The diel activity pattern of small carnivores of Western Ghats, India: a case study at Nelliampathies in Kerala, India

: The diel activity pattern of small carnivores was studied using the camera trap technique at Nelliampathy Reserve Forest, Kerala, India. Six species of small carnivores were recorded during the study. These include Brown Palm Civet Paradoxurus jerdoni, Small Indian Civet Viverricula indica , Stripe-necked Mongoose Herpestes vitticollis , Brown Mongoose Herpestes fuscus, Nilgiri Marten Martes gwatkinsii , and Leopard Cat Prionailurus bengalensis . The maximum diel activity overlap was detected between the Brown Palm Civet and Small Indian Civet, while the activity overlap was minimal between the Stripe-necked Mongoose and Small Indian Civet.


INTRODUCTION
Small carnivores are medium-sized mammals belonging to the order Carnivora. There are 195 species of small carnivores globally belonging to 10 families (Wilson & Mittermiere 2009). The Western Ghats has 14 species, out of which 13 are present in Kerala (Nameer 2015(Nameer , 2020. Understanding the geographical and ecological distributions and abundance of each species is the foundation for effective management. A study on the impact of various factors on habitat selection of Smooth-coated Otters Lutrogale perspicillata in Periyar Tiger Reserve (PTR) found that the otters showed affinity towards areas with less rocky and gently sloped banks with vegetation and adjoining streams (Anoop & Hussain 2004). The central part of the L. perspicillata diet in PTR was fishes (96%), among which tilapia was the primary food item during both lower and higher water levels in the lake (Anoop & Hussain 2005). While comparing the abundance of small carnivores between an intact rainforest and adjoining forest fragments, it was observed that the intact forests have a higher abundance of small carnivores than the fragmented landscapes (Mudappa et al. 2007 (Kalle et al. 2013a). The niche of Brown Mongoose H. fuscus fuscus was greatly influenced by temperature, rain and topography (Raman et al. 2020).
Brown Palm Civet P. jerdoni was believed to have distribution ranges from Kalakkad Mundanthurai Tiger Reserve, Tamil Nadu, to Bhagvan Mahaveer Wildlife Sanctuary in Goa. However, the distribution of Brown Palm Civet has extended further north of Goa up to Satara district of Maharashtra (Bhosale et al. 2013;Sayyed et al. 2019). Punjabi et al. (2014) have reported the northern extension of Stripe-necked Mongoose distribution from Maharashtra and Goa. Recent records of Brown Mongoose from Bilgiri Rangaswamy Temple Tiger Reserve points towards the southeast extension of the distribution of the species from its known range (Suthar et al. 2020).
Studies on the species richness of small carnivores from the Western Ghats reported varying species from the different regions. Kumara & Singh (2006) and Kumara et al. (2014) reported 11 species of small carnivores from the forests of Karnataka. Parmbikulam Tiger Reserve reported 11 species, with Small Indian Civet and Common Palm Civet as the common ones (Sreehari & Nameer 2016). The drier tracts of Wayanad Wildlife Sanctuary reported nine species, and similar to Parambikulam Tiger Reserve, Small Indian Civet was the most frequently sighted species at Wayanad WS. (Sreekumar & Nameer 2018). The high-altitude landscape in Eravikulam National Park recorded nine species, and Jungle Cat and Leopard Cat were the common small carnivores (Nikhil & Nameer 2017). The rain forest landscape of Silent Valley National Park recorded only seven species. The Small Indian Civet was the most common small carnivore in the rainforest habitat (Sanghamithra & Nameer 2018). Anil et al. (2018) reported on the social behaviour, feeding habits, and activity pattern of Martes gwatkinsii from the Pampadum Shola NP.
Diel activity pattern is one of the critical factors which determines the ecological niche of a species. It is also an essential tool for the co-living of the species (Gerber et al. 2012). Interspecific competition is reduced by the chronological separation between the species (Selvan et al. 2019). Variation in activity peaks was observed among sympatric species with similar activity (Su & Sale 2007;Chen et al. 2009).
All the three species of the civets recorded from Wayanad Wildlife Sanctuary in the Western Ghats were nocturnal with varying temporal activities, while the mongooses were diurnal (Sreekumar & Nameer 2018). In Sumatra, a study on the activity pattern of the small carnivores found that all the six species of viverrids in the study area were nocturnal with temporal variations in the activity peaks in an oil palm plantation. At the same time, the Yellow-throated Marten Martes flavigula was diurnal (Solina et al. 2018), the Nilgiri Marten of Pampadum Shola NP in the Western Ghats was also diurnal in habit (Anil et al. 2018). The nocturnal nature of the Small Indian Civets was proved in other studies from the Western Ghats, too (Pillay 2009;Chen et al. 2019;Kalle et al. 2013b). However, Selvan et al. (2019), in a study in the Villupuram district of Tamil Nadu, observed that Small Indian Civets were active during daytime hours.
The present study is expected to gather additional information on the diel activity pattern of the small carnivores of Western Ghats.

Study Area
The study was conducted at Nelliyampathy Reserve Forests (NRF), in the Anamalai Hills, southern Western Ghats, India (Figure 1). The Nelliampathy reserve forest lies between 10.374-10.686 0 N latitudes and 76.518-76.752 0 E longitudes in the Palakkad district, Kerala, and has an extent of 206 km 2 . The altitude varies from 40 m to 1,530 m, and the primary vegetation type is west coast tropical evergreen forest. The dominant trees are Cinnamomum malabatrum, Drypetes roxburghii, Holigarna arnottiana, Mesua ferrea, Palaquim ellipticum, Schleichera oleosa, Syzygium cumini, and Vateria indica. The average temperature ranges 21-41 o C during summer, and the temperature can be as low as 10 o C during the winter in the upper reaches of the Nelliampathies. The mean annual rainfall is 2,500mm (Varghese 2015).

Methods
A total of 30 camera trapping stations were selected in the NRF based on indirect evidence such as scats, pugmarks, and scratches of the small carnivores. We de-ployed camera traps (Cuddeback attack model C1: digital scout cameras with passive infra-red sensors for heat and motion detection) at these locations during January 2019 at the height of 30 cm from the ground, and two cameras were placed at least 250 m from each other (Mudappa et al. 2007;Sreehari & Nameer 2016;Nikhil & Nameer 2017;Sanghamithra & Nameer 2018;Sreekumar & Nameer 2018). The cameras were set up in default settings. The time delay between the pictures during the day was set as fast as possible, and during the night, it was set with a time delay of five seconds. Garmin GPS etrex 30 was used to mark the camera trap stations. The cameras were kept open for 24 hours a day for 28 days at each location. Thus, 840 camera trap days, monitoring for 20,160 hours of trap effort, were carried out in NRF during the study period.

Camera trap success rate
The camera trap success rate is the ratio of independent photo events to the whole camera trap days and the value multiplied by 100 (Rovero & Marshall 2009). The number of independent images of small carnivore camera trapped from NRF was used to calculate the camera trap success rate.

Line transect survey for indirect evidence
The transects were done on the existing trails, forest roads, and streams, searching for indirect evidence of small carnivores. A total of 104km was walked through the various trails in search of indirect evidences of small carnivores. The scats were identified to the family level of small carnivores or the species level (Silveria et al. 2003;Sridhar et al. 2008;Mudappa et al. 2010;Perincherry et al. 2011).

Analysis of diel activity
We recorded species of small carnivores, date, time, and geocoordinates of the trap location for every camera trap image captured. To ensure the independence of the capture, we have defined successive images of the same species at the same camera trap station within a recess of ≤30 minutes as a single event ( The diel activity pattern and activity overlapping were determined using a non-parametric circular Kernal density method. Soothing parameter of 0.8 (sample size <50) was used to generate coefficient of overlap (Δ) (Ridoout & Linkie 2009). The range of coefficient of activity overlap varies from 0 (zero overlaps) to 1 (100% overlap) (Ridoout & Linkie 2009). R-package' OVERLAP' was used to analyze activity patterns of single species and coefficient of overlapping between two species (Meredith & Ridout 2018). To obtain a bias-corrected percentile, we estimated the 95 % confidence interval of Δ with 1,000 bootstrap (Meredith & Ridout 2018).

Diversity of small carnivores at Nelliampathy Reserve Forests, Western Ghats
We recorded six species of small carnivores from NRF representing four families. This comprises two species each of herpestids and viverrids and one species each of felids and mustelids (Table 1). A total of 677 images of 24 species of mammals were obtained during the study period. Two-hundred-and-thirty-one images were of carnivores, out of which 199 (86.15 %) were of small carnivores ( Figure 2 (Figure 3).
The small carnivore camera trap success rate from the evergreen forests of NRF was 22.14 per 100 trap nights. The camera trap success rate of NRF is much higher than earlier camera trap studies from various locations in the Western Ghats. For example, the camera trap success rate of Silent Valley National Park was 10.90 per 100 camera trap nights (Sanghamithra & Nameer 2018), Parambikulam Tiger Reserve was 4.40 (Sreehari & Nameer 2016),   (Mudappa et al. 2007).

Diel activity of small carnivores at Nelliampathy Reserve Forests, Western Ghats
The maximum diel activity overlap was detected between Brown Palm Civet and Small Indian Civet with Δ of 0.81 (0.66-0.92) (Fig. 4a) The most significant diel activity overlap was between Brown Palm Civet and Small Indian Civet. Even though they are similar in size and activity, the competition for resources may be minimized by the dissimilarity in their dietary preferences. Brown Palm Civet is primarily frugivorous (Rajamani et al. 2002;Mudappa et al. 2010), whereas Small Indian Civet is a generalist and omnivorous (Mudappa et al. 2007).
Brown Palm Civet, Small Indian Civet, and Brown Mongoose displayed nocturnal activity patterns, and they have the most significant overlap in the diel activity. However, they all showed varying activity peaks, probably to reduce the competition. Activity peaks of Brown Palm Civet were just before dawn (0400-0600 h) and just after dusk (1800-2000 h), whereas Small Indian Civet had activity peaks were during midnight hours (0000-0100 h) and soon after sunset (1900-2030 h). Brown Mongoose showed peak activity during the midnight hours from 2300 to 0100 h. Similar activity patterns and overlap were observed among the Small Indian Civet and Brown Palm Civet in Wayanad Wildlife Sanctuary (Sreekumar & Nameer 2018). The diel overlap between Common Palm Civet and Small Indian Civet both showed nocturnal activity but varying activity peaks (Su & Sale 2007).
The activity pattern not only depends on factors like limited resources and competition but also on seasonal changes (Ikeda et al. 2016), changes in diurnal temperatures (Fuller et al. 2016) prey-predator interactions (Harmse et al. 2011Linkie & Ridout 2011) and human interventions and human activity (Cruz et al. 2018). It needs to be further investigated to understand how the sympatric species with overlapping diel activity perform the resource partitioning.

CONCLUSION
Depending on the time of the activity of a species, the small carnivores are generally grouped into two, nocturnal and diurnal. Species within the same temporal group have a more significant overlap in their activity. Maximum overlap was observed between two nocturnal small carnivores, Brown Palm Civet and Small Indian Civet. At the same time, the lowest overlap in activity was observed between Stripe-necked Mongoose and Small Indian Civet.
Diel activity patterns are a vital feature of animal behaviour with important implications for a wide range of ecological and physiological processes. Diel activity patterns are an adaptation to environmental variability throughout the day. They reflect a complex compromise  between foraging, resting, predator avoidance, competition, social activities and environmental constraints determining fitness. Thus, the diel activity studies may enable us to perform more robust comparisons of activity patterns and levels across sites and species to better understand ecological and human drivers of these processes and thus have management and conservation significance.