SciELO - Scientific Electronic Library Online

vol.24 número2Análise tricológica de pelos-guarda de roedores do pampa brasileiro, com uma chave de identificaçãoFeeding ecology of Marmosa demerarae (Thomas, 1905) and Marmosops bishopi (Pine, 1981) (Mammalia, Didelphidae) in forest fragments of the southern Amazon índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




  • No hay articulos citadosCitado por SciELO

Links relacionados


Mastozoología neotropical

versión impresa ISSN 0327-9383versión On-line ISSN 1666-0536

Mastozool. neotrop. vol.24 no.2 Mendoza dic. 2017



Effects of patch attributes on the richness of medium- and large-sized mammals in fragmented semi-deciduous forest


Roniel F. Oliveira¹, ² and Wellington Hannibal²,³

¹ Programa de Pós-Graduação em Biodiversidade e Conservação, Instituto Federal Goiano, Rio Verde, Laboratório de Zoologia, Rio Verde, GO, Brazil. [Correspondence: Roniel F. Oliveira <>]
² Laboratório de Ecologia e Biogeografia de Mamíferos, Universidade Estadual de Goiás Quirinópolis, GO, Brazil
³ Programa de Pós-Graduação em Ecologia e Conservação, Universidade Federal do Mato Grosso do Sul, Campo Grande, MS, Brazil

Recibido 28 octubre 2016.
Aceptado 4 mayo 2017.
Editor asociado: T Escalante


We evaluated the effect of patch attributes on the richness of medium- and large-sized mammals in semi-deciduous forest patches in the southern of Goiás state, Brazil. We surveyed 14 semi-deciduous for­est patches (35 to 869 ha) from January to December 2015. Medium- and large-sized mammal were sampled through direct observation, tracks, feces and burrows. In each semi-deciduous forest patch, we measured total area and core area (in hectare), degree of isolation (in meters) and connectivity (patches connected to riparian forest versus isolated patches). We recorded 22 species of medium- and large-sized mammals. Species richness was positively related to the core area of the patches. Patch size had a significant influence on species richness showing that this attribute is important to maintain community structure. Therefore, the fragmented landscape of southern Goiás still harbors a rich fauna of medium- and large-sized mammals.


Efeitos de atributos de mancha na riqueza de mamíferos de médio e grande porte em florestas semidecíduas fragmentadas.

Neste estudo, nós avaliamos o efeito dos atributos da mancha sobre a riqueza de mamíferos de médio e grande porte em manchas de floresta semidecidual no sul do estado de Goiás, Brasil. Quatorze manchas (de 35 a 869 ha) foram estudados durante janeiro a dezembro de 2015. Mamíferos de médio e grande porte foão inventariados por meio de observações diretas, pegadas, fezes e tocas. Em cada mancha de floresta semidecidual foi mensurada a área total e área núcleo (em hectare), grau de isolamento (em metros) e conectividade (manchas conectadas a mata ripária versus manchas isoladas). Vinte e duas espécies de mamíferos de médio e grande porte foram registradas. A riqueza de espécies esteve positivamente correlacionada com a área núcleo do fragmento. O tamanho da área teve uma influência significativa sobre a riqueza de espécies, mostrando que esse atributo é importante para manter a estrutura da comunidade. Portanto, a paisagem frag­mentada do sul de Goiás abriga uma rica e ameaçada fauna de mamíferos de médio e grande porte.

Key words: Conservation; Core area; Degree of isolation; Human-altered landscape.

Palavras-chave: Área núcleo; Conservação; Grau de isolamento; Paisagem alterada pelo homem.


The Cerrado, one of the world’s biodiversity hotspots (Myers et al. 2000) has already lost more than 50% of the vegetation cover (Klink & Machado 2005). Fragmentation processes have been more intense in the southern portion of the Cerrado, leaving only 15.7% of native vegetation in the Quirinópolis micro-region (Rossi 2016), our focal area. Therefore, it is important to study the effects of fragmenta­tion of the medium- and large-sized mammals.

Fragmentation is a change in habitat con­figuration that results in other effects, such as increases in number of patches, reduction of original habitat area, decrease in patch size, and increase in patch isolation (Fahrig 2003), negatively influencing the composition, rich­ness and diversity of the mammal community (Chiarello 1999, 2000; Cáceres et al. 2010; Garmedia et al. 2013).

Several studies have shown that patch size is an important factor for the richness of me­dium- and large-sized mammals, because these species require large areas for maintenance of their populations (Turner 1996; Chiarello 1999; Júnior & Kolher 2009; Garmendia et al. 2013). Furthermore, reduced patch size increases isola­tion between populations and decreases gene flow, leading to extinction of local populations (Chiarello 1999).

We investigated the effect of fragmentation on the richness of medium- and large-sized mammals in semi-deciduous forest patches and addressed the following questions: (1) does patch size and degree of isolation influ­ence the richness of medium- and large-sized mammals?; and (2) the patches connected to riparian forest have more species than isolated patches? We hypothesized that (1) there is a positive relationship between patch size and species richness, and a negative relationship between degree of isolation and species rich­ness; and (2) connected patches presented more species than isolated patches.


Study area

This study was conducted in the municipality of Quirinópolis, southern Goiás State (Fig. 1). The climate is Tropical semi-humid, markedly seasonal, with dry winters (Aw in Köppen climate classifica­tion), mean annual temperatures of 23 °C (max. 40 °C and min. 14 °C), and mean annual rainfall of around 1600-1900 mm. The dry season occurs between April and September, and wet season from October to March (Alvares et al. 2014).

Fig. 1
. Fourteen patches surveyed in southern Goiás state, Brazil.

The landscape surveyed comprised mainly semi-deciduous forest patches, riparian forest (mainly, gallery forests), buriti palm (vereda), and shrubby savanna (cerrado stricto sensu) on the mountain tops, inserted in a matrix composed of sugar-cane plantations and exotic grasses. We surveyed 14 patches of semi-deciduous forest between January and June of 2015, with a mean of two patches by month, and re-surveyed between July and Decem­ber of 2015 (Fig. 1). Therefore, we surveyed each of the 14 patches two times, one in the dry and one wet season.

We select the patches according to: i) spatial con­nectivity: we sampled 1-connected and 1-isolated patch each month, so, we chose patches with this characteristic and minor distance to each other; ii) patch size: in each patch, we measured total area and core area (in hectares), mean of degree of isolation (in meters) and connectivity (presence/ absence). Total area was considered the patch as a whole. Core area was considered to be the area of patch after eliminating 100 m closest to the edge. Degree of isolation was the average distance (in meters) from the central point of the studied patch to the edge of the four nearest patches. The con­nectivity was defined categorically, dividing patches between those that were connected to the riparian forest and those that were isolated (Table 1). The measurements were taken in Google Earth.

Table 1
Characteristics and location of 14 studied patches in southern Goias, Brazil. Patches (P). Degree of isolation (DIM). Connectivity (C): Con = connected, Iso = isolated.

Sampling design

We sampled medium- and large-sized mammals in each patch through active searching in linear transects (1 km) disposed on the edge and two linear transects (500 m) inside the patch. Active searching is the second most effective method for mammal survey (Santos et al. 2013). Each transect/ patch was searched on average one hour/day, during four consecutive days in each sampling phase, an effort of eight hours and 12 km by patch, totaling 112 hours and 168 km, with variation as a func­tion of size of the patch. During active searches, we sampled medium- and large-sized mammals by direct observation, tracks, feces and burrows (order Cingulata). Feces, tracks and other signs were identi­fied according to Borges & Tomas (2004). Deers of the genus Mazama were treated at the generic level due to the difficulty in determining the species by their tracks. We also excluded species that occur exclusively in open areas according to Reis et al. (2011). For taxonomic classification, we followed Páglia et al. (2012).

Data analysis

In each patch, we annotated composition and rich­ness of medium- and large-sized mammals. Total area of the patch was not incuded because it was highly correlated with core area. Linear regression analysis was used for testing the relation between species richness (response variable) to core area of the patch, and mean of degree of isolation (predictor variables). A t-test was used to verify difference of the mean species richness between patches connected to riparian forest and isolated ones. The measures of total area, core area and mean of degree of iso­lation were transformed to natural logs for greater consistency of the values.


Twenty-two species of medium- and large-sized mammals, distributed in 15 families and eight orders, were recorded in the 14 patches (Appendix). Myrmecophaga tridactyla (Linnaeus, 1758), Priodontes maximus (Kerr, 1792), Puma concolor (Linnaeus, 1771), Puma yagouaroundi (É. Geoffroy, 1803) and Tapirus terrestris (Lin­naeus, 1758) are nationally or globally threatened (ICMBio 2014; IUCN 2016; Appendix).

Species richness was positively related to core area (r² = 0.4999; p = 0.0046; Fig. 2a) of the patches, in line with our initial hypothesis that patches with largest areas supported greater richness of medium- and large-sized mammals.

Fig. 2.
Relationship between: a = log core area (ha), b = log (degree of isolation mean) (DIM), to the species richness of medium- and large-size mammals in semi-deciduous forest patches in southern Goiás, Brazil.

However, there was no significant relation be­tween species richness and the mean of degree of isolation (r² = 0.1841; p = 0.1258; Fig. 2b). In addition, there was no significant difference in the species richness between patches connected to forest riparian forests versus isolated patches (t=0.7094; p=0.4916; Fig. 3).  

Fig. 3.
Difference in species richness of the medium- and large-size mammals between connected to riparian forest versus isolated semi-deciduous forest patches in southern Goiás, Brazil.


A study for the region describes 23 species of medium and large mammals (Hannibal et al. 2015). We recorded 22 species, including some not previously recorded in the region: P. maximus, Alouatta caraya, Galictis cuja, and P. yagouaroundi. Our study also highlights the importance of this highly fragmented landscape (Rossi 2016) in harboring five threatened species.

The area also showed a positive relation with species richness of medium- and large-sized mammals, corroborating our initial hypothesis. Smaller patches have relatively large edge areas, resulting in relaxation (Primack & Rodrigues 2001). The edge effect negatively influences mammals (Chiarello 2000), so it is necessary to conserve the core of these areas to maintain this species richness.

Contrary to our initial hypothesis, degree of isolation did not affect species richness of medium- and large-sized mammals. These results can be explained by high dispersal potential presented by most medium and large mammalian species, moving between the patch (Chiarello 1999; Cáceres et al. 2010; Calaça et al. 2010). This dispersion can be dependent of the fragment area, connectivity to riparian forest and isolation. In addition, favoring dispersion of these mammals is important because they need large home ranges (Reis et al. 2011).

The mean species richness of medium- and large-sized mammals was no different be­tween patches connected to riparian forest and isolated ones, also contrary to our initial hypothesis. This result can be explained by the presence of generalist species, which have a high potential of dispersion, and can occur in open or forest environments (Chiarello 1999; Calaça et al. 2010). However, areas connected to riparian forests may present high heterogene­ity, with greater resource availability, important for maintenance of mammalian communities (Magioli et al. 2015). Moreover, riparian for­est act as ecological corridors connecting the forest mammal faunas (Amazon and Atlantic) with the Cerrado of central Brazil (Johnson et al. 1999).

The southern Goiás State, though already strongly fragmented due to monocultures and artificial pastures, still retains a rich and threatened fauna of medium- and large-sized mammals, showing the importance of these human-altered landscapes for mammal con­servation. There was a significant positive relationship between species richness with core area of the patch, indicating the impor­tance of maintaining large core area of native vegetation for the conservation of mammals in central-southern Brazil. We emphasize here the need to create protected areas to promote the conservation of these patches, and encourag­ing research on their roles. This study has the potential to be replicated in other taxonomic groups and ecoregions.


We thank members of the Laboratory of Ecology and Biogeography of Mammals for their help in field work and Dra. Levi Carina Terribile for the English revision. We also thank to the owners of the farms where we performed this work.


1. Alvares, C.A., J. L. Stape, P. J. Sentelhas, M. L. J. Gonçalves, & P. C. Sparovek. 2014. Köppen’s climate classification map for Brazil. Meteorologische Zeitschrift 22:711-728.

2. Borges, P. A. L., & W. M. Tomás. 2004. Guia de rastros e outros vestígios do pantanal. Embrapa pantanal, Corumbá         [ Links ].

3. Cáceres, N. C , R. P. Nápoli, J. Casella, & W. Hannibal. 2010. Mammals in fragmented savannah landscape in South-western Brazil. Jornal of Natural History 44:491-512.         [ Links ]

4. Calaça, A. M., F. R. Melo, P. M. Junior, A. T. A. Jácomo, & L. Silveira. 2010. The influence of fragmentation on the carnivores distribution on a landscape of Cerrado. Neotropical Biology and Conservation 5:31-38.         [ Links ]

5. Chiarello, A. G. 1999. Effects of fragmentation of the Atlantic forest on mammal communities in south-eastern Brazil. Biological Conservation 89:71-82.         [ Links ]

6. Chiarello, A. G. 2000. Conservation value of a native forest fragment in a region of extensive agriculture. Revista Brasileira de Biologia 60:237-247.         [ Links ]

7. Fahrig, L. 2003. Effects of habitat fragmentation on biodiversity. Revista Ecology Systemica 34:487-515.         [ Links ]

8. Garmedia, A., V. Arroyo-Rodrigues, A. Estrada, & E. J. Naranjo. 2013. Landscape and patch attributes impacting medium- and large-sized terrestrial mammals in a fragmented rain forest. Journal of Tropical Ecology 29:331-344.         [ Links ]

9. Hannibal, W. et al. 2015. Mamíferos não-voadores em fragmentos de Cerrado no Sul do estado de Goiás, Brasil. Boletim da Sociedade Brasileira de Mastozoologia 73:103-109.         [ Links ]

10. IUCN. 2016. IUCN Red list of threatened Species. < Acessed in:2016/16/12.         [ Links ]>

11. ICMBio. 2014. ICMBio Instituto Chico Mendes de Conservação da Biodiversidade. < Acessed in: 2015/09/10>         [ Links ]

12. Johnson, M. A., P. M. Saraiva, & D. Coelho. 1999. The role of gallery forests in the distribution of Cerrado mammals. Revista Brasileira de Biologia 59:421-427.         [ Links ]

13. Júnior, E. F. A., & A. Kohler. 2009. Mastofauna de médio e grande porte na RPPN da UNISC, RS, Brasil. Biota Neotropica 9:169-174.         [ Links ]

14. Klink, C. A., & R.B. Machado. 2005. A Conservação do Cerrado brasileiro. Megadiversidade 1:147-155.         [ Links ]

15. Magioli, M., M. C. Ribeiro, K. M. P. M. B. Ferraz, & M. G. Rodrigues. 2015. Thresholds in the relationship between functional diversity and patch size for mammals in the Brasilian Atlantic Forest. Animal Conservation 1-13.         [ Links ]

16. Myers, N., R. A. Mittermier, G. A. B. Fonseca, & J. Kent. 2000. Biodiversidade hotspot for conservation priorities. Nature 403:853-858.         [ Links ]

17. Páglia, A. P. et al. 2012. Lista anotada dos mamíferos do Brasil. Occasional Paper 6:79.         [ Links ]

18. Primack, R. B., & E. Rodrigues. 2001. Biologia da Conservação. Editora Planta. Londrina.         [ Links ]

19. Reis, N. R., A. L. Peracchi, W. A. Pedro, & I. P. Lima. 2011. Mamíferos do Brasil, 2ed, Editora Londrina, Londrina         [ Links ]

20. Rossi, F. R. 2016. Determinantes da dinâmica do uso do solo em um hotspot de biodiversidade: o cerrado do sul de Goiás. Masters Dissertation, University the Brasilia, Brasília, Brazil.         [ Links ]

21. Santos, C. F., B. Bueno, & J. Casella. 2013. Comparação entre métodos de amostragem e eficiência de iscas na atração de mamíferos de médio e grande porte no Cerrado. Neotropical Biology and Conservation 8:156-164.         [ Links ]

22. Turner, I. M. 1996. Species loss in fragments of tropical rain forest: a review of the evidence. Journal of Applied Ecology 33:200-209.         [ Links ]


Species list of medium- and large-sized mammals in 14 semi-deciduous forest patches in southern Goias, Brazil. National threat category: Brazil (ICMBio 2014) and international (IUCN 2016): vulnerable (VU), near threatened (NT), data deficient (DD). Methods: burrow (B), direct observation (D), faeces (F), tracks (T).


Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons