Espécies mais frequentes de formigas (Hymenotera - Formicidae) como indicadoras da composição de espécies no tempo em uma floresta ombrófila densa na Amazônia Central

Abstract

The little of knowledge about ecosystems and the constant degradation related mainly to anthropogenic action makes difficult the collection of data by researchers in a timely manner, as a consequence many organisms can disappear before they are even described. As a way of expediting a survey and management of these areas, an alternative would be to focus on a smaller number, surrogate taxa, which are able to provide satisfactory answers within a short time and to know how these species are distributed in that environment. One group that can help with this are the ants, which represent a large proportion of animal biomass and play an important role in the trophic networks in which they participate. The study was carried out in a 25 km2 trails system in the Ducke Reserve, the ants were collected through pit-fall traps, in the years of 2006 and 2012, identified at species level through specific keys. In order to investigate how the exclusion of rare / infrequent species affects the prediction power of the reduced assemblages in relation to the total assemblage, a dissimilarity analysis was first carried out, followed by a principal coordinate analysis (PCoA) and then the matrices were compared with the Procrustes test. It was also evaluated if the matrices with reduced number of species and those with the total of species had similar responses against the environmental variables. Multivariate analysis of variance (MANOVA) was used to estimate how much of the variation in response data (matrices with number of reduced species and total species) is explained jointly by the environmental variables mentioned above. Correlations of Procrustes for the years 2006 and 2012 indicated that the reduced datasets were able to predict the main pattern observed with the original matrix. MANOVA indicated that the ground-dwelling ants assemblage is related to the variables clay content, slope of terrain and leaf litter volume, this result is maintained even with the removal of the most infrequent species independently of the year. Despite the reduction in the original matrix and considering the values obtained for the most 9frequent species, it was still possible to observe a composition similar to the total collected assembly and the relationships with the environmental variables had results comparable to the maximum number of species. The present study addressed for the first time the possibility of the use of frequent species of ants as surrogate taxa for assemblages of species in the Amazon. This tool helps in future work the reduction of time, and costs of sorting and identification of material collected, in addition to potentializing area management.