ABSTRACT
BACKGROUND: Since 1996, when Vivo questioned how many species of mammals occur in Brazil, there has been a huge effort to assess this biodiversity. In this contribution, we present new records for rare species of the sigmodontine rodent genera Rhagomys and Neusticomys previously unknown to Brazilian Amazon. We provided detailed information on the morphologic variation to allow the proper identification of these species. We also furnished updated information on their collection, aiming to establish hypothesis of their geographic distribution, based on SDM's, aiming to hypothesize potential occurrence areas for these species. METHODS: Rodent specimens were sampled in separate inventories in two sites of Rondônia State (Hydroelectric Dam Jirau and Parque Nacional de Pacaás Novos) and one site in Pará State (Pacajá), Brazil, and were compared to specimens from museum collections to apply appropriate names. The SDM were conducted using two algorithms for rare species, MaxEnt and randomForest (RF), and were based on seven localities for Rhagomys, and 10 for Neusticomys. RESULTS: All specimens were collected with pitfall traps. One specimen of genus Rhagomys was trapped in the Hydroelectric Dam Jirau. We identified this specimen as R. longilingua, and the SDM species indicates suitable areas for its occurrence at high elevations near on the Andes and lowlands of Amazon Basin to the South of the Rio Amazonas. Two specimens of Neusticomys were recorded, and we identified the specimen from Pacaás Novos as N. peruviensis, with SDM suggesting main areas of occurrence on Western Amazon. We applied the name N. ferreirai to the specimen from Pacajá, with SDM recovering suitable areas in Eastern Amazon. DISCUSSION: We reinforced the importance of pitfall traps on the study of Neotropical rodents. We described morphologic variation within and among all species that do not invalidate their specific status, but in the near future a re-evaluation will be mandatory. The new records extended the species distribution considerably. SDM was successful to predict their distributions, as the two algorithms presented important differences in range size recovered by the models that can be explained by differences in the thresholds used for the construction of the models. Most suitable areas coincide with the areas facing most of the deforestation in Amazon. We added two rare species of sigmodontine rodents to the list of Brazilian Mammals, which now comprises 722 species (or 775 valid nominal taxa). Although more information is available than in 1996, it is essential that mammal experts maintain inventory and revisionary programs to update and revise this information. This is even more important, as changes in Brazilian environmental legislation are being discussed, suggesting reduced need for environmental impact reports prior to beginning commercial enterprises, resulting in the loss of information about native biodiversity in the affected areas.
ABSTRACT
The contribution of small mammal ecology to the understanding of macroecological patterns of biodiversity, population dynamics, and community assembly has been hindered by the absence of large datasets of small mammal communities from tropical regions. Here we compile the largest dataset of inventories of small mammal communities for the Neotropical region. The dataset reviews small mammal communities from the Atlantic forest of South America, one of the regions with the highest diversity of small mammals and a global biodiversity hotspot, though currently covering less than 12% of its original area due to anthropogenic pressures. The dataset comprises 136 references from 300 locations covering seven vegetation types of tropical and subtropical Atlantic forests of South America, and presents data on species composition, richness, and relative abundance (captures/trap-nights). One paper was published more than 70 yr ago, but 80% of them were published after 2000. The dataset comprises 53,518 individuals of 124 species of small mammals, including 30 species of marsupials and 94 species of rodents. Species richness averaged 8.2 species (1-21) per site. Only two species occurred in more than 50% of the sites (the common opossum, Didelphis aurita and black-footed pigmy rice rat Oligoryzomys nigripes). Mean species abundance varied 430-fold, from 4.3 to 0.01 individuals/trap-night. The dataset also revealed a hyper-dominance of 22 species that comprised 78.29% of all individuals captured, with only seven species representing 44% of all captures. The information contained on this dataset can be applied in the study of macroecological patterns of biodiversity, communities, and populations, but also to evaluate the ecological consequences of fragmentation and defaunation, and predict disease outbreaks, trophic interactions and community dynamics in this biodiversity hotspot.
