M. V. R. (2017). Landscape changes in no 2.166-67, de 24 de agosto. a neotropical forest-savanna ecotone zone in central Bregman, T. P., Sekercioglu, C. H., & Tobias, J. A. (2014). Brazil: The role of protected areas in the maintenance of Global patterns and predictors of bird species responses to native vegetation. Journal of Environmental Management, forest fragmentation: Implication for ecosystem function 187,16–23. and conservation. Biological Conservation, 169,372–383. Geneletti, D. (2004). A GIS-based decision support system Brockerhoﬀ, E. G., Jactel, H., Parrotta, J. A., & Ferraz, S. F. B. to identify nature conservation priorities in an alpine (2013). Role of eucalypt and other planted forests in valley. Land Use Policy, 21(2), 149–160. biodiversity conservation and the provision of Honorato, R., Crouzeilles, R., Ferreira, M. S., & biodiversity-related ecosystem services. Forest Ecology Grelle, C. E. V. (2015). The eﬀects of habitat availability and Management, 301,43–50. and quality on small mammals abundance in the Carranza,M.L., Hoyos, L.,Frate,L., Acosta,A.T.R.,& Brazilian Atlantic Forest. Brazilia Journal of Nature Cabido, M. (2015). Measuring forest fragmentation Conservation, 13, 133–138. using multitemporal forest cover maps: Forest loss Instituto Chico Mendes De Conservação Da Biodiversidade and spatial pattern analysis in the Gran Chaco, (ICMBio). (2017). Plano de Manejo da FLONA de Capão Central Argentina. Landscape and Urban Planning, Bonito: Versão preliminar. Capão Bonito: ICMBIO 2017. 143,238–247. PLANO DE MANEJO DA FLORESTA NACIONAL DE Carrara, E., Arroyo-Rodríguez, V., Vega-Rivera, J. H., CAPÃO BONITO.São Paulo. Volume I. 230p Volume Schondube, J. E., de Freitas, S. M., & Fahrig, L. (2015). II.212p. Impact of landscape composition and conﬁguration on Jara, T., Hylander, K., & Nemomissa, S. (2017). Tree diver- forest specialist and generalist bird species in the frag- sity across diﬀerent tropical agricultural land use types. mented Lacandona rainforest, Mexico. Agriculture, Ecosystems and Environment, 240,92–100. BiologicalConservation, 184, 117–126. Joly, C. A., Metzger, J. P., & Tabarelli, M. (2014). Centro De Pesquisas Meteorológicas E Climáticas Aplicadas Experiences from the Brazilian Atlantic Forest: À Agricultura (CEPAGRI). (2013). Condições Ecological ﬁndings and conservation initiatives. New meteorológicas.Disponívelem http://www.cepagri.uni Phytologist, 204, 459–473. camp.br Landis, J. R., & Koch, G. G. (1977). The measurement of Chazdon, R. L. (2017). Tropical forest regeneration. observer agreement for categorical data. Biometrics, 33 Reference module in life sciences. Encyclopedia of (1), 159–174. Biodiversity, 2013, 277–286. Latawiec, A. E., Strassburg, B. B. N., Brancalion, P. H. S., Congalton, R. G., & Green, K. (1998). Assessing the accuracy Rodrigues, R. R., & Gardner, T. (2015). Creating space for of remotely sensed data: Principles and practices. large-scale restoration in tropical agricultural landscapes. New York: Lewis Publishers. Frontiers in Ecology Environment, 13(4), 211–218. 30 T. P. V. DE MATOS ET AL. Leite, M. S., Tambosi, L. R., Romitelli, I., & Metzger, J. P. Peixoto, C. A. B., & Theodorovicz, A. (2009). (2013). Landscape ecology perspective in restoration pro- Geodiversidade SP. BRASIL, C.-S. G. D. Brasília, jects for biodiversity conservation: A review. 2010 Peixoto, Carlos Augusto Brasil. Retrieved from Natureza&Conservação, 1(2), 108–118. http://www.cprm.gov.b r/publique/media/ Marcantonio, M., Rocchini, D., Geri, F., Bacaro, G., & Geodiversidade_SP.pdf Amici, V. (2013). Biodiversity, roads, & landscape frag- Peña-Domene, M., Minor, E. S., & Howe, H. F. (2016). mentation: Two Mediterranean cases. Applied Geography, Restored connectivity facilitates recruitment by an ende- 42,63–72. mic large-seeded tree in a fragmented tropical landscape. Matos, F. A. R., Magnago, L. F. S., Gastauer, M., Ecology, 97(9), 2511–2517. Carreiras, J. M. B., Simonelli, M., Meira-Neto, J. A. A., Périco, E., & Cemin, G. (2006). Caracterização da paisagem & Edwards, D. P. (2017). Eﬀects of landscape conﬁgura- do município de Arvorezinha, RS, com ênfase na tion and composition on phylogenetic diversity of trees in dinâmica dos fragmentos ﬂorestais, por meio de sistemas a highly fragmented tropical forest. Journal of Ecology, de informações geográﬁcas (SIGs). ScientiaForestalis, 70 105, 265–276. (1), 9–21. McGarigal, K., SA Cushman, and E Ene. 2012. FRAGSTATS Ponte, E., Roch, M., Leinenkugel, P., Dech, S., & Kuenzer, C. v4: Spatial Pattern Analysis Program for Categorical and (2017). Paraguay´s Atlantic Forest cover loss – Satellite- Continuous Maps. Computer software program pro- based change detection and fragmentation analysis duced by the authors at the University of Massachusetts, between 2003 and 2013. AppliedGeography, 79,37–49. Amherst. Available at the following web site: http://www. R Core Team. R. (2013). A language and environment for umass.edu/landeco/research/fragstats/fragstats.html. statistical computing. Vienna: R Foundation for Statistical Mello, K., Toppa, R. H., & Abessa, D. M. (2011). Dynamics Computing. of the landscape of the city of Cubatão: Growth among Rezende, C. L., Scarano, F. R., Assad, E. D., Joly, C. A., shores, industry facilities and the mountain range Serra Metzger, J. P., Strassburg, B. B. N., . . . do Mar. O Mundo Da Saúde, 35,42–46. Mittermeier, R. A. (2018). From hotspot to hopespot: Mello, K., Toppa, R. H., & Cardoso-Leite, E. (2016). Priority An opportunity for the Brazilian Atlantic Forest. areas for forest conservation in an urban landscape at the Perspectives in Ecology and Conservation, 16(4), 208–214. transition between Atlantic forest and Cerrado. CERNE Ribeiro, C. R., Metzger, J. P., Martensen, A. C., [Online], 22(3), ISSN 0104-7760, 277–288. Ponzoni, F. J., & Hirota, M. M. (2009). The Brazilian Mendonça-Lima, A., Duarte, L. S., & Hartz, S. M. (2014). Atlantic Forest: How much is left, and how is the remain- Comparing diversity and dispersal traits of tree commu- ing forest distributed? Implications for conservation. nities in plantations and native forests in Southern Brazil. BiologicalConservation, 142, 1141–1153. Brazilian Journal of Nature Conservation, 12(1), 24–29. Rocha-Santos, L.; Pessoa, M. S.; Cassano, C. R.; Talora, D. Metzger, J. P., Bernacci, L. C., & Goldenberg, R. (1997). C.; Orihuela, R. L.L.; Mariano-Neto, E.; Morante-Filho, J. Pattern of tree species diversity in riparian forest frag- C.; Faria, D.; Cazetta, E. .2016. The shrinkage of a forest: ments of diﬀerent widths (SE Brazil). Plant Ecology, 133 Landscape-scale deforestation leading to overall changes (2), 135–152. in local forest structure. Biological Conservation, v. 196, Metzger, J. P., & Brancalion, H. S. (2013). Challenges and p. 1-9. opportunities in applying a landscape ecology perspective Ross, J. L. S., & Moroz, I. (1997). Mapa geomorfológico in ecological restoration: A powerful approach to shape do estado de São Paulo - Escala 1:500.000 (pp. 64). São neolandscapes. Natureza&Conservação, 11(2), 103–107. Paulo: Departamento de Geograﬁa-FFLCH-USP/IPT/ Milan, E., & Moro, R. S. (2012). Padrões de fragmentação FAPESP. ﬂorestal natural no parque estadual de Vila Velha, Ponta Rubio, L., & Saura, S. (2012). Assessing the importance of Grossa (PR). Ambiência, 8(5), 685–697. individual habitat patches as irreplaceable connectivity Molin, P. G., Chazdon, R., Ferraz, S. F. B., & providers: And analysis of simulated and real landscape Brancalion, P. H. S. (2018). A landscape approach for data. Ecological Complexity, 11,28–37. cost-eﬀective large-scale forest restoration. Journal of Rudnick, D., Beier, P., Cushman, S., Dieﬀenbach, F., Applied Ecology, 55, 2767–2778. Epps, C. W., Gerber, L., & Trombulak, S. C. (2012). The Moraes, M. C. P., Mello, K., & Toppa, R. H. (2017). role of landscape connectivity in planning and implement- Protected areas and agricultural expansion: Biodiversity ing conservation and restoration Priorities. Issues in conservation versus economic growth in the Southeast of ecology Report No. 16. Ecological Society of America. Brazil. Journal of Environmental Management, 188, Washington, DC. 73–84. Santos Junior, P. C. A., Marques, F. C., Lima, M. R., & Dos Newman, M. E., McLaren, K. P., & Wilson, B. S. (2014). Anjos, L. (2016). The importance of restoration areas to Assessing deforestation and fragmentation in a tropical conserve Bird species in a highly fragmented Atlantic moist forest over 68 years; the impact of roads and legal forest landscape. Brazilian Journal of Nature protection in the Cockpit Country, Jamaica. Forest Conservation, 14,1–7. Ecology and Management, 315, 138–152. Santos, R. S., Ribeiro, C. A. A. S., Peluzio, T. M., Nichol, J. E., Abbas, S., & Fischer, G. A. (2017). Spatial Peluzio, J. B. E., Queiroz, V. T., Branco, E. R., & patterns of degraded tropical forest and biodiversity Almeida, S. L. H. (2016). Geotechnology and landscape restoration over 70-years of succession. Global Ecology ecology applied to the selection of potential forest frag- and Conservation, 11, 134–145. ments for seed harvesting. Journal of Environmental Oakleaf, J. R., Matsumoto, M., Kennedy, C. M., Management, 183, 1050–1063. Baumgarten, L., Miteva, D., Sochi, K., & Kiesecker, J. Saura, S., Bertzky, B., Bastin, L., Battistella, L., Mandrici, A., (2017). LegalGEO: Conservation tool to guide the siting & Dubois, G. (2018). Protected area connectivity: of legal reserves under the Brazilian Forest Code. Applied Shortfalls in global targets and country-level priorities. Geography, 86,53–65. Biological Conservation, 219,53–67. GEOLOGY, ECOLOGY, AND LANDSCAPES 31 Sekercioglu, C. H., Loarie, S., Oviedo-Brenes, F., Uezu,A.,&Metzger,J.P.(2016). Time-lag in Mendenhall,C.,Daily, G.R.,& Ehrlich,P.(2015). Tropical responses of birds to Atlantic Forest fragmentation: countryside riparian corridors provide critical habitat and Restoration opportunity and urgency. PloS One, 11 connectivity for seed-dispersing forest birds in a fragmented (1), e0147909. landscape. Journal of Ornithology, 56(S1), 343–353. Valente, R. O. A., & Vettorazzi, C. A. (2005). Comparação Silva, P.H.M.,Poggiani,F.,Sebbenn,A. M.,&Mori,E.S. entre métodos de avaliação multicriterial, em ambiente (2011). Can Eucalyptus invade native forest fragments close SIG, para a conservação e a preservação ﬂorestal. Scientia to commercial stands? Forest Ecology and Management, 261, Forestalis, 69,51–61. 2075–2080. Valente, R. O. A., & Vettorazzi, C. A. (2008). Deﬁnition of Singh, S., & Mishra, A. (2014). Spatiotemporal analysis of priority areas for forest conservation through the ordered the eﬀects of forest covers on stream water quality in weighted averaging method. Forest Ecology and Western Ghats of peninsular India. Journal of Management, 256(6), 1408–1417. Hydrology, 519, 214–224. Veloso, H.P.(2012). Sistema Fitogeográﬁco. In Instituto Slade, E. M., Merckx, T., Riutta, T., Bebber, D. P., Redhead, D., Brasileiro De Geograﬁa E Estatística. Manual Técnico Riordan, P., & Macdonald, D. W. (2013). Life-history traits da Vegetação Brasileira (2., pp. 9–38). Rio de Janeiro: and landscape characteristics predict macro-moth responses IBGE. to forest fragmentation. Ecology, 94(7), 1519–1530. Vettorazzi, C. A., & Valente, R. A. (2016). Priority areas for Stanturf, J. A., Palik, B. J., & Dumroese, R. K. (2014). forest restoration aiming at the conservation of water Contemporary forest restoration: A review emphasizing resources. EcologicalEngineering, 94, 255–267. function. Forest Ecology and Management, 331, 292–323. Zamorano-Elgueta, C., Benayas, J. M. R., Cayuela, L., Strassburg, B. B. N., Beyer, H. L., Crouzeilles, R., Hantson, S., & Armenteras, D. (2015). Native forest Iribarrem, A., Barros, F., Siqueira, M. F., . . . Uriarte, M. replacement by exotic plantations in southern Chile (2019). Strategic approaches to restoring ecosystems can (1985–2011) and partial compensation by natural triple conservation gains and halve costs. Nature Ecology regeneration. Forest Ecology and Management, 345, & Evolution, 3,62–70. 10–20. Tambosi, L., Martensen, A., Ribeiro, M., & Metzger, J. Zhang, C., & Fu, S. (2009). Allelopathic eﬀects of Eucalyptus (2014). A framework to optimize biodiversity restoration and the establishment of mixed stands of Eucalyptus and eﬀorts based on habitat amount and landscape native species. Forest Ecology and Management, 258(7), connectivity. RestorationEcology, 22, 169–177. 1391–1396.

Journal

Geology Ecology and Landscapes – Taylor & Francis

Published: Jan 2, 2021

Keywords: Landscape connectivity; forest conservation; atlantic forest; land use; buffer zone

References

Estratégicas para a conservação da biodiversidade biológica em florestas fragmentadas
Ecosystem restoration is now a global priority: Time to roll up our sleeves
Evaluation of the effectiveness of a wildlife roadkill mitigation system in wetland habitat
The confounded effects of habitat disturbance at the local, patch and landscape scale on understorey birds of the Atlantic Forest: Implications for the development of landscape-based indicators
Comparing species and measures of landscape structure as indicators of conservation importance
Spatial assessment and mapping of biodiversity and conservation priorities in a heavily modified and fragmented production landscape in north-central Victoria, Australia
Roads, deforestation, and the mitigating effect of protected areas in the Amazon
Unidades demonstrativas de restauração ecológica através de técnicas nucleadoras: Floresta estacional semidecidual, Cerrado e Restinga
Sixty years of tracking conservation progress using the world database on protected areas
Suitability index for restoration in landscapes: An alternative proposal for restoration projects
Global patterns and predictors of bird species responses to forest fragmentation: Implication for ecosystem function and conservation
Role of eucalypt and other planted forests in biodiversity conservation and the provision of biodiversity-related ecosystem services
Measuring forest fragmentation using multitemporal forest cover maps: Forest loss and spatial pattern analysis in the Gran Chaco, Central Argentina
Impact of landscape composition and configuration on forest specialist and generalist bird species in the fragmented Lacandona rainforest, Mexico
Condições meteorológicas
Tropical forest regeneration. Reference module in life sciences
Multi-criteria decision analysis to select metrics for design and monitoring of sustainable ecosystem restorations
Structure of forest fragments of the river basin Poxim-SE, as subsidy to ecological restoration
Protected areas effectiveness in maintaining viable giant anteater (Myrmecophagatridactyla) populations in an agricultural frontier
Targeted reforestation could reverse declines in connectivity for understory birds in a tropical habitat corridor
Effects of habitat fragmentation on biodiversity
How good are tropical forest patches for ecosystem services provisioning?
Towards an ecologically-sustainable forestry in the Atlantic Forest
Landscape changes in a neotropical forest-savanna ecotone zone in central Brazil: The role of protected areas in the maintenance of native vegetation
A GIS-based decision support system to identify nature conservation priorities in an alpine valley
The effects of habitat availability and quality on small mammals abundance in the Brazilian Atlantic Forest
Tree diversity across different tropical agricultural land use types
Experiences from the Brazilian Atlantic Forest: Ecological findings and conservation initiatives
The measurement of observer agreement for categorical data
Creating space for large-scale restoration in tropical agricultural landscapes
Landscape ecology perspective in restoration projects for biodiversity conservation: A review
Biodiversity, roads, & landscape fragmentation: Two Mediterranean cases
Effects of landscape configuration and composition on phylogenetic diversity of trees in a highly fragmented tropical forest
Dynamics of the landscape of the city of Cubatão: Growth among shores, industry facilities and the mountain range Serra do Mar
Priority areas for forest conservation in an urban landscape at the transition between Atlantic forest and Cerrado
Comparing diversity and dispersal traits of tree communities in plantations and native forests in Southern Brazil
Pattern of tree species diversity in riparian forest fragments of different widths (SE Brazil)
Challenges and opportunities in applying a landscape ecology perspective in ecological restoration: A powerful approach to shape neolandscapes
Padrões de fragmentação florestal natural no parque estadual de Vila Velha, Ponta Grossa (PR)
A landscape approach for cost-effective large-scale forest restoration
Protected areas and agricultural expansion: Biodiversity conservation versus economic growth in the Southeast of Brazil
Assessing deforestation and fragmentation in a tropical moist forest over 68 years; the impact of roads and legal protection in the Cockpit Country, Jamaica
Spatial patterns of degraded tropical forest and biodiversity restoration over 70-years of succession
LegalGEO: Conservation tool to guide the siting of legal reserves under the Brazilian Forest Code
Geodiversidade SP. BRASIL, C.-S. G. D. Brasília, 2010 Peixoto, Carlos Augusto Brasil
Restored connectivity facilitates recruitment by an endemic large-seeded tree in a fragmented tropical landscape
Caracterização da paisagem do município de Arvorezinha, RS, com ênfase na dinâmica dos fragmentos florestais, por meio de sistemas de informações geográficas (SIGs)
Paraguay´s Atlantic Forest cover loss – Satellite-based change detection and fragmentation analysis between 2003 and 2013
From hotspot to hopespot: An opportunity for the Brazilian Atlantic Forest
The Brazilian Atlantic Forest: How much is left, and how is the remaining forest distributed? Implications for conservation
Assessing the importance of individual habitat patches as irreplaceable connectivity providers: And analysis of simulated and real landscape data
The role of landscape connectivity in planning and implementing conservation and restoration Priorities. Issues in ecology
The importance of restoration areas to conserve Bird species in a highly fragmented Atlantic forest landscape
Geotechnology and landscape ecology applied to the selection of potential forest fragments for seed harvesting
Protected area connectivity: Shortfalls in global targets and country-level priorities
Tropical countryside riparian corridors provide critical habitat and connectivity for seed-dispersing forest birds in a fragmented landscape
Can Eucalyptus invade native forest fragments close to commercial stands?
Spatiotemporal analysis of the effects of forest covers on stream water quality in Western Ghats of peninsular India
Life-history traits and landscape characteristics predict macro-moth responses to forest fragmentation
Contemporary forest restoration: A review emphasizing function
Strategic approaches to restoring ecosystems can triple conservation gains and halve costs
A framework to optimize biodiversity restoration efforts based on habitat amount and landscape connectivity
Time-lag in responses of birds to Atlantic Forest fragmentation: Restoration opportunity and urgency
Comparação entre métodos de avaliação multicriterial, em ambiente SIG, para a conservação e a preservação florestal
Definition of priority areas for forest conservation through the ordered weighted averaging method
Priority areas for forest restoration aiming at the conservation of water resources
Native forest replacement by exotic plantations in southern Chile (1985–2011) and partial compensation by natural regeneration
Allelopathic effects of Eucalyptus and the establishment of mixed stands of Eucalyptus and native species

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Protected areas and forest fragmentation: sustainability index for prioritizing fragments for landscape restoration

Protected areas and forest fragmentation: sustainability index for prioritizing fragments for landscape restoration

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Protected areas and forest fragmentation: sustainability index for prioritizing fragments for landscape restoration

Protected areas and forest fragmentation: sustainability index for prioritizing fragments for landscape restoration

References

Abstract

Journal

Recommended Articles

References

Our policy towards the use of cookies