All of the habits clearly presented sigmoidal relationship even if log-transforming area and you can variety (Fig 2)
Assemblage specific models varied in the predicted species richness based on their sensitivity to deforestation. Pair-wise permutation tests comparing predicted species richness between the entire small mammal assemblage, native species forest assemblage, and forest-specialist assemblage for the Atlantic Forest of Paraguay were highly significant (P < 0.0001; Fig 2). SESARS models for the entire and native species forest assemblages showed parallel patterns and, as expected, the entire assemblage showed consistently higher species richness throughout the region. For Atlantic Forest remnants in Paraguay that were 0.50 ha and greater, species richness estimates varied from 6 to 12 species for the entire assemblage, 5 to 10 species for the native species forest assemblage, and 2 to 5 species for the forest-specialist assemblage (including all the likely models: logistic, power, exponential, and rational). Species richness and area plots of the raw data showed that most of the species accumulations appeared at relatively small areas (Fig 2C) and with different hypothetical sampling efforts for the entire assemblage and native species forest assemblage, and most species accumulation was reached for forest specialists when forest area was considerably larger.
The new plots reveal An excellent) log-city and predicted variety matchmaking; B) urban area and you will predict kinds richness dating, and this reveal that extremely kinds accumulations was in fact achieved at seemingly short forest parts; and C) new log urban area and you can log types matchmaking which might be beneficial to have analysis out-of patterns out of varieties accumulations
Plots represent the best-complement patterns for the entire assemblage regarding short mammals (SppEntire), the new local species tree (SppForest), and you will four types-area patterns to have tree specialists (logistic: SppLog, power: SppPow, exponential: SppExp, and you can proportion: SppRat). More