|
||||||||||
|
Alvarenga, G.C. | |
Spatial dynamics of medium- and large-sized mammal assemblages in v rzea and terra firme forests, Central Amazonia, Brazil | |
2017 Full Book | |
Studies that try to understand how mammals use the space are extremely important to improve conservation decisions. Despite the high knowledge of mammals in the Amazon forest, little is known about this group in the varzea forests. These forests are located in the lowland inundated regions of the Amazon basin, and are seasonally inundated by white-water rivers. Varzea forests were, and still are, intensively used by humans, because of the easy access and the high fertility of it soils. Beyond that, climatic changes and the actual Brazilian politics, which encourages constructions of hydroelectric plants, are direct threats to the varzea and the animals that live there. The goal of this study is understand how variations on the vegetation influence the distribution of medium and large-sized mammals in the varzea forest. For this, we executed two consecutive years of camera traps surveys during dry season in Sustainable Development Reserves AmanÆ and Mamirau , Central Amazonia. Specifically, we compared the community composition of mammals between an upland forest (AmanÆ) and a varzea forest (Mamirau ), and we tested the response of mammal's species to the vegetation variation in the varzea forest. The sampling grid in each reserve consisted of up to 50 baited camera trap stations and an overall sampling effort of 4075 camera trap*days. We used a pre-exist landscape mapping of Mamirau Reserve and from it we calculated the area (kmý) of the habitats inside two different scales (500 m and 1000 m) around each camera trap station. The comparison between the mammal species of upland and varzea was made with an ordination by NMDS of two dimensions and after that a Similarity Analysis (ANOSIM), both using Bray-Curtis indice. To test the influence of the varzea vegetation's in the mammal distribution we used Generalized Linear Models (GLM) and, for the species with too many zeros, ZeroInflated Models (ZIP/ZINB) or Hurdle Models (ZAP/ZANB). During the study, we recorded 21 medium and large-sized mammal species of 3443 records. Among this species, seven are classified as 'vulnerable' globally or in Brazil. As expected the community composition of mammals were different between upland and varzea forest. At the varzea we recorded only six species, while upland recorded 20 species. The mammal's community of v rzea forest presented a hierarchical subgroup pattern of the upland community, and all the species registered in the v rzea have semi-arboreal habits. Tree species recorded in varzea forest were more represented there than in upland forest, which could be related with biggest populations at this region. The varzea vegetation's were related with almost all species distributions, the exception was _Leopardus wiedii_. The total number of records and the composition of species were influenced by the habitat chavascal, such there were fewer records with chavascal area increase. At the same way, _Didelphis marsupialis_ and _Nasua nasua_ avoided areas of chavascal. _N. nasua_ avoided too the open fields of herbaceous vegetation. The species _Coendou prehensilis_ was positively influenced by constant water habitats and high varzea forest, while _Panthera onca_ avoided high varzea forest. |
|
PDF files are only accessible to Friends of the Cat Group. Joining Friends of the Cat Group gives you unlimited access and downloads in the Cat SG Library for one year, and allows you to receive our newsletter Cat News (2 regular issues per year plus special issues). More information how to join here |
(c) IUCN/SSC Cat Specialist Group ( IUCN - The World Conservation Union) |