Results cont...

Track Census

Tracks were recorded for a total of 15 species although there was only sufficient data available to adequately graph 7 of them; 2 cats (L. pardalis, P. onca), 3 ungulates (T. tajacu, M. americana, T. terrestris), and 2 rodents (D. variegata, A. paca). These species yielded 330 separate track encounters in designated track traps. Paired t-tests revealed that the abundance of jaguar (P. onca) tracks were greater along T transects than C transects. The results for this species (Figure 3.18) give the impression that it is restricted almost entirely to tourist trails! This however is not the case, the results are probably an artefact of the methodology and the behaviour of this species. It was noted that jaguars did walk along C but much less frequently and for shorter distances than along T, thus significantly reducing the chances of them being recorded. Furthermore, it has been known for some time that this and other cat species show a significant propensity to walk along man-made tails for long distances (Emmons 1984, 1990). The methodology is therefore significantly biased towards over-recording of species which prefer to walk along open clear trails for long distances. The trend in track abundance between T and C for 3 species were significantly correlated: T. tajacu,(Spearman = +0.87), L. pardalis (Spearman = +1.00) and A. paca (Spearman = +1.00). Further Spearman correlation analyses (Table 3.16) showed that all species except T. tajacu were negatively correlated with traffic along tourist transects, however only M. americana, T. terrestris and L. pardalis were significantly affected it appears. No correlation was evident between track abundance and relative abundance, as calculated from observational data, for any species.

Table 3.16. Results from a Spearman correlation analysis between traffic and track abundance along T. * p < 0.05

Figure 3.18. Relative abundance of Jaguar (P. onca) tracks.

Figure 3.19. Relative abundance of Ocelot (L. pardalis) tracks.

Figure 3.20. Relative abundance of Collared Peccary (T. tajacu) tracks.

Figure 3.21. Relative abundance of Brazilian Tapir (T. terrestris) tracks.

Figure 3.22. Relative abundance of Red Brocket Deer (M. americana) tracks.

Figure 3.23. Relative abundance of Brown Agouti (D. variegata) tracks.

Figure 3.24. Relative abundance of Paca (A. paca) tracks.

Visibility Levels

There is the potential for the abundance of a species, when determined with distance sampling methodology, to vary from place to place simply due to variations in visibility levels brought about by variations in vegetation characteristics. To investigate this we chose to analyse visibility in terms of the average straight-line observer-animal distance (AD) for the saddleback tamarin (S. fuscicollis) and the brown agouti (D. variegata), two commonly observed mammals. Each is either entirely arboreal or terrestrial thus allowing for an analysis visibility at the level of the low to mid canopy as well as the undergrowth. The results (Tables 3.17 and 3.18 ) and t-tests showed that only in the case of the brown agouti at EI was there a difference in visibility between T and C. When data was grouped no significant differences between lodges as a whole was evident.

Table 3.17. Straight-line observer-animal distance (m) for S. fuscicollis.

Table 3.18. Straight-line observer-animal distance (m) for D. variegata. * p < 0.01.

Discussion