2. METHODS

2.1 The study area

The study area, Chinnar Wildlife Sanctuary, is located between 10°15’- 10°21’N latitude and 77°16’- 77°5’ E longitudes in the Idukki district of Kerala state (Figure 1) and has a total extent of 90.44 km2. The significant variation in altitude and rainfall leads to a wide array of habitat types like deciduous forests, dry thorny scrub, riparian forests, shola forests and grasslands that are interspersed with plains, hillocks, rocks and cliffs, which provide microhabitats for varied forms of life (Management Plan of Chinnar Wildlife Sanctuary., 2012).

2.2 Observation methods

The information on the composition and seasonal variation in the GGS diet was collected through foraging observations. Focal Animal Sampling was used following (Altmann, 1974). Each encountered individual was followed, and observation on the time spent on feeding, plant species eaten, plant part eaten, pick up rates of plant parts such as fruit, leaves and flowers were also recorded. The activities occur in bouts that are periods of, for example, feeding activity within a food source or movement between two sources. During the period spent in the food source, some of the time was allotted to searching for a food item, selecting by smelling, remaining inactive or other grooming activities. The length of the bout was judged as the period between entry into and exit from the source. The incidence of feeding on different food items across different seasons was categorised. This was done for the three different seasons summer (December to May), southwest monsoon (June to September) and northeast monsoon (October to November).
Observations were made for 10 months between April 2013 to May 2014, and within a month, at least one week was spent in the field. Five transects were selected after reconnaissance, and observations were made for each transect both in the forenoon and afternoon for an equal duration. Squirrels were observed using 8x40 Olympus standard binoculars (Olympus Global, Tokyo, Japan). Different feeding postures and feeding techniques used by the GGS were interpreted by direct observation during the sampling.

2.3 Data analysis

To understand the food composition of GGS, collected data were analysed by three complementary approaches. The percentage contribution of different food items to the squirrel diet was analysed based on the duration of feeding on a particular item and the number of times of feeding incidence on a particular item. The non-parametric test, “Mann Whitney U” was performed to determine whether there was any significant difference in the feeding bout and duration between the forenoon and afternoon hours. ᵡ2 test for association was used to find out the seasonal variation on the preferred tree species and the plant parts fed by the GGS across different seasons. A constrained analysis, species as response variable against the environmental variables (Residualized predictors were permuted), canonical correspondence analysis (CCA) was performed to check the food choice and season using Canoco 5 (Braak & Smilauer, 2012; Jiangshan et al., 2014).