**2. Methods**

#### **2.1 Study area**

Mudumalai Tiger Reserve is located in the Nilgiris District of Tamil Nadu (11° 32′ and 11° 42´ N and 76° 20′ and 76° 45′ E; **Figure 1**). It extends over an area of 321 km<sup>2</sup> and forms a part of the Nilgiri Biosphere Reserve. The sanctuary is located in the Western Ghats, which is one of the 34 Biodiversity hotspots of the world [14]. The altitude varies from 485 to 1226 m above MSL with a general elevation of about 900 to 1000 m. The annual rainfall is varied from 1001 mm to 1648 mm. Sanctuary receives rain both from southwest (May to August) and northeast (September to December)

#### **Figure 1.** *Map showing Mudumalai tiger reserve and transect locations.*

#### *Population Density and Age-Sex Composition of Large Mammals in Tropical Forests, Southern… DOI: http://dx.doi.org/10.5772/intechopen.109567*

monsoons. The study area has three major forest type namely tropical moist deciduous forest, dry deciduous forest and tropical thorn forest [15]. The dominant tree species associations are "*Acacia*-*Ziziphus*" association characterising tropical thorn forest, "*Anogeissus* – *Tectona*" and "*Anogessius* – *Terminalia* – *Tectona*" association describing the tropical dry deciduous forest and "*Lagerstoroemia* – *Terminalia* – *Tectona*" characterising the tropical moist deciduous forest. The presence of areas of open grassland, numerous permanent water resources and salt licks has resulted in highly productive landscape. The large herbivores include elephant (*Elephus maximus*), three species of cervids, Chital (*Axis axis*), Sambar Deer (*Rusa unicolor*) and Barking Deer (*Muntiacus muntjak*), two species of antelopes, *viz*., the Four-horned Antelope (*Tetracerus quadricornis*) and the Blackbuck (*Antilope cervicapra*) are the dominant mammals of this area. In addition predators like Tiger (*Panthera tigiris*), Leopard (*Panthera pardus*) and Wild dog (*Cuon alpinus*) can be seen. The Gaur and other herbivores are threatened by habitat degradation from overgrazing and human disturbances.

Densities of herbivores were estimated using distance sampling [16–18]. A total of 24 transects, each 2–4 km in length were monitored within a 321 km2 survey area encompassing the different habitats from 2004 to 2007. Transects were placed randomly, with stratification by broad forest type (Moist deciduous forest, Dry deciduous forest and Thorn forest; **Figure 1**). Twelve surveys were repeatedly conducted on 24 line transects established throughout the sanctuary from 2004 to 2007 during the dry season in December to March and this provided adequate spatial replicates of the study area (643 km total), covering all vegetation types. On every walk, species, group size, age-sex composition, sighting angle using hand held sighting compass and sighting distance measured by a manual range finder. Population density of large herbivores was estimated using the software Distance v.6 Release 2 [19]. To model detection functions to estimate species density the data for each species per transect was examined for signs of evasive movement and peaking at distances from the line of walk. Following this, the data were either truncated or re-classed so as to ensure a reliable fit of key functions and adjustment terms to the data. The Akaike Information Criterion (AIC) and χ<sup>2</sup> goodness-of-fit test were used to judge the fit of the model. Using the model thus selected, estimates of group density, group size and individual density were derived [19].

In addition to transects, road-strip counts were conducted on a monthly basis in the different vegetation types (1262.5 km total). A total distance of 830, 308 and 124.5 km was sampled in DDF, MDF and TF respectively. In each sighting of gaur, we recorded group size and composition. Only groups where all the individuals counted were used for group size analysis. The sighting data from transect and road-strip count were pooled together to compute season-wise and habitat wise estimate on mean group size and sex ratios and median group size. Mean group sizes obtained from line transect counts actually indicate the size of 'clusters' [16] of animal species being sampled, rather than of social groups. Though these clusters are not always identical to social groups it has been used for population analysis [7]. In addition group size data from transects, counts from road-side counts, foot trails and waterholes were also used to calculate mean group size and group structure. For comparing grouping tendencies of different species the percentage of observed groups in categories such as: solitary animals, family associations (2–3 individuals) consisting of pairs of adult females with their yearlings and young, small groups (4–10 individuals), medium sized groups (11–30 individuals) and large groups of above 30 individuals.

#### **2.2 Population structure**

The proportion of different age-sex classes in the population was calculated using transects count supplemented with counts from road-side counts, foot trails and waterholes. The physical characteristics described in the literature [20–22] and observation in captive animals to establish the age-sex categories for classification. Chital, Sambar and gaur were classified as adult male (AM), adult female (AF), Sub-adult male (SAM), Sub-adult female (SAF) and young (YG). Younger elephants (<15 years) were classified by comparing their height to the oldest adult female in the group. Elephants were placed in broad age groups such as calves (<1 year), juveniles (1–5 years), sub-adults (5–15 years) and adults (>15 years) [7, 10].

The mean biomass densities (in kg/km<sup>2</sup> ) in the study area were derived from multiplying mean ecological density for each species by its average unit weight. Biomass density was estimated from published data on body weights [20, 23, 24] and population structure data from study area.

### **3. Results**

Barking deer and sambar are essentially non-social species with solitary individuals, pairs or family association comprising 80% and 84% groups seen, respectively (**Table 1**). Though solitary individuals accounted for 16% and 25% of elephant and gaur most of these individuals are appeared to be adult males. These two species seems to occupy an intermediate position in terms of sociality with the group size frequencies being relatively evenly distributed among family associations, small groups and medium sized groups. At the other end of the scale, chital and common langur were clearly social groups with solitary individuals comprising less than 5% of all animal classified. The exceptionally large groups of 148 for chital and 32 for gaur were probably temporary agglomerations of several groups.

The study area harboured high ungulate density of 55.4 ± 5.83 animals/km2 . Among the different species chital was the most abundant species with the highest density (29.8/km2 ; **Table 2**). The next abundant species was Common langur 11.9 ± 2.31/km2 . Gaur and sambar were the abundant species and their densities were 7.1 ± 1.47/km2 and 4.2 ± 0.62/km2 respectively. Barking deer density (0.6 ± 0.20/km2 ) was relatively lower than other large herbivores. Among different species chital had the highest group density (3.6 ± 0.55/km2 ). Species such as Gaur and sambar had higher density but with lower group density (1.3 ± 0.18/km2 and 1.7 ± 0.22/km2 ). The ecological density varied with group size of different herbivore species. Though there is relatively less variation in the group density among Gaur, sambar and elephant the density estimate varied because of variation in the group size. Among the different species chital and Gaur had higher group size of 8.3 ± 0.85 individuals and 5.59 ± 0.84 individuals respectively. The solitary species such as barking deer, giant squirrel and black naped hare density did not show any variation between the group density and actual density. Overall elephant density was 4.8 ± 0.92/km2 with mean group size of 4.3 ± 0.40 individuals (**Table 2**).

#### **3.1 Population structure**

The adult sex ratios in ungulate species seem to be female biased. The adult male– female sex ratios favour females by 1:33.8 for elephant, 1:4 for gaur and 1:2 for chital in exception equal (1,1) sex ratio was observed in sambar. The composition of adult male *Population Density and Age-Sex Composition of Large Mammals in Tropical Forests, Southern… DOI: http://dx.doi.org/10.5772/intechopen.109567*


#### **Table 1.**

*Grouping characteristics among large herbivore species: Group size classes, range of observed group size, n = sample size in Mudumalai Tiger Reserve.*


#### **Table 2.**

*Ecological density (±standard error), group density (±SE) and group size (±SE) of large mammals in Mudumalai Tiger Reserve during the study period.*

**Figure 2.**

*Percentage of different age-sex classes among four ungulate species (elephant and gaur; sambar and chital) in Mudumalai Tiger Reserve.*

was lowest (1.65%) among the elephants (**Figure 2a**). Totally 120 gaurs were sighted during the study period. The composition of adult female (50.79%) was highest in the group. The sub-adult male and female consisted of 16% of the populations. Totally 78 sambar deer were sighted and in which 85% consisted of Adult male and adult female (**Figure 2b**). The calf consisted of 10.34%, the sub-adult male had the lowest composition of 0.49%. While considering age distribution, the proportion of prereproductive age classes (sub-adults, juveniles and calves) is 53% in elephant, 36.6% in gaur, 26.4% in chital and 14% in sambar.
