Estimation of tiger densities in India using photographic captures and recaptures

Ecology, Dec, 1998 by K. Ullas Karanth, James D. Nichols

Alluvial grasslands such as Kaziranga provide optimal habitats where tigers probably attain their highest ecological densities. Although Chitwan also is an alluvial grassland, tiger densities appear to be lower there than in Kaziranga, Nagarahole, and Kanha. Recent extirpation of large-sized prey such as wild buffalo and barasingha from Chitwan (Sunquist 1981) may be a contributory factor. Our study site in Nagarahole included both moist deciduous forests, with high prey densities, and dry deciduous forests, with lower prey densities (Karanth and Sunquist 1992). If tiger densities (excluding cubs) in moist deciduous forests of Nagarahole are considered separately, they (14.7 tigers/100 [km.sup.2]; Karanth 1995) seem comparable to the densities in Kaziranga ([Mathematical Expression Omitted] = 16.8 [ or -] 2.96 tigers/100 [km.sup.2]). Therefore, it appears as though moist deciduous forests and alluvial grasslands of tropical Asia represent the upper limits for ecological densities attained by wild tiger populations, whereas the temperate forests of the Russian Far East may represent the lower limit.

At Kanha, the tiger density estimated by us is considerably higher than the density reported by Schaller (1967) 30 years earlier. Conservation measures implemented during the intervening period, such as effective antipoaching operations, relocation of human settlements, and elimination of forest product exploitation (Panwar 1987), appear to have substantially increased the capacity of the Kanha area to support tiger populations. A similar rebound of tiger populations following comparable conservation interventions was also observed in Nagarahole (Karanth et al., 1998).

Wikramanayake et al. (1998) have synthesized remotely sensed vegetation maps with rapid assessment of tiger distributions to estimate the size of potential habitat blocks called "Tiger Conservation Units" (TCU). The sizes of TCUs containing the specific habitat types in which we sampled tiger densities range from 3600 to 23 900 [km.sup.2]. Although these forest blocks are largely composed of suboptimal tiger habitats under multiple uses, they contain better quality habitat patches in wildlife protected areas ranging in size from 488 [km.sup.2] to 7013 [km.sup.2] (Wikramanayake et al., 1998). Assuming a typical, reasonably protected, but insular tiger habitat patch size of 1000 [km.sup.2], and using the lower and upper limits of tiger densities estimated by us in this study (Table 4), we can estimate the sizes of typical surviving wild tiger populations in such better protected habitat blocks at 41-168 tigers, exclusive of cubs.

Influence of prey community abundance and structure on densities of large felids

Based on our data (Table 5), Kaziranga had the highest densities of large prey (16.9 animals/[km.sup.2]), followed by Pench (11.0 animals/[km.sup.2]), Nagarahole (8.7 animals/[km.sup.2]), and Kanha (4.5 animals/[km.sup.2]). The densities of medium-sized prey at these sites were, respectively, 41.2, 52.8, 41.4, and 52.2 animals/[km.sup.2]. According to the predictions of Karanth and Sunquist (1995), Nagarahole, Pench, and Kaziranga, which support abundant prey in both size classes, should have leopards coexisting with tigers at relatively high densities. In Kanha, because of tigers switching to medium-sized prey, leopards should be relatively scarce. Our relative abundance index for leopards (Results) shows that this pattern holds true except for Kaziranga, which has a low index of leopard abundance. This exception to the predicted pattern could arise because ecological factors other than prey community structure may also influence leopard abundance in Kaziranga. Seasonal flooding or scarcity of trees that provide escape cover during interspecific interactions with tigers in the grasslands (K. U. Karanth, personal observations) could be among the factors constraining coexistence of leopards with tigers in Kaziranga.