LARGE-MEDIUM MAMMAL DIVERSITY IN FRAGMENTED HABITAT: A CAMERA TRAP INVENTORY SURROUNDING A RURAL COMMUNITY IN
THE UPPER PARANÁ ATLANTIC FOREST OF PARAGUAY
DIVERSIDAD DE MAMÍFEROS EN UN HÁBITAT FRAGMENTADO: RELEVAMIENTO CON CÁMARA-TRAMPA EN LAS INMEDIACIONES DE UNA
COMUNIDAD RURAL EN EL BOSQUE ATLÁNTICO DEL ALTO PARANÁ DE PARAGUAY
*Nicholas Gengler
*Peace Corps Paraguay, 162 Chaco Boreal casi Mariscal López, Asunción, Paraguay. Email: [email protected].
Abstract.- A camera trap inventory of terrestrial vertebrates was carried out in the Atlantic Forest of Paraguay around the small rural community of Isla Alta in the Department of Paraguarí. One Reconxy HC600 camera trap was used over the course of 398 camera trap nights in 12 different sites around the periphery of Isla Alta. The area surveyed covers roughly 500 ha within an 8,600 ha forest fragment connected with Ybycuí National Park (YNP). A total of 22 wild vertebrate species were identified: one reptile, nine birds, and twelve mammals. Noteworthy species registered were margay (Leopardus wiedii), greater naked-tailed armadillo (Cabassous tatouay), and Azara’s agouti (Dasyprocta azarae). These three species have not been recorded from YNP and are of potential concern for conservation. However, the absence of certain large mammals from the forest surrounding Isla Alta which are registered within YNP may indicate impact from local hunting pressure occurring outside of the park.
Resumen.- Un relevamiento, utilizando una cámara trampa, fue llevado a cabo en el Bosque Atlántico del Alto Paraná, en el departamento de Paraguarí, Paraguay alrededor de Isla Alta, compañía de La Colmena. Una Reconxy HC600 fue utilizada sobre 398 noches en 12 sitios distintos cubriendo 500 ha de un fragmento de bosque de 8.600 ha que está conectado con el Parque Nacional Ybycuí (PNY). Un total de 22 especies de vertebrados terrestres fueron registradas: un reptil, nueve aves, y 12 mamíferos. Las especies detectadas más notables fueron tirika (Leopardus wiedii), tatú de rabo molle (Cabassous tatouay), y akuti say’ju (Dasyprocta azarae), que no han sido registrados en el PNY, y son de preocu-pación potencial para su conservación. Sin embargo, la ausencia de ciertos mamíferos grandes del área de estudio en los alrededores del PNY puede indicar una presión significante de cacería local.
Palabras Clave: fragmentación del bosque, presión de cacería, Leopardus wiedii, Cabassous tatouay, Dasyprocta azarae.
The Atlantic Forest is one of the most diverse and also threatened ecosystems in the world, ranking as one of the five most vulnerable biodiversity “hot spots” (Myers et al., 2000). It extends westward from the Atlantic coast, and southward from the equator in Brazil to temperate latitudes in Paraguay and Argentina. The long north to south extension results in a heterogeneous ecosystem complex, comprised of 15 ecoregions (Fundación Vida Silvestre Argentina and WWF, 2017). Its largest ecore-
gion, the Upper Paraná Atlantic Forest (UPAF), is found around the tri-national border region of Brazil, Argentina, and Paraguay. The UPAF climate is subtropical with an average annual temperature of 16-22°C and 1,000-2,200 mm of rainfall per year mostly in the austral summer (Di Bitetti et al., 2003).
The UPAF has lost 92% of its original forest cover, mostly to agricultural land conversion (Fundación Vida Silvestre Argentina and WWF, 2017). The remaining 8% of the UPAF cover
Boletín del Museo nacional de Historia natural del Paraguay Vol. 22, Nº 2 (dicieMBre 2018)60
Nicholas GeNGler
is severely fragmented, with almost half of the total remaining area comprised of fragments of <100 km² (Di Bitetti et al., 2003). This has significantly impacted habitat and reduced bio-diversity within much of the UPAF ecosystem. Fragmentation in the Atlantic Forest leads to a disproportionate decrease in mammalian bio-mass (Chiarello, 2000), and diminished local biodiversity, particularly in bird assemblages (Ribon et al., 2003; Turner, 1996).
In addition to providing inadequate habitat for many species, fragmentation exposes the UPAF to greater risk of edge effect impacts, which can include tree mortality, changes in forest structure, and increased accessibility to humans and invasive species (Harper et al., 2005).
Given the extensive habitat loss in the UPAF, it is important to evaluate which species can adapt to various degrees of degradation and pressure. To date, little species inventory work to date has been done in the UPAF of Paraguay outside of a few protected national parks (Lowen et al., 1996; Brooks et al., 1993; Fragano et al.,
2003). The study presented here provides spe-cies inventory data on forest fragments within an agricultural matrix, which is the type of landscape now most common throughout the Paraguayan UPAF.
METHODSStudy AreaYbycuí National Park, located in the department of Paraguarí and managed by the Secretariat of the Environment (SEAM), is a small remnant of the UPAF in Paraguay. This park consists of 5,124 ha under strict national protection (FCA-UNA, 2015).
A meandering and mostly continuous for-est corridor not part of YNP, extends northeast from the park (Fig. 1). This 8,600 ha corridor, roughly 15km long and 1-3km wide, is part of a patchwork agricultural matrix of mostly small-scale, family-level agriculture. National regulations require the preservation of forests on steep slopes to prevent erosion. Because much of this forest corridor is situated on steep terrain (up to 20% grade), this regulation, along with
Figure 1. Fragment area. Yellow diamond indicates the location of Isla Alta. Red square indicates the approximate boundaries of the Ybycui National Park (YNP). Map data: Google, Landsat/Copernicus.
Boletín del Museo nacional de Historia natural del Paraguay Vol. 22, Nº 2 (dicieMBre 2018) 61
Large-medium mammaL diversity in fragmented habitat: a camera trap inventory
some voluntary conservation of private forested properties, may explain why much of this forest corridor has been preserved.
The community of Isla Alta, within the mu-nicipality of Ybycuí, is situated at the northern terminus of this forest corridor. Isla Alta has a population of around 200 individuals, the major-ity of which engage in small-scale agriculture. Many families also voluntarily preserve about 1-20 ha of native forest along with their land under agricultural production.
The natural vegetation of the study area is typical of the transition zone between UPAF and the humid Chaco ecosystem to the west. The study area is defined as isolated UPAF (Di Bitetti et al., 2003; FCA-UNA, 2015). The UPAF in this area is a dense, semideciduous, subtropical forest intermixed with patches of natural grassland. Vines, bromeliads, and bam-boos are common understory features in this area. Small streams are common in the forests around Isla Alta.
Most of the private forested properties around Isla Alta are used to range a small amount of cattle throughout the year. Many residents also selectively harvest native trees for lumber, firewood, or charcoal production according to personal or financial need. Frequent hunting/trapping occurs around Isla Alta, primarily of nine-banded armadillo (Dasypus novemcinctus), Azara’s agouti (Dasyproctus azarae), coati (Na-sua nasua), and rusty-margined guan (Penelope superciliaris).
Based on interviews with the property own-ers of the camera trap site locations, hunters enter each forested property an average of eight times a month (range: 3-28). The hunters often enter the property without the expressed permission and usually against the wishes of the property owners. Hunters tend to prefer the use of traps over rifles and tend to hunt more in the summer (Oct-Feb). Puma (Puma concolor), and mesopredators such as foxes (Cerdocyon thous) and tegu (Salvator merianae) which are perceived to be a threat to domestic poultry, are
also sometimes killed by the general public if detected.
Data collection One HC600 Reconxy camera trap (Reconyx Inc. Holmen, WI) was used from December 2014 to October 2016 to gather a total of 2,687 photos of animals from the forest surrounding Isla Alta. Twelve camera trap site locations were established opportunistically as appro-priate locations were discovered (Fig. 2). No formal spatial analysis of Isla Alta was made before the study, and locations were not chosen to specifically monitor any species, habitat, or ecological process. The primary intention of using the camera trap was to promote environ-mental awareness amongst local community members. However, camera trapping has been shown to be an effective method for registering biodiversity in tropical ecosystems (Sollmann el at., 2013; Tobler et al., 2008; O’Connell et al., 2011; Srbek-Araujo et al., 2005). Locations, trails, and properties were selected as appropri-ate to capture the most comprehensive record of animal diversity possible.
All camera trap sites were under complete forest cover and set to monitor a cattle, wild-life, or human trail on the private property of a community member of Isla Alta. Camera trap sites were set around the periphery of Isla Alta on average 320m (range: 65-992m) from the nearest residence, and on average 124m (range: 26-360m) from the nearest forest edge (Fig. 2). Maximum distance between sites was 2,500 m. Distances were estimated using Google Earth.
The average duration of camera trap ac-tivity for the sites was 33 camera trap nights (range: 7-80 nights). The camera trap was used continuously throughout the study. Because only one camera trap was available, sites were never monitored simultaneously. The camera trap was set at 50-150cm height on the most suitable available tree for monitoring the target trail. Vegetation was cleared when necessary to provide a full view of the trail being monitored.
Boletín del Museo nacional de Historia natural del Paraguay Vol. 22, Nº 2 (dicieMBre 2018)62
Nicholas GeNGler
RESULTS22 species of wild vertebrates were identified in this study (Table 1). Three domesticated species and humans were also occasionally detected. The number of distinct capture events of each species were recorded. A capture event indicates when an individual animal passed in front of the camera, and the number of capture events does not necessarily indicate the number of photos obtained of that species (i.e. one individual ani-mal while passing the camera one time may have paused in front of the camera for several min-utes triggering the camera trap to take several photos). A gap of 15 minutes between photos of the same species was used as the standard for an independent capture event. If multiple individu-als of the same species appear together in a photo each individual was counted as a capture event (i.e. a flock of three guans appearing together counts for three capture events). The number of difference sites around Isla Alta in which a spe-cies was registered in was also recorded.
DISCUSSION AND CONCLUSIONIt is probable that the 12 species of mammals registered in this study represent the majority of medium-large (>1kg) terrestrial mammal spe-cies present in the study area as newly encoun-tered mammal species tapered off substantially over the course of the study (Figure 3). A few bird and reptile species were also registered incidentally throughout this study. The fact that this study only set the camera to monitor clearly established trails could bias the study to favor species that readily use trails. However, camera traps have shown a clear tendency to register more medium-large mammals on trails than off of them, even for rarely registered species (Reilly, 2015; Kolowski et al., 2017).
Of the species registered around Isla Alta, the margay, greater naked-tailed armadillo, and Azara’s agouti are of particular interest due to their conservation status. Margays (L. wiedii) are classified as Near Threatened by the IUCN and with a decreasing population are likely to
Figure 2. Study Area. Twelve camera trap locations occupying a circular area of roughly 5km². Yellow diamond indicates the center of Isla Alta at 25° 56' 37"S and 56° 46' 47"W, 416m elevation. Map data: Google, CNES/Airbus.
Boletín del Museo nacional de Historia natural del Paraguay Vol. 22, Nº 2 (dicieMBre 2018) 63
Large-medium mammaL diversity in fragmented habitat: a camera trap inventory
classify as Vulnerable in the near future. This feline species is highly arboreal and likely to suffer further from a continued loss of forest cover (Bou, 2012; Henriquez et al., 2002). Greater naked-tailed armadillos (C. tatouay), although not threatened, are generally difficult to register due to their highly fossorial habits (Smith, 2008). Azara’s agoutis (D. azarae) are classified as Data Deficient by the IUCN. These important seed dispersers of Atlantic forest trees are listed as potentially Vulnerable in Argentina but appear to retain a population around Isla Alta and have been recorded throughout the UPAF of
Paraguay (Brooks et al., 1993). None of these three species of interest have been recorded in YNP (FCA-UNA, 2015). However, it is likely that they are in fact present within the park. Their absence from park records might be explained by the fact that no species inventory has been carried out in the park using a camera trap, or at least to date have not been published.
The other species around Isla Alta detected in this study are classified as Least Concern by the IUCN and are generally considered com-mon in other Paraguayan and Brazilian Atlantic Forest fragments, including YNP (FCA-UNA,
Table 1. Species capture events within Isla Alta.
species name common name (English) total encounters number of sites
Salvator merianae black and white tegu 21 6Penelope superciliaris rusty-margined guan 12 7Nyctidromus albicollis pauraque 2 1Crypturellus tataupa tataupa tinamou 41 6Leptotila verreauxi white-tipped dove 246 12Turdus rufiventris rufous-bellied thrush 7 2
domestic chicken 7 1domestic dog 14 5domestic cow 46 4
human 2 1
Boletín del Museo nacional de Historia natural del Paraguay Vol. 22, Nº 2 (dicieMBre 2018)64
Nicholas GeNGler
2015; Brooks et al., 1993; Albuquerque et al., 2013; Rocha et al., 2015; Machado et al., 2016; Hannibal, 2014).
However, no large terrestrial mammals nor-mally common in the UPAF such as brocket deer (Mazama spp.), puma, collared peccary (Pecary tajacu), and tapir (Tapirus terrestris) were recorded in this study. Moreover, residents of Isla Alta anecdotally support these camera trap findings, reporting that roughly two decades ago brocket deer, collared peccary, and puma were regularly encountered in the area, and are now absent. Brocket deer and collared peccary are known to currently persist in YNP (FCA-UNA, 2015), a nearby connected and nearly identical forest habitat.
Other Atlantic Forest fragments in Paraguay have shown a conspicuous decrease of mamma-lian diversity in areas experiencing unregulated hunting pressure (Brooks et al., 1993; Paviolo et al., 2009), and Brazilian Atlantic Forest frag-ments have shown a 37% decrease in bird and mammalian abundance due to heavy hunting pressure, particularly affecting brocket deer and
nine-banded armadillo (Cullen et al., 2000). Consequently, local hunting pressure may be responsible for a substantial reduction, if not the elimination, of these large mammals within the study area around Isla Alta, and may continue to threaten certain species.
This study in Isla Alta and others in the At-lantic Forest show that medium to small-sized forest fragments can still provide habitat for a variety of mammals (Turner, 1996), including some of concern for conservation. These re-maining fragments may also serve as valuable starting points for habitat restoration efforts to reconstruct connectivity between larger frag-ments of the UPAF (Di Bitetti et al., 2003). However, efforts to work with communities to manage local hunting practices around Isla Alta and similar regions may be an important step towards allowing these small fragments to serve as viable habitat for a number of species.
The absence of certain large mammals around Isla Alta demonstrates the vital impor-tance of protected areas as reserves for species susceptible to over hunting.
R² = 0,8932
0
2
4
6
8
10
12
14
0 50 100 150 200 250 300 350 400 450
tota
l >1k
g m
amm
al sp
p ob
serv
ed
Total study camera trap nights
large-medium mammal accumula�on curve
Figure 3. Accumulation curve for medium to large sized mammal species detected throughout the study.
Boletín del Museo nacional de Historia natural del Paraguay Vol. 22, Nº 2 (dicieMBre 2018) 65
Large-medium mammaL diversity in fragmented habitat: a camera trap inventory
ACKNOWLEDGEMENTSThe biggest thank you goes to the residents of Isla Alta for welcoming me into their community during this study. I would also like to thank Paul Smith from Fauna Paraguay and Para La Tierra for his help in this study. Another big thank you to Dr. Michelle Reilly from the United States Fish and Wildlife service and Dr. Paul Beier from Northern Arizona University for the loan of camera trap equipment. Thank you to Lic. Alejandra Gill from WWF-Paraguay for her help with GIS analysis of the study area, and to Sheila McCluskey for her assistance in gathering information on hunting pressure.
LITERATUREAlbuquerque, H. G., Martins, P. G., Pessoa, F. S.,
Modesto, T.C., Luz, J.L., Raíces, D.S.L., Ardente, N.C., Lessa, I.C.M., Attias, N., Jordão-Nogueira, T., Enrici, M.C. & H.G. Bergallo. 2013. Mammals of a forest frag-ment in Cambuci municipality, state of Rio de Janeiro, Brazil. Check List, 9(6): 1505-1509.
Bou, N. 2012. Analisis de la presencia del Margay (Leopardus wiedii) (Carnivora: Felidae) en el area protegida Quebrada de los Cuervos (Uruguay): 12pp.
Brooks, T.M., Barnes, R., Bartrina, L., Butchart, S.H.M., Clay, R.P., Esquivel, E.Z., Etch-everry, N.I., Lowen, J. C. & J. Vincent. 1993. Bird Surveys and Conservation in the Paraguayan Atlantic Forest. Project CANOPY ’92: Final Report: 154pp.
Chiarello, A. G. 2000. Density and Popula-tion Size of Mammals in Remnants of Brazilian Atlantic Forest. Conservation Biology, 14: 1649-1657.
Cullen L.Jr., Bodmer, R.E. & C.V. Pádua. 2000. Effects of hunting in fragments of the Atlantic forests, Brazil. Biological Con-servation, 95: 49-56.
Di Bitetti, M.S., Placci, G. & L.A. Dietz. 2003. A Biodiversity Vision for the Upper Paraná Atlantic Forest Ecoregion: Designing a
Biodiversity Conservation Landscape and Setting Priorities for Conservation Action. Washington, D.C., World Wildlife Fund: 147pp.
FCA-UNA (Facultad de Ciencias Agrarias-Uni-versidad Nacional de Asunción). 2015. Plan de Manejo del Parque Nacional Ybycuí 2015-2025. Asunción: SEAM/FCBT: 210pp.
Fragano, F. & R. Clay. 2003. Biodiversity Status of the Interior Atlantic Forest of Paraguay. State of the Hotspots, The Atlantic Forest of South America Biodiversity Status, Threats, and Outlook: 288-309.
Fundación Vida Silvestre Argentina and WWF (2017). State of the Atlantic Forest: Three Countries, 148 Million People, One of the Richest Forests on Earth. Puerto Iguazú, Argentina: 148pp.
Hannibal, W. 2014. Mammals of medium and large size from a fragmented seasonal forest landscape in Mato Grosso do Sul state, central-western Brazil. Check List, 10 (6): 1430-1434.
Harper, K.A., MacDonald, S.E., Burton, P.J., Chen, J., Brosofske, K.D., Saunders, S.C., Euskirchen, E.S., Roberts, D., Jaiteh, M.S. & P.A. Esseen. 2005. Edge Influence on Forest Structure and Composition in Fragmented Landscapes. Conservation Biology, 19: 768-782.
Henriquez, V.E. & V.G. Ramirez Henriquez. 2002. Preferencia de Habitat y Patrones de Movimiento de Leopardus wiedii, “tigrillo”, en el Parque Nacional El Im-posible, Sector San Benito Ahuachapan. El Salvador. Universidad de El Salva-dor Facultad de Ciencias Naturales y Matematica, Escuela de Biologia: 78pp.
Kolowski, J.M., T.D. Forrester. 2017. Camera trap placement and the potential for bias due to trails and other features. PLoS ONE 12(10): e0186679. https://doi.org/10.1371/journal.pone.0186679: 20pp.
Boletín del Museo nacional de Historia natural del Paraguay Vol. 22, Nº 2 (dicieMBre 2018)66
Nicholas GeNGler
Lowen, J.C., Bartrina, L., Brooks, T.M., Clay, R. & J. Tobias. 1996. Project Yacutinga’95: bird surveys and conservation priorities in eastern Paraguay. Cotinga, (5):14-19.
Machado, F.S., Almeida, A.F., Barros, D.A., Pereira, J.A.A., Silva, R.A. & A.A.S. Pereira. 2016. Diversity of médium-sized and large mammals from Atlantic Forest remnants in southern Minas Gerais state, Brazil. Check List, 12 (5): 1962-1967.
Myers, N., Mittermeir, R.A., Mittermeir, C.G., Fonseca, G.A.B. & J. Kent. 2000. Biodi-versity hotspots for conservation priori-ties. Nature, 403:853-858.
O’Connell, A.F., Nichols, J.D. & K.U. Karanth. 2011. Camera Traps in Animal Ecology, Methods and Analyses. Springer 2011: 263pp.
Paviolo, A., De Angelo, C., Di Blanco, Y., Agos-tini, I., Pizzio, E., Melzew, R., Ferrari, C., Palacio, L. & M.S. Di Bitetti. 2009. Efecto de la caza y el nivel de proteccion en la abundancion de los grandes mam-iferos del Bosque Atlantico de Misiones. Biodiversidad y manejo de recursos na-turales: 237-254.
Reilly, M. 2015. The Effects on Non-Motorized Human Recreation on Mid-Size and Large mammals in the San Francisco Bay Ecoregion. Northern Arizona University: 170pp.
Ribon, R., Simon, J.E. & G.T. Mattos. 2003.
Bird Extinctions in Atlantic Forest Frag-ments of the Viçosa Region, Southeastern Brazil. Conservation Biology, 17: 1827-1839.
Rocha, E.C., Soares, K.L. & I.M. Pereira. 2015. Medium and Large sized animals in Mata Altantic State Park, southeastern Goias, Brazil. Check list, 11 (6): 7pp.
Smith, P. 2008. Greater naked-tailed arma-dillo Cabassous tatouay. Fauna Paraguay Handbook of the Mammals of Paraguay: 9pp.
Sollmann, R., Mohamed, A., Samejima, H. & A. Wilting. 2013. Risky business or simple solution – Relative abundance indices from camera-trapping. Biological Con-servation, 159: 405-412.
Srbek-Araujo, A.C. & A.G. Chiarello. 2005. Is camera-trapping an efficient method for surveying mammals in Neotropical for-ests? A case study in south-eastern Brazil. Journal of Tropical Ecology, 21: 121-125.
Tobler, M.W., Carrillo-Percastergui, S.E., Pitman, R.L., Mares, R. & G. Powell. 2008. An evaluation of camera traps for inventorying large- and medium-sized terrestrial rainforest mammals. Animal Conservation, 11: 169-178.
Turner, I. M. 1996. Species loss in fragments of tropical rain forest: a review of the evidence. Journal of Applied Ecology, 33: 200-209.
