Influência de fatores abióticos sobre a distribuição de colêmbolos (Collembola: Entomobryomorpha) edáficos e redução do esforço amostral em floresta ombrófila densa de terra-firme na Amazônia Central, Brasil

Abstract

From ecological studies with Collembola, the potential of the information obtained has been strengthened. Such information, in first instance, can be used in basic ecological studies, or directed to applied studies aiming to understand, monitor and maintain the stability of a community. However, studies with Collembola, due to the diverse nature and abundance of these organisms are time demanding and require high financial costs. The present study investigated the relationships of Collembola Entomobryomorpha edaphics community with Carbon, Nitrogen e Phosphorus (C, N, and P) concentrations in the soil, and the possibility of sampling effort reduction was estimated. Using a 3.5 cm² soil core, five sub-samples were collected from soil and litter in 30 plots of 250 m, to extract the springtails in the modified Berlese-Tullgren extraction. The ecological relationships were tested using multiple regression between each one of the first two axes resulted from PCoA with seven possible models using C, N and P. For the reduction of the sampling effort, the congruence of species composition at each reduction level in relation to the maximum effort composed by five subsamples per plot was estimated with the Mantel test. The efficiency of the ecological response to C, N and P at each reduction level related to the maximum effort was estimated from values of the explained variance obtained by multiple regression. A total of 1,031 individuals of Entomobryidae and Paronellinae and a richness of 13 genera and 30 species (86.7% morphospecies) was recorded. The two regression models that best explained the variation in species composition of Collembola edaphics showed that N, P and C explained 14% to 19% of the distribution of these animals in the landscape. The substitution pattern of species as a function of the gradients of C, N and P was graphically visualized. The sampling effort per plot could be reduced to four subsamples, i.e., a reduction of 20%, reducing financial costs and 60 working days. The changes of C, N and P concentrations influence the distribution of species, indicating that the stoichiometric ecological processes can be used to predict changes in the community. With the use of reduced sampling effort, the financial resources and time saved can be directed to purchase materials for the laboratory and also assist other field collections and increase the number of sampled plots.