Gaius Wilson

We examined changes in a community of seedlings/saplings 10–150 cm tall associated with the presence of a widely invasive plant, Lantana camara and environmental covariates along 67 randomly located transects, in Mudumalai, India. We compared plant species assemblage and grass cover in L. camara-invaded and uninvaded plots in three habitats. Multivariate analyses revealed a significant association of all environmental covariates with plant species assemblage. Pairwise tests indicated that L. camara was significantly associated with changes in plant species assemblage and grass cover within the moist and dry deciduous forest, but not in the thorn forest. The relationship between L. camara and that of elephant browse plants varied with species. A linear regression analysis indicated that L. camara invasion was the only significant predictor of grass occupancy. Our results indicate that in addition to other factors, L. camara was associated with altering plant species assemblage, some elephant browse plants and grass cover in the moist and dry deciduous forest. It appears that L. camara can have a major effect on diversity within this reserve, but whether this effect is by L. camara driving the change or being associated with other habitat change requires further experimental evidence.

Lantana camara is a widespread exotic invasive species in India, capable of dominating and displacing native forage species. We investigated whether L. camara was associated with variation in elephant foraging behavior in Mudumalai Tiger Reserve, India. The behavioral responses of elephants to L. camara were assessed from feeding and stepping rates. Elephants were never observed to feed on L. camara, but rather fed on grass and browse present within and around L. camara patches. A multiple regression analysis showed that feeding rates were negatively associated with L. camara invasion (F_{1, 55} = 4.26, R² = 0.07), but not stepping rates. Instead, grass cover and browse density were associated with stepping rates (F_{2, 55} = 11.16, R² = 0.30). Path analysis indicated that the total effect of L. camara on feeding rates was 11 percent (β = −0.24) less than the direct negative association (β = −0.27) owing to a positive indirect association of L. camara with feeding rates through grass cover and browse density (β = 0.03), while stepping rates were negatively associated with grass cover (β = −0.39) and positively associated with browse density (β = 0.38). Our results indicate that L. camara appears capable of modifying feeding rates of elephants, likely through a loss of grass areas due to L. camara invasion. Experimental work is needed to test for causal relationships among the variables we measured, to enhance our understanding of how invasive weeds modify elephant behavior.

Invasive weeds likeLantana camarahave a range of effects on animals such as elephant. These plants are not edible by the Asian elephant (Elephas maximus). They also compete for space with elephant food plants and take over large areas of elephant habitat. We tested whether the addition ofL. camarato a model consisting of measured environmental variables improved predictions of habitat use by elephant in Mudumalai Tiger Reserve, India. Elephant dung density was used to assess elephant habitat use from 62 line transects 1-km in length. Results indicated that habitat and impact of human settlements significantly influenced elephant habitat use at a landscape scale. However, we found no evidence for the hypothesis that the addition ofL. camarasignificantly predicted elephant habitat use at the landscape level. We then tested the association ofL. camaraon elephant habitat use in the dry deciduous forest (DDF) where there was a significant interaction between DDF andL. camara. In the DDF,L. camarasignificantly predicted elephant habitat use. We conclude that while no significant effects ofL.camarawere seen at the level of an entire reserve, at a finer scale and in specific habitats negative effects of this invasive plant on elephant habitat use were observed.