Aims: In dry tropical forests, herbaceous species may have dormancy mechanisms and form persistent and transient seed banks in the soil. Evolutionarily acquired, these mechanisms are efficient for the establishment and survival of these herbs, especially in forests with unpredictable climates, such as the Caatinga. Thus, our objective was to verify whether the studied herbaceous species adopt the physical dormancy mechanism and how these natural barriers are overcome, to understand the temporal dynamics existing in the soil seed bank from a Brazilian dry tropical forest.Methods: Seeds of five native herbaceous species from the Caatinga forests were selected and submitted to pre-germinative treatments for verifying the presence of physical dormancy. We collected soil samples in the rainy and dry seasons for four consecutive years and monitored the emergence of the selected herbaceous in the greenhouse. We verified the differences in germination and seed bank emergence in the soil by generalized linear models.Important Findings: The presence and absence of physical dormancy were observed in seeds from Caatinga herbaceous species. We found intraspecific and interspecific differences in the herbaceous emergence from soil seed banks between years and climatic seasons. In perennial herbs, consecutive lack of emergence between seasons and years was frequent, which suggests a direct relationship with the mechanism of physical dormancy and the environmental conditions necessary to overcome integument barriers. In these species, seed dimorphism and dormancy may confer additional advantages to their survival. Moreover, presenting intermediate levels of physical dormancy in an annual species may be an evolutionary adjustment to rainfall unpredictability. In contrast, we found that the annual herb without dormancy is more sensitive to seasonal and interannual climate changes, as evidenced by the increase and significant reduction of its emergence in the soil seed bank. These differences acquired evolutionarily are advantageous for the establishment of herbaceous populations, mainly in semiarid regions with an unpredictable climate.
Soil seed bank contributes to the maintenance and renewal of the plant community. Species richness and density of the soil seed bank are affected by the temporal and spatial variations occurring in dry environments. Thus, this study aimed to characterize the effect of time (climatic season and annual precipitation) and space (soil depth and microhabitat type) on floristic composition, species richness, and seed density in the soil bank from a Brazilian dry forest over seven consecutive years. During the study period, 0–5 cm soil samples were collected in three types of microhabitats (Flat, Rocky, and Riverside) from the litter and soil, totaling 2940 samples. The seed bank was quantified using the seedling emergence method; each collected sample was monitored for six months. All data on species richness and seedling emergence were analyzed using the GLM and Anosim test for floristic composition. Results show that the floristic composition, species richness, and seed density differ between climatic seasons, between years of study, between soil depths, and between the three types of microhabitats found in the studied area. However, the explanatory power of each variable was low, demonstrating that other variables such as temperature, wind action, and predation may also be influencing seed storage in the soil bank. In summary, this study demonstrated that the richness, species composition, and emergence of seedlings recorded in soil seed banks are influenced by the spatiotemporal variation found in dry environments. Thus, all the tested variables partially explained the behavior observed in the seed bank.
The Tropical dry forests cover large regions of the world and have been extensively altered by human activities. This study was aimed at characterizing and comparing the density, diversity and structural parameters of a human-altered secondary dry within an interval of five years of regeneration. Two hundred plots were established in the forest and two monitoring surveys were carried out, with interval of five years between them. All individuals in each plot with diameter at the ground level ≥ 3 cm were measured. After five years, no differences were observed in the number of species. Total density and mean basal area decreased, as a result of the maturation of the forest and possibly due to a prolonged drought in 2012. In 2008, the total basal area was 10.59 m².ha-1, increasing to 11.01 m².ha-1 in 2013. The mean height of the community did not increase significantly, from 3.59 m in 2008 to 3.65 min 2013. The number of individuals in the larger diameter classes reduced between 2008 and 2013, while the opposite was observed for smaller classes. Some structural parameters increased or decreased depending on the group of species analyzed and there was still a group that did not present significant changes. Therefore, age since abandonment can affect the regeneration process and the resilience of the forest. We realized, the speed in which these alterations occurred in an interval of five years can be considered slow and changes in the recovery process of a 16-year-old forest may not be detected.