Fauna edáfica sob modelos em estágio inicial de restauração de floresta subtropical

The goal of this study was to compare areas under different forest restoration technologies in relation to abundance, richness, diversity and composition of the present fauna in the litter and soil. The treatments evaluated were: natural regeneration (RN); high diversity tree plantations (L) and nucleation (N). An area of secondary forest was included in the study as a reference of soil conditions. The experimental design was in randomized block with four replications. Samples were collected for extraction of mesofauna (October/2012; July and October/2013) and macrofauna (June and October/2013) in six points of each plot, totaling 24 samples per treatment. For collecting soil macrofauna was used TSBF method. The mesofauna was collected with a metal cylinder and extracted by Berlese-Tüllgren funnel. Litter and soil were collected separately at each point and the fauna was identified level of class/taxonomic order. The springtails were classified using morphotypes. In total, considering the mesofauna, macrofauna and three times collected were accounted 28618 organisms. In relation soil mesofauna, the evaluated technologies did not differ, after three years of restoration, in relation to total abundance of organisms and community composition. The Shannon diversity index (H), in soil mesofauna, followed a human impact gradient. This index was higher in natural regeneration, which not was undergone technical interventions and showed higher moisture in the soil. The tree planting technology, under the control of volunteer plants in total area, showed lower H index. In the case of litter mesofauna, the technologies did not differ in relation the mean richness, total abundance of organisms and community composition. Considering edaphic macrofauna, technologies did not differ in relation to the abundance and richness, and in the evaluation of June/2013, RN showed higher H index and differed in relation to the community composition of other technologies. In October/2013 evaluation, the differences between the technologies in relation to H index were narrower and these did not differ in terms of composition of soil macrofauna community. In litter macrofauna, in June/2013, the RN presented greater richness and H index when compared to other technologies and in evaluation October/2013 technologies did not differ in relation to community composition, richness and mean abundance of organisms. In the case of springtails, technologies after three years in the restoration process, did not differ in relation to the abundance, richness and composition of Collembola community for different morphotypes. The secondary forest, in relation to forest restoration technologies, presented greater abundance of saprophages, predators and greater diversity of morphotypes of springtails. From these results it, we recommended to natural regeneration by to have the lowest cost of deployment, followed by nucleation and online planting. The animals should be monitored over time, in the restoration technologies, as well as the physical and chemical characteristics of the soil, in order to understand the possible changes in the composition and diversity of organisms.