The Brazilian Amazonia is a region covered by an extensive mosaic of tropical forests conditioned by different topographical and hydro-edaphic features. Although studies relating environmental determinants of structure and floristic composition are systematically evolving in the region, there is no doubt that there are still information gaps due to the lack of research in peripheric areas of the Amazonia. The seasonally flooded areas of the state of Roraima situated on rio Branco-rio Negro basin, northern Brazilian Amazonia, still are deprived of such information. In this way, this work had as objective determine the physical and soil chemical attributes, and the flooding periodicity that characterize different forest types dispersed in a topographic gradient located in an area on the north of rio Branco-rio Negro basin. Soil samples (0-60 cm) were collected along a 2.7 km transect (31.1-64.8 m a.s.l.) crossing three different forest types: (i) mosaic between treed and forested shade-loving (La+Ld), (ii) area of ecological tension between forested shade-loving and open ombrophilous forest (LO) and (iii) open ombrophilous forest (Ab+As). The results indicated different soil classes and flooding periodicity for each forest type observed: Entisols Fluvents (La+Ld, 3-4 months flooded), Entisols Quartzipsamments (LO, 1-2 months) and Yellow Ultisols (Ab+As, no flooding). All analyzed soils were defined as nutrient-poor areas, especially those located on low altitude, characterized for higher hydrological restrictions (seasonal flooding) aggregating forest types of lower structural patterns (e.g. La+Ld). Soils on low altitude were also characterized as those with the highest percentage of fine sand and silt, while soil free of seasonal flooding (Yellow Ultisols) presented the highest levels of clay and coarse sand, always associated with the ombrophilous forests (higher structural patterns). These results improve our understanding of the environmental factors conditioning different forest types in this peripheral region of Amazonia, suggesting that ecosystems with higher hydro-edaphic restrictions are a strong indicator of forest types with lower structural patterns.

Aims: In the Brazilian Amazon, the practices and use of agroforestry systems (SAF) are increasingly used, the proper management of them helps to improve soil properties and also to prevent their degradation. In this context, the objective of this work was to evaluate the biological attributes of the soil in the oil palm cultivation systems with intercropping. Study Design: The experimental design used was completely randomized with four repetitions and six treatments: Oil palm (Elaeis guineense Jacq.) interspersed with pineapple (OPi), bean (OBe), banana (OBa), yucca (OYu) and Brachiaria humidicola (OPa), as well as adjacent area only with Brachiaria humidicola as a witness (Pa). The Tukey test was used at a level of 5% probability in samples analyzed at a depth of 0-0.0 m, to compare the means of the variables evaluated. Place and Duration of Study: The experimental area is located of São João da Baliza, vicinal 26, km 12, with geographic coordinates of reference 00º.51'13.3''N and 60º00'19.8''W, the altitude of 100 msnm and, distant to 352 km from the capital Boa Vista, state of Roraima realized in 2016. Results: The TOC presented values between 4.70 and 9.45 g kg-1, being the highest values found in the interim systems OYu, Pa, OBa, highlighting the intermediate system OPi that presented the lowest levels. The highest basal respiration values ​​of the soil (RBS) (23.50 mg C-CO2 kg-1 soil h-1) and carbon from microbial biomass (C-BMS) (116.0 mg C microbiano kg-1 soil) were verified in the pasture system. Likewise, for the urease and acid phosphatase activity, the grass system stands out as a control with values of (148.42 g NH4+ g-1 soil 2 h-1) y (230 μg de p-nitrofenol g-1 soil h-1) followed by palm with grass and yucca systems. However, the β-glucosidase activity (51.22 μg p-nitrofenol g-1 h-1) it was positively influenced by the oil palm system with yucca. On the other hand, the system interspersed with pineapple showed a higher metabolic coefficient (qCO2) (0.36 mg C-CO2 g-1 C-BMS h-1). It can be concluded that the pasture system (Pa) is presented as a more stable environment, followed by interspersed systems of oil palm with grass (OPa) and yucca (OYu).

No abstract