Abstract Conservation soil management systems can promote beneficial changes in the edaphic fauna, which is important in improving and maintaining soil quality. Hence, this study aimed to evaluate the edaphic fauna in two areas with soils managed under no-till with and without terraces for four years. The edaphic fauna was evaluated by installing 32 pitfall trapsin each experimental plot. After seven days, the traps were removed, and the sampled individuals were classified at the level of major taxonomic groups. Collembola, Acari, Formicidae, Coleoptera, and Araneae were the most abundant in both study areas. The mean abundance of the order Coleoptera stood out in the no-till system with terraces in 2021, while the orders Collembola and Acari were more abundant without terraces in 2019 and 2021, and 2021 and 2022, respectively. There was a greater abundance of organisms for both areas in 2021, with significant equitability in the no-till system with terraces. Thus, the results showed that only some groups are positively affected by mechanical erosion control. quality Hence notill no till years 3 plot days removed Formicidae 201 202 2022 respectively Thus control 20 2

RESUMO Neste estudo, o objetivo foi avaliar a fauna epiedáfica em diferentes agroecossistemas utilizando dois métodos de coleta. Os organismos foram amostrados em quatro agroecossistemas: (i) cultivo de milho, (ii) cultivo de Brachiaria ruziziensis, (iii) sistema silvipastoril, e (iv) remanescente florestal secundário. Em cada agroecossistema, 5 armadilhas do método Pitfall traps e 5 do método Provid foram instaladas, espaçadas em 10 metros entre si e mantidas em campo por três dias. Os organismos coletados foram classificados ao menor nível taxonômico possível e os dados obtidos, avaliados pelos índices de diversidade de Shannon e de dominância de Simpson, e análise de componentes principais. O método Pitfall traps resultou em maior abundância da comunidade epiedáfica em relação ao método Provid. A área cultivada com milho teve maior abundância em relação aos demais agroecossistemas avaliados. Os melhores valores de diversidade foram observados no cultivo de B. ruziziensis, por ambos os métodos. As áreas de cultivo de milho e sistema silvipastoril foram separadas pelos dois métodos pela análise de componentes principais. O maior número de grupos da fauna epidáfica ocorreu na área cultivada com B. ruziziensis e ao remanescente florestal. estudo coleta i (i ii (ii iii (iii iv (iv secundário agroecossistema instaladas 1 dias obtidos Simpson principais B

ABSTRACT In this study, the objective was to evaluate the epiedaphic fauna in different agroecosystems using two sampling methods. Organisms were sampled in four agroecosystems: (i) corn cultivation, (ii) Brachiaria ruziziensis cultivation, (iii) silvopastoral system, and (iv) secondary forest remnant. In each agroecosystem, 5 Pitfall traps and 5 Provid traps were installed at a distance of 10 meters and kept in the field for three days. The sampled organisms were classified at the lowest possible taxonomic level and the data obtained were analyzed using Shannon’s diversity and Simpson’s dominance indices, and principal component analysis. The Pitfall traps method resulted in greater abundance of the epiedaphic community compared to the Provid traps method. The corn cultivation area had greater abundance compared to the other agroecosystems evaluated. The best ecological index was observed in the B. ruziziensis crop, using both methods. The areas of corn cultivation and silvopastoral system were separated by the two methods, by principal component analysis, with a greater number of groups associated with the area cultivated with B. ruziziensis and the secondary forest remnant. study methods i (i ii (ii iii (iii iv (iv remnant agroecosystem 1 days Shannons Shannon s Simpsons Simpson indices analysis evaluated B crop