Habitat partitioning among sympatric tinamous in semiarid woodlands of central Argentina.

Sympatric, phylogenetically related and morphologically similar species that overlap in their distributions at a regional scale display different patterns of co-occurrence at local assemblages. Occurrence of each species at local scales might be the result of interspecific competition for limiting resources. However, these patterns could also arise from species-specific habitat preferences along the abiotic or land use gradients. To assess the role of these mechanisms, we investigated niche partitioning among sympatric tinamou species occurring in semiarid woodlands of central Argentina. We used occupancy models incorporating habitat characteristics and interspecific interactions, while accounting for detectability, to examine the spatial overlap among elegant crested tinamous (Eudromia elegans), brushland tinamous (Nothoprocta cinerascens), and nothura tinamous (Nothura darwinii and N. maculosa) across a wide regional scale. In addition, we investigated time partitioning among these species by estimating the degree of overlap in their daily activity patterns. The regional distribution of the three species was influenced by the gradient in plant productivity and vegetation structure, in agreement with their broad habitat requirements. We also found that the occurrence of each species was presumably affected by the presence of one or two predator species. Models including interactions among tinamou species found weak negative and positive interactions among species pairs, suggesting that co-occurrence patterns were mainly driven by species-specific habitat use rather than interspecific competition. The three species were diurnal, showing two main peaks of activity, and overlapped widely in their overall diel activity, although subordinate species tended to shift their activity patterns to reduce encounters with the dominant tinamou species, suggesting some segregation in this niche dimension. Projected changes in rainfall seasonality and warmer conditions in this region could benefit elegant crested tinamous over the other two species, although climate and land use changes will likely have a negative impact on all tinamou species.

Assessing carnivore spatial co-occurrence and temporal overlap in the face of human interference in a semiarid forest.

Apex predators drive top-down effects in ecosystems and the loss of such species can trigger mesopredator release. This ecological process has been well documented in human-modified small areas, but for management and conservation of ecological communities, it is important to know which human factors affect apex predator occurrence and which mediate mesopredators release at large scales. We hypothesized that mesopredators would avoid spatial and temporal overlap with the apex predator, the puma; but that human perturbations (i.e., cattle raising and trophy hunting) would dampen top-down effects and mediate habitat use. We installed 16 camera traps in each of 45, 10 × 10 km grid cells in the Caldén forest region of central Argentina resulting in 706 total stations covering 61,611 km2 . We used single-season occupancy and two-species co-occurrence models and calculated the species interaction factor (SIF) to explore the contributions of habitat, biotic, and anthropic variables in explaining co-occurrence between carnivore pairs. We also used kernel density estimation techniques to analyze temporal overlap in activity patterns of the carnivore guild. We found that puma habitat use increased with abundance of large prey and with proximity to protected areas. Geoffroy's cats and skunks spatially avoided pumas and this effect was strong and mediated by distance to protected areas and game reserves, but pumas did not influence pampas fox and pampas cat space use. At medium and low levels of puma occupancy, we found evidence of spatial avoidance between three pairs of mesocarnivores. All predators were mostly nocturnal and crepuscular across seasons and mesopredators showed little consistent evidence of changing activity patterns with varying levels of puma occupancy or human interference. We found potential for mesopredator release at large scale, especially on the spatial niche axis. Our results suggest that a combination of interacting factors, in conjunction with habitat features and intervening human activities, may make mesopredator release unlikely or difficult to discern at broad scales. Overall, we believe that promoting the creation of new protected areas linked by small forest patches would likely lead to increased predator and prey abundances, as well as the interactions among carnivores inside and outside of protected areas.