ARTÍCULO ORIGINAL

Food habits of Geoffroy's cat (Leopardus geoffroyi) in agroecosystem habitats of Buenos Aires, Argentina

 

Juan S. Guidobono¹*; Jimena Muñoz¹; Emiliano Muschetto¹; Pablo Teta² & María Busch¹

¹ Departamento de Ecología, Genética y Evolución, Instituto de Ecología, Genética y Evolución de Buenos Aires (CONICET-UBA), Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, C.A.B.A., Argentina.
² Consejo Nacional de Investigaciones Científicas y Tecnológicas, Centro Nacional Patagónico (U9120ACD)Puerto Madryn, Chubut, Argentina.
* jsguidobono@ege.fcen.uba.ar

Editor asociado: Alejandro Farji-Brener

Recibido: 18 de mayo de 2015 Aceptado: 17 de diciembre de 2015

 

---

Abstract

Geoffroy's cat (Leopardus geoffroyi), a small wild cat of South America, showed in the last years a decline in population abundance mainly related to both habitat destruction and hunting. The increasing human population densities and the consequent demand for food and natural resources will rapidly erode and fragment its remaining habitats. In Argentina most studies about Geoffroy's cat diet have been conducted in protected areas or in their surroundings, while there are no studies conducted in purely agricultural systems. We described the diet composition of Leopardus geoffroyi and its seasonal variations in an agricultural ecosystem to contribute knowledge of its ecology and natural history. The study was carried out from spring 2008 to spring 2011. A total of 494 scats were analyzed and at least 695 prey-items were identified. The diet included 81.58% of mammal prey, representing 93.32% of the potential biomass ingested. Within mammals, the principal prey were rodents, representing 79.57% of the total prey items and 59.39% of the total potential biomass consumed. Small rodents account for a great proportion of prey items, while larger species, as the caviid Cavia aperea, contributed more biomass. The diet plasticity may enhance the persistence of Geoffroy's cat populations in variable and anthropized ecosystems.

Keywords:  Small wild cats; Diet; Mammal prey; Agricultural ecosystems

Resumen

Hábitos alimentarios del gato montés (Leopardus geofroyi) en agroecosistemas de Buenos Aires, Argentina. La abundancia del gato montés (Leopardus geofroyi), un pequeño felino silvestre sudamericano, se ha reducido principalmente por la destrucción de su hábitat y por la caza. En la Argentina, la mayoría de los estudios sobre la dieta del gato montés se llevaron a cabo en áreas protegidas y en sus alrededores, mientras que no existen estudios en sistemas puramente agrícolas. Describimos la composición de la dieta de L. geofroyi y sus variaciones estacionales en un sistema agrícola con el fn de contribuir al conocimiento de su ecología e historia natural. Desde la primavera de 2008 hasta la primavera de 2011 analizamos un total de 494 heces e identificamos 695 ítems presa. La dieta incluyó 81.6% de mamíferos, lo cual representó 93.3% de la biomasa potencialmente ingerida. Entre los mamíferos, la presa principal fueron los roedores, que representaron 79.6% del total de los ítems presa y 59.4% del total de la biomasa potencialmente ingerida. Los pequeños roedores significaron una gran proporción de los ítems presa, mientras que las especies de mayor tamaño, como el cávido Cavia aperea, contribuyeron mayormente en función de la biomasa ingerida. La plasticidad de la dieta del gato montés podría permitir la persistencia de sus poblaciones en ecosistemas variables y antropizados.

Palabras clave: Pequeños felinos silvestres; Dieta; Mamíferos presa; Ecosistemas agrícolas

---

 

Introduction

Global analyses show that among terrestrial mammals, many carnivores are the most threatened, and many within the families Felidae (Fischer de Waldheim 1817), Canidae (Fischer 1817) and Ursidae (Fischer de Waldheim 1817) appear to be suffering the severest population declines and range contractions (Ceballos et al. 2005; Schipper et al. 2008). One of the causes may be the human intolerance of such predatory carnivores which is frequently based on misconceptions about the potential risk they pose to livestock and humans (Treves & Karanth 2003). Additionally, other species are threatened because they occupy unique ecological niches, or because they are habitat or diet specialists (Karanth & Chellam 2009).

Populations of vulnerable carnivore species are unable to persist under human land uses that engender conflicts over habitat, space or prey (Treves & Karanth 2003). The increasing human population densities and the consequent demand for food and natural resources will rapidly erode and fragment remaining carnivore habitats. A common consequence of such conflict has been the elimination of carnivores from human-dominated landscapes (Karanth & Chellam 2009). In the Pampean region, the increase of agriculture produced a decrease in natural habitat areas, which frequently remain as linear corridors along roadsides and railways (Bonaventura et al. 1988; Ellis et al. 1997; Busch et al. 2001). One of the most important changes observed during the 20th century was the replacement of livestock fields by crops, with the consequent fragmentation of natural habitats. According to some estimates in the Rolling Pampas, approximately 90% of the land is devoted to crops as soybean, wheat, corn and sunflower (Viglizzo et al. 2001; Paruelo et al. 2005). These habitat changes may affect prey availability for predators due to changes in species identity, relative abundances and vulnerability (Fedriani et al. 2001; Fedriani & Kohn 2001; Sovada et al. 2001; Pia et al. 2003; Farias & Kittlein 2008). The effect of land use may vary according to the predator species considered, its habitat use, food requirements and degree of specialization. While some species may decrease their hunting efficiency in agricultural habitats, other species may be benefited by food subsidies associated to increases in prey abundance as a consequence of human activities (Mukherjee et al. 2004).

Land use changes may be reflected in the diet composition of predators, but although diet and its variations are the most studied aspects of the ecology of carnivore predators (e.g., Pereira et al. 2006; Farias & Kittlein 2008; Radović & Kovačić 2010; Bisceglia et al. 2011; Pereira et al. 2011; 2012; Trigo et al. 2013) most studies were conducted in natural or low disturbed habitats. Studies about foraging behavior and diet composition of carnivores in urban or agricultural systems are less frequent (e.g., Gehrt 2004; Biró et al. 2005; Eagan et al. 2011; Krauze-Gryz et al. 2012). In Argentina, most studies about carnivores diet have also been conducted in protected areas or in their surroundings, with natural vegetation and low human intervention (e.g., Vuillermoz 2001; Canepuccia et al. 2007; Bisceglia et al. 2008; Varela et al. 2008; Pereira 2009; 2010; Bisceglia et al. 2011; Pereira et al. 2012; Santillán et al. 2014), although studies in small reserves or in areas with different levels of human intervention have quantified the effect of livestock and agriculture on the diet of carnivores (Pia et al. 2003; Farias & Kittlein 2008; Pereira 2009; Pereira et al. 2012).

The Geoffroy's cat (Leopardus geoffroyi, d´Orbigny and Gervais, 1844) is a solitary, primarily nocturnal small felid, distributed in South America from southern Bolivia and the Paraná basin of southern Brazil to the southern tip of Patagonia (Nowell & Jackson 1996; Cuellar et al. 2006; Canepuccia et al. 2007). The Geoffroy's cat is a habitat generalist, occupying open woodland, bushy areas, open savannas, marshes (Cuellar et al. 2006; Canepuccia et al. 2007, 2008; Caruso et al. 2012), river deltas (Pereira et al. 2005) and tall grasslands (Manfredi et al. 2006). This species usually disposes its faeces in latrines placed in conspicuous sites, over trees, rocks, or clumps of grasses (Manfredi et al. 2006; Soler et al. 2009). L. geoffroyi has been described as an opportunistic predator that consumes mainly mammals, especially small rodents and medium to big sized birds (Manfredi et al. 2004; Bisceglia et al. 2008; Canepuccia et al. 2007; Sousa & Bager 2008). It can also prey on larger prey as hares (Sousa & Bager 2008). There are no studies about the Geoffroy's cat diet conducted exclusively in agricultural systems, although Castillo et al. (2008) registered its presence in these habitats in the Argentine pampas. Our aim was to describe the diet composition of L. geoffroyi and its seasonal variations in an agricultural ecosystem to contribute knowledge of its ecology and its natural history.

Materials and Methods

Study Area

We studied the food habits of Geoffroy's cat in agroecosystems in Exaltación de la Cruz Department (34°19' S and 59°14' W), Buenos Aires province, Argentina. The study area is located in the Rolling Pampas, eco-region that includes part of Buenos Aires, Córdoba and Santa Fe provinces, within banks of La Plata and Paraná rivers at northeast, Salado River at southwest and Matanza river at southeast. This region presents a highly rolling topography in comparison with other neighbor areas, forming an exoreic drain system (Bilenca & Miñarro 2004). The climate is sub-humid temperate, with a mean annual temperature of 16 °C and an annual precipitation ~1000 mm. Originally, the area was covered by grasslands, with a vegetation structure that corresponded to a prairie in humid years and pseudo-steppe or steppe during dry years (Soriano et al. 1991). The vegetation was represented by tall perennial grasses, as Nassella spp. (Trin., É. Desv.) and Piptochaetium spp. (J. Presl.). Nowadays, 90% of the land is devoted to crops as soybean, wheat, corn and sunflower (Viglizzo et al. 2001; Paruelo et al. 2005), so the landscape is a matrix of cropfields with grassland corridors along their edges (borders) and a mixture of exotic and native woody species forming woodlots. Along borders, a spontaneous and particular flora had developed (Soriano et al. 1991), dominated by species such as Nasellaneesiana (Trin. & Rupr., Barkworth), Jaravaplumosa (Spreng., S.W.L. Jacobs & J. Everett), Paspalum dilatatum (Poiret), Bromus catharticus (Vahl), the forbs Solidago chilensis (Meyen) and Senecio grisebachii (Bacher) and the thistles Carduus acanthoides (Linnaeus), Cirsium vulgare (Savi and Tenore) and Cynara cardunculus (Linnaeus) (Bonaventura & Cagnoni 1995; Bilenca & Kravetz 1998). Woodlots are small patches of about 0.5 to 2.5 ha composed by trees of about 5 m high, and their saplings. Woodlots are highly variable according to the dominant species of tree. The most frequent species are eucalyptus (Eucalyptus sp., Smith), ombú (Phytolacca dioica, Linnaeus), privet (Ligustrum sp., Linnaeus), willows (Salix spp., Linnaeus) and the honey locust (Gleditsia triacanthos, Linnaeus). Seasonal changes in plant phenology and in the stage of development of crops caused seasonal qualitative and quantitative variations in resources, both in cropfields and borders (Busch et al. 1997; Hodara & Busch 2006). The borders have abundant vegetation cover throughout the year, while cropfield cover varies from low cover after ploughing and sowing to high cover when crops are mature (Busch et al. 1997; Hodara & Busch 2006).

According to studies developed in and around the study area, potential mammalian prey species for the Geoffroy's cat in the study area include sigmodontine, caviid and commensal rodent species (Busch & Kravetz 1992; Bilenca & Kravetz 1995; Busch et al. 2000; Hodara & Busch 2006; Bilenca et al. 2007). Other potential preys for the Geoffroy's cat are European hares (Lepus europaeus, Pallas 1778), opossums and many passerines and raptor species. Geoffroy's cat eventually consumed small reptiles, amphibians and insects (Sousa& Bager 2008).

Sampling Design

Diet. Sampling was carried out seasonally from November 2008 to December 2011 (a total of 26 sampling events). In each sampling, an area of 300 km² was covered during 3 days, searching Geoffroy's cat latrines and scats within woodlots, roads and railways. Cropfields and its borders were discarded because very few latrines or scats were found during preliminary samplings. Characteristics of woodlots and surrounding fields were recorded. In woodlots Geoffroy's cat faeces were found on medium and large-sized trees, like Ombú and Eucalyptus. Tree-forks were 1-2 m high and, in general, trees were pruned or broken, generating additional surfaces that were used by Geoffroy's cat. Woodlots with juvenile trees did not present Geoffroy's cat latrines. Faeces were assigned to L. geoffroyi by size and form, since there are no other wild felids with similar weight and aspect of faeces in northern Buenos Aires province. Although the domestic cat is present in the area, it is usually found near human houses, while Geoffroy's cat faeces were found in small woodlots distant from human dwellings. On the other hand, domestic cats bury or semibury their scats (Lozano & Urra 2007), while Geoffroy's cat deposes its scats in conspicuous sites (Soler et al. 2009). In road and railways, we found both L. geoffroyi and Pampas Fox (Lycalopex gymnocercus) scats, which were differentiated by form.

Rodent community. Rodent samplings were conducted simultaneously with fecal samplings in three cropfields and their borders of the study area placed within the home range area of Geoffroy's cats detected by the presence of latrines. Sampling sites were located at a distance greater than 200 m in order to give independent estimates of rodent abundance and species composition. At each site, we placed 25 trapping stations spaced at 10 m intervals in one border and another two lines of 25 trapping stations in the field perpendicular to the border trapping line. There was one Sherman trap (23x7.7x9.1 cm) at each trapping station, which was baited with a mixture of peanut butter, bovine fat and rolled oats. Traps were checked every morning for three consecutive days. Animals were handled according to the 14346 Argentina National law of animal care and ASM guidelines (Sikes et al. 2011). Captured individuals were identified to species level, individually marked with ear tags, and released at the capture site.

Data analysis

The diet composition of Geoffroy's cat was determined on basis of prey remains, like bones, feathers and insect remains present in scats. We considered three levels for classification of prey: prey type, prey item and prey species. The first level (prey types) included "Mammals", "Birds", "Reptiles", "Insects","Unidentified vertebrates" and "Unidentified remains". The second level (prey items) for mammalian types was differentiated according to order in "Rodents", "Marsupials" and "Lagomorphs". For birds we differentiated between "Passerines" and "Raptors" (Falconiforms and Strigiforms). Reptile prey items were classified as "Scaled reptiles". Insect remains were not differenti­ated at any level. The third level (prey spe­cies) was differentiated according to genus or species, if possible.

Since sample sizes for some of the seasons were low, the data from the three years of study were combined, obtaining one estimative value of the contribution of each prey category to the diet per season. The number of individuals of each prey category per season was also added in order to estimate the global contribution along the studied period.

Except for insects, which were only registered as present or absent in the diet, the quantification of the contribution of the rest of the prey categories to the diet was analyzed in function of its occurrence in the scats and the potential biomass ingested, in order to consider the large variation in size of prey categories. The occurrence of each item was quantified as a percentage of occurrence (PO, Equation 1). We used this term following Sousa & Bager (2008) and Pereira et al. (2012) in order to allow comparisons, where ni is the number of individuals of prey category i (Sousa & Bager 2008).

PO = (n_i / Σn_i) x 100 Equation 1

The number of individuals of each prey category found in the fecal pellets was determined from the number of jaws and teeth for mammals; peaks, claws and large bones for birds; and scales and bones for reptiles (assuming a single individual in the scat from the presence of scales and bones). The presence of insects was recognized from elytra and heads.

The estimation of the potential biomass ingested (PBI, Equation 2) was calculated for each prey species following Figueroa Rojas & Corales Stappung (2004) and Santos Moreno & Alfaro Espinola (2009), where PBIi is the percentage potential biomass ingested of prey species i, PO_i is the percentage of occurrence of prey species i and w_i is the mean mass of prey species i. PBI for each prey item was calculated as the sum of its corresponding PBI prey species. PBI for each prey type was calculated as the sum of its corresponding PBI prey items.

PBI%_i = [(PO_i x wi) / ∑(PO_i x w_i)] x 100 Equation 2

Body mass of mammalian and reptile prey species was obtained from literature (Miño et al. 2001; Gómez Villafañe et al. 2004; Fraschina 2011; Palacios et al. 2012). For birds, the mean body mass of passerines and raptors was estimated considering the body mass of the most frequent species in the study area, according to Naroski & Yzurieta (2012).

Statistical analysis

Variation in diet composition along seasons (spring, summer, autumn and winter) was evaluated by a Log-Linear test (Norman & Streiner 1996), using the R software version 2.11.1 (R Development Core Team 2010). In this test, the significance of the main effects evaluate if marginal values are different, while the interaction between prey item and season analyze if there is a variation in diet composition along the year.

Variation in diet composition was analyzed at both prey item and species level. Since the number of individuals (n_i) for some of the prey items (Marsupials, Passerines, Raptors and Scaled reptiles) was below the minimum required for statistical analysis (n=5), they were grouped in new categories as follows: Passerines and Raptors as "Birds" and Marsupials and Scaled reptiles as "Others". Unidentified vertebrates were also included in the analysis under the category "Others". Thus, the categories considered were: Rodents, Lagomorphs, Birds and Others. "Unidentified remains" was not included in this analysis. At prey species level, the only ones suitable for comparison were the rodent species.

To estimate the degree of feeding specialization of Geoffroy's cat, the breadth of the food niche in each sampling was calculated using the "B" index of Levins (1968), standardized according to Cowell & Futuyma (1971) "Bsta", and averaged by season. This index varies between 0 (when only one prey is consumed) and 1 (all preys are equally consumed). Niche values close to 1 mean a uniformly distributed diet, with no single prey predominating (i.e., generalist diet). Values close to 0 indicate that few preys are consumed in high proportions (i.e., specialist diet, Sousa & Bager 2008). The index was calculated in function of PO for each prey species, except for birds where PO for prey type was used. All unidentified categories were not included in this analysis. In order to compare Bsta values among seasons we estimated confidence intervals using InfoStat statistical program in its 2014 version (Di Rienzo et al. 2014).

Results

Diet

A total of 494 scats were analyzed and at least 695 remains were identified. Most scats were found under trees or on their forks within small woodlots, while few were found along railways or roadsides. Considering the global contribution, 89.5% of the diet remains was identified. The main prey type consumed were mammals, followed by birds. Reptiles were poorly represented while insects were present in all sampling periods, although few remains per scat were found (Table 1). Diet remains that could not be identified (10.5%) corre­sponded to unidentified vertebrates (5.9%) and unidentified remains (4.6%). The global diet included 81.6% of mammalian prey, representing 93.3% of the potential biomass ingested. Within mammals, the principal prey were rodents, representing 79.6% of the total prey items and 59.39% of the total potential biomass consumed (Table 1), while lagomorphs represented only 1.7% of the total number of items found but 32.2% of the potential biomass ingested. Marsupials were the least consumed item representing 0.3% of total items and 1.7% of PBI. Considering bird remains, 6.9% were assigned to Passerines and 0.72% to Raptors, representing similar values of potential biomass ingested (0.97% and 1.01% respectively). Reptiles accounted for 0.29% of the remains, representing 4.70% of the potential biomass ingested.

Table 1. Diet composition of Leopardus geoffroyi during the three studied years. The values in the table represent the percent numerical frequency (PO) and the potential biomass ingested (PBI) of each prey item in the bulk of faeces collected in each season. Values on Bsta confidence interval row are the niche breadth. Numbers in parentheses (n) in the first line indicate the number of scats analyzed at each season. For insects, p=presence, a=absence.
Tabla 1. Composición de la dieta de Leopardus geoffroyi durante los tres años estudiados. Los valores en las tablas representan la frecuencia numérica porcentual (PO) y la biomasa potencialmente ingerida (PBI) de cada ítem presa en el total de heces recolectadas en cada estación del año. Los valores del intervalo de confianza en la línea "Bsta" corresponden a la amplitud de nicho. Los números entre paréntesis (n) en la primera línea indican el número de heces analizadas en cada estación del año. Para los insectos, p=presencia, a=ausencia

Rodent remains were assigned to seven categories of prey species/genus: Calomys spp. (that includes the non-distinguishable remains of Calomys laucha and Calomys musculinus), Rattus spp. (that includes the non-distinguishable remains of Rattus nowegicus and Rattus rattus), Akodon azarae, Oligoryzomys flavescens, Holochilus vulpinus, Cavia aperea and "Unidentified rodents". Among them, Calomys spp. showed the highest percentage of occurrence in the diet (29.21%), followed by C. aperea (13.96%). Each of the remaining rodent categories accounted for less than 10% of the total number of items (Table 1). Considering the potential biomass ingested, C. aperea was the most important prey specie (42.04%), followed by Rattus spp. (12.30%). Each of the remaining rodent categories represented less than 3% of the potential biomass ingested (Table 1). The remains of marsupials and lagomorphs were assigned to Didelphys albiventris and L. europaeus, respectively. Salvator merianae (Duméril & Bibron, 1839) was the only Scaled reptile species found.

The proportion of prey items in the diet varied among seasons (Prey-item x Season: Dev=22.79, n=16, df=9, P<0.01, log-linear analysis) (Table 2). Rodents were the most consumed prey item throughout the year, followed by Birds. Lagomorphs and Others were the least consumed items. Geoffroy's cat consumed the two last items in similar numbers in spring, but Lagomorphs were more consumed in autumn, and Others were more consumed in summer and winter. For rodents, prey species were consumed differentially among seasons (Prey-species x Season: Dev=58.41, n=20, df=12, P<0.01, log-linear analysis) (Table 3). Calomys spp. was the most consumed prey species throughout the year, except in winter when the number of consumed individuals (ni) of Calomys spp. and C. aperea were similar.

Table 2. Log-Linear results for the variation in diet composition of Leopardus geoffroyi (rodents, birds, lagomorphs and others) according to the seasons (Spring, Summer, Autumn and Winter).
Tabla 2. Resultados del análisis Log-Linear para la variación de la composición de la dieta de Leopardus geoffroyi (roedores, aves, lagomorfos y otros) en función de las estaciones del año (primavera, verano, otoño, invierno).

Table 3. Log-linear results for the variation in the proportion of different rodents consumed by Leopardus geoffroyi (Akodon azarae, Calomys spp., Oligoryzomys flavescens, Cavia aperea, Rattus spp.), according to the season (Spring, Summer, Autumn and Winter).
Tabla 3. Resultados del análisis log-linear para la proporción de los diferentes roedores consumidos por Leopardus geoffroyi (Akodon azarae, Calomys spp., Oligoryzomys flavescens, Cavia aperea, Rattus spp.) en función de las estaciones del año (primavera, verano, otoño, invierno)

Considering the non-rodent prey species, the marsupial D. albiventris was only consumed in summer, while the lagomorph L. europaeus was ingested in all seasons except in winter. Although the PO representation of these mammalian species was poor in all seasons (PO_i<2%), in terms of potential biomass L. europaeus represented the 29.27%, 47.35% and 45.93% in spring, summer and autumn, respectively. The marsupial D. albiventris reached 9.86% of the PBI when it was consumed in summer (Table 1). Among birds, passerines were consumed throughout all seasons, representing less than 12% of the PO and less than 2% of the PBI. Raptors were only ingested in spring and winter reaching 0.63% and 2.35% of the PO, and 0.73% and 3.97% of the PBI in spring and winter, respectively. The reptile S. merianae was consumed only in spring and represented 1.26% of the PO and 17.07% of the PBI in the diet in this season (Table 1).

In relation to the degree of feeding specialization of Geoffroy's cat, the Bsta index showed that the diet was narrower in summer, autumn and winter than in spring, when it reached the highest breadth (Table 1). Confidence intervals for this index never reached the most extreme values (i.e., 0 and 1), suggesting an intermediate level of specialization.

Rodent community

Along the 16 sampling sessions, we captured a total of 344 individual rodents (218 A. azarae, 107 Calomys spp. 84 C. laucha and 23 C. musculinus and 19 O. flavecens). Calomys spp. represented more than 50% of the captures all along the study, although their representation decreased from spring- summer to autumn-winter, along with an increase in A. azarae and O. flavescens. Of the 2 species of the genus Calomys, most captures corresponded to C. laucha (C. musculinus did not exceed 1%).

Discussion

Most studies about the diet of carnivores have been conducted in protected areas or in their surroundings, with natural vegetation and low human intervention. In the present work, we studied the Geoffroy's cat diet exclusively in agricultural systems. Mammals, and specially rodents, were the main prey of the Geoffroy's cat in the studied agroecosystems. Among rodents, Calomys spp. were the most consumed species followed by C. aperea. However, in terms of PBI, the latter was the most important prey species. Similar food habits of this small felid were observed in Pampas grassland (Vuillermoz 2001; Manfredi et al. 2004), Monte scrubland (Bisceglia et al. 2008; 2011; Pereira et al. 2012), the Patagonian steppe (Novaro et al. 2000) and in Rio Grande do Sul fields (Trigo et al. 2013), while in wetlands the main prey were waterbirds (Canepuccia et al. 2007). Both according to Bisceglia et al. (2008) and in this work, the consumption of small mammals was higher than 60%, with low representation of European hares, contrasting to the diet in other regions as the Chilean and Argentina Patagonian steppes, where the percentages of occurrence of hares ranged from 22.8 to 57.4% (Johnson and Franklin 1991; Novaro et al. 2000). On the other hand, Manfredi et al. (2004), found that European hares are frequent in the diet of Geoffroy's cat with an occurrence that varied from 5% to 16%. Nevertheless, in terms of rodent consumption our results agree with the findings of Manfredi and co-authors (2004).

Relative importance of birds in the diet of Geoffroy's cat changed throughout the year, increasing during spring and summer. These seasonal variations were also observed by Vuillermoz (2001) in Pampean grasslands of Argentina, Bisceglia et al. (2008) for the Monte scrublands of Argentina and Sousa and Bager (2007) for southern coastal plains in Brazil. These authors, however, found a higher rep­resentation of birds in the cat diet with respect to our study. On the other hand, Canepuccia et al. (2007) observed a high representation of waterbirds in the cat diet in a coastal lagoon of the Pampean grassland, probably as a con­sequence of the local offer of food.

Although previous studies in the study area reported strong seasonal variations in abundance of rodents (Miño et al. 2001; Gómez Villafañe et al. 2004; Busch & Hodara 2006; Fraschina 2011), both their percentage of occurrence in the cat diet and the potential biomass ingested of each prey category did not show great variations. These results suggest that rodents are the basis of the cat diet independently of their abundance variations, as was observed by Pereira et al. (2012) who proposed that Geoffroy's cats is not opportunistic predator with respect to sigmodontines. These authors concluded that, in ranches of Monte scrublands, the cat positively selects rodents throughout the year at the expense of other potential preys, such as hares or birds. This behavior could be related to the active search behavior, one of the hunting strategies used by small and medium sized felids (Emmons 1987; Jedrzejewski et al. 2002). Such behavior was observed in the study of Pereira et al. (2012), which determined an increase in the movement rate of Geoffroy's cats as an attempt to increase the encounter rate with prey, which appeared to be less abundant due to the human activities.

The high representation of Calomys spp. in the scats and its relative abundance in the local rodent community suggests that these rodents are consumed opportunistically. Rodents that use open habitats, like Calomys spp., suffer a periodic loss of vegetation cover in their habitat that could increase its vulnerability, facilitating their capture. Geoffoy's cats, as well as lions (Panthera leo) (Hopcraft et al. 2005), Canada lynxes (Lynx Canadensis) (Fuller et al. 2007) and leopard cats (Prionailurus bengalensis) (Rajaratnam et al. 2007), prefer to hunt in habitats where prey are easier to capture (Pereira et al. 2012). Such an opportunistic behavior in the Geoffroy´s cat was already proposed by other authors (Bisceglia et al. 2008; Novaro et al. 2000; Manfredi et al. 2004; Sousa & Bager. 2008).

The composition of the diet suggests that the Geoffroy´s cat hunts not only in fields (where Calomys spp. are more abundant) but also along other habitats such as field borders, where C. aperea, A. azarae and O. flavescens are found, and in woodlots, where O. flavescens is the most frequent rodent species (Fraschina 2011). The use of forested habitats like woodlots was reported for L. geoffroyi by Johnson & Franklin (1991), Manfredi et al. (2006) and Trigo et al. (2013). The Geoffoy's cat seems to be able to use many different habitats, with documented use of habitats greatly altered by deforestation, agriculture, ranching and planting of exotic trees (Trigo et al. 2013). On the other hand, the low percentage of prey typical of domestic and peridomestic habitats (as rats, house mice and poultry birds) suggests that the Geoffroy´s cat does not frequently hunt in these habitats, although rats were present in all seasons in the diet.

The diet of the Geoffroy's cat reflects consumption oriented to rodents. However, it is generalist in relation to the particular mammal/rodent species consumed. Differences in the niche breadth between spring and the other seasons were consistent with an opportunistic feeding behavior that increase the number of prey consumed according to their availability when the dominant is in low abundance (Elmhagen et al. 2002). An alternative point of view may be that the wider diet in spring is consequence of a change in habitat use, with a higher use of borders when cropfields are with low plant cover, before the development of summer crops.

In spite of the high representation of Calomys spp. remains in the diet, the main prey in terms of biomass ingested were C. aperea and Rattus spp. These species are the largest rodents in the study area (mean body weight=300 g for Rattus spp. and 645 g for C. aperea, in contrast to 20 g for the smaller rodent species), and may be selected because represent a more profit­able prey than smaller rodent species. Rattus spp. is rare in cropfield habitats frequented by Geoffroy´s cats (Mills et al. 1991; Busch & Kravetz 1992; Fraschina et al. 2011), but cavies are common along field and road edges (Mills et al. 1991). Considering the potential biomass ingested and the percentage of occurrence, C. aperea is a fundamental prey in the diet com­position of the Geoffroy's cat in agricultural areas. A similar conclusion was suggested by Palacios et al. (2012) to these felids, but for individuals of the Auca Mahuida Provincial Reserve of Neuquen province, Argentina.

The variable diet composition of L. geoffroyi among different studies suggest trophic adaptability of this species, which seems to be able to adjust its predatory behavior to exploit locally abundant food resources (Manfredi et al. 2004; Canepuccia et al. 2007; Palacios et al. 2012; Pereira et al. 2012; Trigo et al. 2013). Consequently, L. geoffroyi, may not be actually limited by food. Further studies about its ecology in human modified habitats are needed in order to assess the risk for its conservation.

Acknowledgements. We are greatly indebted to the cropfield owners who allowed us to work in their properties. We would like to thank Eliana Melignani for the revision of the manuscript and we also thank the comments of the anonymous reviewers that helped to improve the manuscript. This work was funded by Universidad de Buenos Aires.

References

1. Bilenca, DN & FO Kravetz. 1995. Patrones de abundancia relativa en ensambles de pequeños roedores de la región pampeana. Ecol Austral, 5(1):21-30.         [ Links ]

2. Bilenca, DN & FO Kravetz. 1998. Seasonal variations in microhabitat use and feeding habits of the pampas mouse Akodon azarae in agroecosystems of central Argentina. Acta Theriol, 43(2):193-205.         [ Links ]

3. Bilenca, DN & F MiNarro. 2004. Áreas valiosas de pastizal (AVPs) en las pampas y campo de Argentina, Uruguay y sur de Brasil. Fundación Vida Silvetre Argentina. Buenos Aires.         [ Links ]

4. Bilenca, DN; CM González-Fischer; P Teta & M Zamero. 2007. Agricultural intensification and small mammal assemblages in agroecosystems of the Rolling Pampas, central Argentina. Agr Ecosyst Environ, 121(4):371-375.         [ Links ]

5. Biró, Z; J Lanszki; L Szemethy; M Heltai & E Randi. 2005. Feeding habits of feral domestic cats (Felis catus), wild cats (Felis silvestris) and their hybrids: trophic niche overlap among cat groups in Hungary. / Zool, 266(2):187-196.         [ Links ]

6. Bisceglia, SBC; JA Pereira; P Teta & RD Quintana. 2008. Food habits of Geoffroy's cat (Leopardus geoffroyi) in the central Monte desert of Argentina. / Arid Environ, 72(6):1120-1126.         [ Links ]

7. Bisceglia, SBC; JA Pereira; P Teta & RD Quintan a. 2011. Rodent selection by Geofroy's cats in a semi-arid scrubland of central Argentina. J Arid Environ, 75(11):1024-1028.         [ Links ]

8. Bonaventura, SM; MI Bellocq & FO Kravetz. 1988. Selección de hábitat por roedores en campos de cultivo. Physis, 46:61-66.         [ Links ]

9. Bonaventura, SM & MC Cagnoni. 1995. La vegetación de los bordes de caminos en agroecosistemas. Physis (Buenos Aires Sección C), 50:63-71.         [ Links ]

10. Busch, M & FO Kravetz. 1992. Competitive interactions among rodents (Akodon azarae, Calomys laucha, C. musculinus and Oligoryzomys favescens) in a two habitat system. I. Spatial and numerical relationships. Mammalia, 56(1):45-56.         [ Links ]

11. Busch, M; MR Álvarez; AE Cittadino & FO Kravetz. 1997. Habitat selection and interspecific competition in rodents in pampean agroecosystems. Mammalia, 61(2):167-184.         [ Links ]

12. Busch, M; MH Miño; JR Dadon & K Hodara. 2000. Habitat selection by Calomys musuclinus (Muridae, Sigmodontinae) in crop areas of the pampean region, Argentina. Ecol Austral, 10(1):15-16.         [ Links ]

13. Busch, M; MH Miño; JR Dadon & K Hodara. 2001. Habitat selection by Akodon azarae and Calomys laucha (Rodentia, Muridae) in pampean agroecosystems. Mammalia, 65(1):167-184.         [ Links ]

14. Busch, M & K Hodara. 2006. Uso y selección de hábitat y competencia interespecífica en roedores sigmodontinos de la región pampeana. Pp. 147-172 in: Polop, J & M Busch (eds.). Biología y ecología de pequeños roedores en la región pampeana de Argentina. Editorial de la Universidad Nacional de Córdoba, Córdoba, Argentina.         [ Links ]

15. Castillo, D; E LuengosVidal; M Lucherini & E Casanave. 2008. First report on the Geoffroy's cat in a highly modified rural area of the Argentine pampas. Cat News, 49:27-28.         [ Links ]

16. Canepuccia, AD; MM Martínez & AI Vassallo. 2007. Selection of waterbirds by Geoffroy's cat: Effects of prey abundance, size, and distance. Mam Biol, 72(3):163-173.         [ Links ]

17. Canepuccia, AD; AA Farias; AH Escalante; O Iribarne; A Novaro; et al. 2008. Differential responses of marsh predators to rainfall- induced habitat loss and subsequent variations in prey availability. Can J Zool, 86(5):407-418.         [ Links ]

18. Caruso, N; M Sotelo & EM Luengo Vidal. 2012. Primer registro de presencia del gato montés, Leopardus geoffroyi, en la reserva natural Bahía Blanca, Bahía Falsa, Bahía Verde (RNBBBFBV), provincial de Buenos Aires. BioScriba, 5(1): 54-59.         [ Links ]

19. Ceballos, G; PR Erlich; J Soberon; I Salazar & JP Fay. 2005. Global mammal conservation: What must we manage? Science, 309(5734):603-607.         [ Links ]

20. Colwell, RK & DJ Futuyma. 1971. On the measurement of niche breadth and overlap. Ecology, 52(4):567-576.         [ Links ]

21. Cuellar, E; L Maffei; R Arispe & A Noss. 2006. Geoffroy's cats at the northern limit of their range: activity patterns and density estimates from camera trapping in Bolivian dry forests. Stud Neotrop Fauna E, 41(3):169-177.         [ Links ]

22. Di Rienzo, JA; F Casanoves; MG Balzarini; L González; M Tablada; et al. 2014. InfoStat versión 2014. Grupo InfoStat, FCA, Universidad Nacional de Córdoba, Argentina. URL: http://www.infostat.com.ar.         [ Links ]

23. Eagan, TS; JC Beasley; ZH Olson & OE Rhodes. 2011. Impacts of generalist mesopredators on the demography of small-mammal populations in fragmented landscapes. Can J Zool, 89(8):724-731.         [ Links ]

24. Ellis, BA; JN Mills; JE Childs; MC Muzzini; KT McKee, Jr; et al. 1997. Structure and floristics of habitats associated with fve rodent species in an agroecosystem in Central Argentina. J Zool, 243(3):437-460.         [ Links ]

25. Elmhagen, B; M Tannerfeld & A Angerbjörn. 2002. Food niche overlap between arctic and red foxes. Can J Zool, 80(7): 1274- 1285.         [ Links ]

26. Emmons, LH. 1987. Comparative feeding ecology of felids in a Neotropical rainforest. Behav Ecol Sociobiol, 20(4):271 -283.         [ Links ]

27. Farias, AA & KJ Kittlein. 2008. Small-scale spatial variability in the diet of pampas foxes (Pseudalopex gymnocercus) and human-induced changes in prey base. Ecol Res, 23(3):543-550.         [ Links ]

28. Fedriani, JM; TK Fuller & RM Sauvajot. 2001. Does availability of anthropogenic food enhance densities of omnivorous mammals? An example with coyotes in southern California. Ecography, 24(3):325-331.         [ Links ]

29. Fedriani, JM & MH Kohn. 2001. Genotyping faeces links individuals to their diet. Ecol Lett, 4(5):477-483.         [ Links ]

30. Figueroa Rojas, RA & S Corales Stappung. 2004. Summer diet comparison between the American kestrel (Falco sparverius) and Aplomado falcon (Falco femoralis) in an agricultural area of Araucania, southern Chile. Hornero, 19:53-60.         [ Links ]

31. Fraschina, J. 2011. Efectos de cambios en el uso de la tierra sobre ensambles de roedores en agroecosistemas pampeanos. Dissertation, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires.         [ Links ]

32. Fraschina, F; VA León & M Busch. 2011. Long-term variations in rodents abundance in a rural landscape of the Pampas, Aregentina. Ecol Res, 27(1):191-202.         [ Links ]

33. Fuller, AK; DJ Harrison & JH Vashon. 2007. Winter habitat selection by Canada lynx in Maine: prey abundance or accessibility? J Wildlife Manage, 71(6):1980-1986.         [ Links ]

34. Gehrt, SD. 2004. Ecology and Management of Striped Skunks, Raccoons, and Coyotes in Urban Landscapes. Pp 81-104 in: Fascione, N; A Delach & M Smith (eds.). People and Predators: From Confict to Coexistence. Island Press Washington, DC, USA.         [ Links ]

35. Gómez Villafañe, IE; F Miñarro; M Ribicich; CA Rossetti; D Rossotti; et al. 2004. Assessment of the risks of rats (Rattus norvegicus) and opossums (Didelphis albiventris) in different poultry-rearing areas in Argentina. Braz J Microbiol, 35(4): 359-363.         [ Links ]

36. Hodara, K & M Busch. 2006. Return of preferred habitats (edges) as a function of distance in Akodon azarae (Rodentia, Muridae) in cropfeld-edge systems of central Argentina. J Ethol, 21(2):141-145.         [ Links ]

37. Hopcraft, JGC; ARE Sinclair & C Packer. 2005. Planning for success: Serengeti lions seek prey accessibility rather than abundance. J Anim Ecol, 74(3):559-566.         [ Links ]

38. Jedrzejewski, W; K Schmidt; H Okarma & R Kowalczyk. 2002. Movement pattern and home range use by the Eurasian lynx in Bialowieza Primeval Forest (Poland). Ann Zool Fenn, 39(1):29-41.         [ Links ]

39. Johnson, WE & WL Franklin. 1991. Feeding and spatial ecology of Felis geoffroyi in Southern Patagonia. J Mammal, 72(4):815-820.         [ Links ]

40. Karanth, KU & R Chellam. 2009. Carnivore conservation at the crossroads. Oryx, 43(1):1-2.         [ Links ]

41. Krauze-Gryz, D; J Gryz & J Goszczynski. 2012. Predation by domestic cats in rural areas of central poland: an assessment base don two methods. / Zool, 288(4):260-266.         [ Links ]

42. Levins, R. 1968. Evolution in changing environments. Princeton Univ. Press, Princeton, New Jersey.         [ Links ]

43. Lozano, J & F Urra. 2007. El gato domestico, Felis catus Linnaeus, 1758. Galemys, 19(l):35-38.         [ Links ]

44. Manfredi, C; M Lucherini; AD Canepuccia & EB Casanave. 2004. Geographical variation in the diet of geofroy's cat (Oncifelis geofroyi) in pampas grassland of Argentina./ Mammal, 85(6):1111-1115.         [ Links ]

45. Manfredi, C; L Soler; M Lucherini & EB Casanave. 2006. Home range and habitat use by Geofroy's cat (Oncifelis geofroyi) in a wet grassland in Argentina. / Zool, 268(4):381-387.         [ Links ]

46. Mino, MH; R Cavia; IE Gomez Villafane; DN Bilenca; EA Cittadino; et al. 2001. Estructura y diversidad de dos comunidades de pequeños roedores en agroecosistemas de la provincia de Buenos Aires, Argentina. Bol Soc Biol Concepción, 72:73-81.         [ Links ]

47. Mills, JN; AE Ellis; KT McKee; JI Maiztegui & JE Childs. 1991. Habitat associations and relative densities of rodent populations in cultivated areas of central Argentina. / Mamm, 72(3):470-479.         [ Links ]

48. Mukherjee, S; SP Goyal; AJT Johnsingh & MRP Leite Pitman. 2004. The importance of rodents in the diet of jungle cat (Felis chaus), caracal (Caracal caracal) and golden jackal (Canis aureus) in Sariska Tiger Reserve, Rajasthan, India. / Zool, 262(4):405-411.         [ Links ]

49. Naroski, T & D Yzurieta. 2012. Guía de Aves de Argentina y Uruguay. Editorial Vazquez Massini. Buenos Aires.         [ Links ]

50. Norman, GL & DL Streiner. 1996. Bioestadística. Time Mirror International Publishers. Division Iberoamericana. Madrid.         [ Links ]

51. Novaro, AJ; MC Funes & SR Walker. 2000. Ecological extinction of native prey of a carnivore assemblage in Argentine Patagonia. Biol Conserv, 92(l):25-33.         [ Links ]

52. Nowell, K & P Jackson. 1996. The Americas. Geofroy's cat, Oncifelis geofroyi (D'orbigny and Gervais, 1844). Pp. 129-130 in: Nowell, K & P Jackson (eds.). Wild cats Status survey and Conservation Action Plan. IUCN, Gland, Switzerland.         [ Links ]

53. Palacios, R; SR Walker & AJ Novaro. 2012. Diferences in diet and trophic interactions of Patagonian carnivores between areas with mostly native or exotic prey. Mamm Biol, 77(3):183-189.         [ Links ]

54. Paruelo, JM; JP Guerschman & SR Veron. 2005. Expansión agrícola y cambios en el uso del suelo. Ciencia Hoy, 15: 14-23.         [ Links ]

55. Pereira, JA; DM Varela & L Raffo. 2005. Relevamiento de los felinos silvestres en la región de parque nacional Pre-Delta. Facena, 21:69-77.         [ Links ]

56. Pereira, JA; NG Fracassi & MM Uhart. 2006. Numerical and spatial responses of Geofroy's cat (Oncifelis geofroyi) to prey decline in Argentina. J Mammal, 87(6):1132-1139.         [ Links ]

57. Pereira, JA. 2009. Efectos del manejo ganadero y disturbio asociados sobre la ecología trofica y la demografía del gato montés (Leopardus geofroyi) en el desierto del monte, Argentina. Dissertation, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires.         [ Links ]

58. Pereira, JA. 2010. Activity pattern of Geofroy's cats (Leopardus geofroyi) during a period of food shortage. J Arid Environ, 74(9):1106-1109.         [ Links ]

59. Pereira, JA; MS Di Bitetti; NG Fracassi; A Paviolo; CD De Angelo; et al. 2011. Population density of Geofroy's cat in scrublands of central Argentina.J Zool, 283(l):37-44.         [ Links ]

60. Pereira, JA; RS Walker & AJ Novaro. 2012. Efects of livestock on the feeding and spatial ecology of Geofroy's cat. J Arid Environ, 76(l):36-42.         [ Links ]

61. Pia, MV; MS Lopez & AJ Novaro. 2003. Efects of livestock on the feeding ecology of endemic culpeo foxes (Pseudalopex culpaeus smithersi) in central Argentina. Rev Chil Hist Nat, 76(2):313-321.         [ Links ]

62. Radovic, A& D Kovacic. 2010. Diet composition of the golden jackal (Canis aureus L.) on the Pejesäc Peninsula, Dalmatia, Croatia. Period Biol, 112(2):219-224.         [ Links ]

63. Rajaratnam, R; M Sunquist; L Rajaratnam & L Ambu. 2007. Diet and habitat selection of the leopard cat (Prionailurus bengalensis borneoensis) in an agricultural landscape in Sabah, Malaysian Borneo. J Trop Ecol, 23(2):209-217.         [ Links ]

64. R Development Core Team. 2010. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0. URL: http://www.R-project.org.         [ Links ]

65. Santillán, MA; AS Costán & JI Zanón-Martinez. 2014. Predation by Geoffroy's cat on ferruginous pygmy-olw in Caldén semiarid forest, Argentina. J Raptor Res, 48(2):192- 193.         [ Links ]

66. Santos Moreno, A & AM Alfaro Espinola. 2009. Mammalian prey of barn owl (Tyto alba) in southeastern Oaxaca, México. Acta Zool Mex, 25(1):143-149.         [ Links ]

67. Schipper, J; JS Chanson; F Chiozza; JN1A Cox; M Hoffmann; et al. 2008. The status of the world's land and marine mammals: diversity, threat and knowledge. Science, 322(5899):225-230.         [ Links ]

68. Sikes, RS; WL Gannon & The Animal Care and Use Committee of the American Society of Mammalocists. 2011. Guidelines of the American Society of Mammalogists for the use of wild mammals in research. Journal of Mammalogy, 92(1): 235-253.         [ Links ]

69. Soler, L; M Lucherini; C Manfredi; M Ciuccio & EB Casanave. 2009. Characteristics of defecation sites of the Geoffroy's cat Leopardus geoffroyi. Mastozool Neotrop, 16(2):485-489        [ Links ]

70. Soriano, A; KJC León; OB Sala; JxS Lavado; VA Oeregibus; et al. 1991. Río de la Plata grasslands. Pp 367-407 in: Coupland, RT (ed.). Natural grasslands Introduction and Western Hemisphere. Elsevier, Ecosystems of the World, Amsterdam, Netherlands.         [ Links ]

71. Sousa, K & A Bager. 2008. Feeding habits of Geoffroy's cat (Leopardus geoffroyi) in southern Brazil. Mamm Biol, 73(4): 303-308.         [ Links ]

72. Sovada, MA; CC Roy & DJ Telesco. 2001. Seasonal Food Habits of Swift Fox (Vulpes velox) in Cropland and Rangeland Landscapes in Western Kansas. Am Midi Nat, 145(1):101-111.         [ Links ]

73. Treves, A & KU Karanth. 2003. Human-carnivore conflict and perspectives on carnivore management worldwide. Conservation Biology, 17(6):1491-1499.         [ Links ]

74. Trigo, TC; F Tirelli; L Machado; F Bortolotto Peters; C Indrusiak; et al. 2013. Geographic distribuition and food habits of Leopardus tigrinus and L. geoffroyi (Carnivora, Falidae) at their geographic contact zone in southern Brazil. Stud Neotrop Fauna E 48(1):56-67.         [ Links ]

75. Varela, O; A Cormenzana-Méndez; L Krapovickas & EH Bucher. 2008. Seasonal Diet of the Pampas Fox (Lycalopex gymnocercus) in the Chaco Dry Woodland, Northwestern Argentina. / Mammal, 89(4):1012-1019.         [ Links ]

76. Viglizzo, EF; F Lertora; AJ 1 ordomingo; JJN1 Bernardos; R Ze; et al. 2001. Ecological lessons and applications from one century of low external-input farming in the Pampas of Argentina. Agric Ecosyst Environ, 83(1-2):65-81.         [ Links ]

77. Vuillermoz, PA. 2001. Dieta estacional y selección de presas del gato montés (Oncifelis geofroyi) y zorro pampeano (Pseudalopex gymnocercus) en la Reserva de Vida Silvestre "Campos del Tuyu" (Bahía Samborombón). Dissertation. Universidad de Buenos Aires, Argentina.         [ Links ]