Species–area relationship of Neotropical waterbird assemblages in remnant wetlands: looking at the mechanisms

Aim  We investigated the relative role of area, isolation, microhabitat diversity and number of individuals as explanatory factors defining the richness of waterbirds in wetland remnants. Location  Freshwater marshes along the Atlantic coastal zone of South Brazil. (30°56′–30°22′S; 50°58–50°22′W; Fig. 1). 1The study area (dashed line), in the Atlantic coastal zone of Brazil, is characterized by a high concentration of remnant wetlands used by resident and migratory waterbirds. Waterbirds were surveyed monthly along 2003 in 42 randomly selected wetland remnants.Methods  We surveyed waterbirds monthly for 1 year in 42 wetland remnants. We measured area, diversity of microhabitats, and isolation using aerial photographs and Landsat images. We compared the fit of the relationship between waterbird richness and wetland area across three models. We evaluated the ‘passive sampling effect’ comparing the observed species–area relationship with area‐based and individual‐based null models, performed with randomization without replacement. We used Path Analysis to measure the relative importance of direct and indirect effects of wetland area, isolation, microhabitat diversity and the total number of individuals on species richness. Results  Power functions showed that larger wetlands had proportionally fewer species, individuals and microhabitats as compared to smaller wetlands. Species richness was consistently lower than predicted by area‐based and individual‐based random placement null models. Path analysis showed significant direct effects of remnant area and microhabitat diversity on species richness, but no effect of isolation on species richness. When number of individuals was included in the model, it became the most important factor explaining species richness, followed by wetland area. The number of individuals was significantly influenced by area, isolation and microhabitat diversity. Main conclusions  We demonstrated that waterbird richness in fragmented landscapes is not a random placement phenomenon, but a biological one, significantly affected by the complex interplay of remnant area and isolation and diversity of microhabitats. We showed that the number of individuals plays a central role mediating the effect of these physical factors on species richness.