The ecology of jaguars (panthera onca) in a human-influenced landscape

Despite intense persecution over the last century, the jaguar (Panthera onca) has sustained a wide geographic distribution, perhaps due to its elusive nature and rather flexible ecology. This study investigated jaguar ecology under anthropogenic pressures in Belize, Central America. A suite of methods including camera-trap surveys, diet analysis, discussions with local stakeholders, and population simulations were used to study a population of jaguars spanning the boundary of a protected forest. Camera-trap data combined with capture-recapture population models are increasingly used to estimate the density of mammals such as jaguars with individually identifiable coat patterns. A review of current methods highlighted problems associated with estimating the sizes of lowdensity populations. Simulations to assess the robustness of the method found that camera failure can negatively or positively bias the abundance estimate, depending on the particular nature of capture histories. The most commonly used model estimator in the literature was nevertheless robust to failures of up to 10% of trap-occasions. Pooling trap-occasions reduced the effect of camera failure. Sub-sampling data from large-scale surveys indicated a threshold survey area of ~170 km2, below which estimates of density were inflated and unreliable. For surveys exceeding this threshold size, jaguar density varied across the landscape from the protected forest to the human-influenced lands such that <30% contiguous forest precipitated reduction. Reduced densities with distance from contiguous forest and proximity to human habitation may result principally from direct conflicts with people. The influence of anthropogenic factors on the coexistence of jaguars and pumas (Puma concolor) was investigated by comparing their habitat use and feeding ecology. Diet was analysed from the largest sample to date of scats from one area identified to species. Jaguars and pumas made similar use of the secondary rainforest, despite differences in diet. Although both cats relied heavily on one species of small prey (5-10 kg), for jaguars this was the nine-banded armadillo (Dasypus novemcinctus) while for pumas it was the paca (Agouti paca). Both cats took some larger prey, mainly white-lipped peccaries (Dictolyes pecari) by jaguars and red brocket deer (Mazama americana) by pumas. Energetics models indicated that reproduction may be limited for either species if large prey are unavailable for females with dependents. Outside the forest block, jaguars rarely ate large wild prey species; instead, a diet of smaller wild prey was supplemented with large domestic stock. Pumas were scarce outside the protected forest, possibly reflecting a reluctance to utilise domestic species near human developments and competition with humans for their preferred prey of paca and deer, which are also prized regionally as game species. Human-induced mortality of jaguars outside the protected forest was mainly associated with livestock predation. Both sexes were equally active on pastures and were persecuted at a similar rate. Many of those killed were young individuals in good body condition, suggesting high turnover rates augmented by immigration. Population simulations indicated that the observed levels of human-induced mortality could be maintained only with immigration from the protected forest. Without natal dispersers (2-4 year olds) immigrating in, the hunted population had zero probability of persisting beyond 20 years. Simulations indicated that the jaguar populations inhabiting the two main protected forest blocks in Belize could persist in isolation and maintain low levels of emigration to the unprotected population. However the probability of all three populations persisting for 100 years fell to ~50% if the migration of natal dispersers from the protected to unprotected population exceeded ~12% per year.

Name dc:type.qualificationname

Ph.D.

Level dc:type.qualificationlevel

doctoral

Grantor dc:publisher.institution

University of Southampton

Year dc:date.issued

2008

Author dc:creator

• Foster, Rebecca

Advisor dc:contributor.advisor

• Doncaster, C.P.