A large fraction of tree species forming persistent soil seed-banks and with dormant seeds are expected to be found in strongly seasonal ecosystems such as Neotropical savannas, where seedling recruitment could be highly variable. In the savannas of Central Brazil, we studied seed characteristics (type of dormancy, longevity and moisture content) of 14 representative woody species differing in seed dispersal season. We also studied the dynamics of soil seed-banks and similarity patterns in woody species composition among seed rain, soil seed-bank, seedling bank and standing vegetation along shallow topographic gradients that differ in canopy cover. Woody species composition of the soil seed-bank largely differed from the standing vegetation, the seed rain and the seedling bank species composition, suggesting low recruitment of woody species from the soil seed-bank. Seeds of the 14 woody species remained viable for less than 16 months in laboratory dry-storage conditions. Of those, most seeds dispersed in the dry season were dormant and exhibited low moisture content, while most seeds dispersed in the wet season were non-dormant and exhibited high moisture content. Longevity of these seeds dispersed in the dry and the wet seasons did not differ significantly. This study shows that both timing of seed dispersal and dormancy appear to control timing of seed germination and seedling recruitment of most Neotropical savanna woody species, which did not form persistent soil seed-banks. This study contributes to the understanding of tree/grass coexistence and tree density variations along topographic gradients in tropical savannas.
Summary1. Seedling dynamics are crucial for understanding spatial plant distribution patterns, yet little is known about seedling establishment in Neotropical savannas because empirical studies at the community level are scarce. 2. Over 2 years, we studied the recruitment and survival of an initial seedling assemblage and three cohorts of recruits of woody plants within 216 plots of 1 m 2 located along a tree density gradient in the savannas of central Brazil. These savannas differ in tree density and canopy cover, from closed (high canopy cover) to open savannas (low canopy cover), and are located along shallow topographic gradients. 3. We measured community-wide seedling limitation (i.e. proportion of 1-m 2 plots without seedlings of any woody species), photosynthetic photon flux density, litter cover, soil moisture and soil nutrients in each savanna type. Because closed savannas had lower PPFD and higher leaf litter cover than open savannas, we evaluated the effects of light level and litter cover on seedling emergence of nine dominant savanna woody species under controlled conditions in a glasshouse. 4. Density, recruitment and survival of seedlings decreased over time because of mortality in all savanna types, but they were consistently higher in closed than in open savannas. Community-wide seedling limitation was significantly lower in closed (0.16 ± 0.03) than in open (0.30 ± 0.05) savannas. 5. In the glasshouse, high litter cover and very low light levels reduced seedling emergence of most species, suggesting an adaptation to delay seed germination until the wet season when soil water availability is high and leaf litter rapidly decomposes. 6. Synthesis: In Neotropical savannas, tree canopy cover facilitates seedling establishment of woody species by reducing stressful environmental conditions. In particular, low irradiance and high litter cover in closed savannas enhance the recruitment and survival of woody seedlings relative to open savannas by reducing soil water deficits and increasing nutrient availability in the upper soil layers. The higher seedling limitation of tree species in open than in closed savannas contributes to maintain relatively different balances between trees and herbaceous plants along topographic gradients in Neotropical savannas and helps to explain spatial distribution patterns of woody species in these ecosystems.
The failure of seeds to arrive at all suitable sites (seed limitation) greatly affects plant distribution and abundance. In contrast to tropical forests, the degree of seed limitation in Neotropical savannas is unclear because empirical studies at the community level are scarce. We estimated seed limitation of 23 woody species from annual seed rain measurements along a tree density gradient in the savannas of Central Brazil. These savannas differ in tree density and canopy cover, from closed to open savannas, and are located along shallow topographic gradients. We also studied post-dispersal seed predation and removal of 17 representative woody species, and seed viability loss over time of 12 common woody species under dry-storage conditions. Annual seed rain was lower in open (410 seeds/m 2 ) than in closed savannas (773 seeds/m 2 ). Average seed limitation across woody species was higher than 80% along the tree density gradient. More than 60% of seeds of the studied woody species were predated or removed within 30-45 days in all savannah types. Seeds of most common woody species (66%) lost their viability in less than 12 months of dry storage. This study shows that Neotropical savannah woody plants are strongly seedlimited because of low and poor distribution of seeds among sites, post-dispersal seed removal, and short seed longevity. The high seed limitation of tree species in Neotropical savannas, particularly in open savannas, also may contribute to maintain their relatively low tree densities and help to explain the spatial variation of tree abundance along topographic gradients.
Questions What are the temporal and spatial variations of groundwater depth along topographic gradients in Neotropical savannas? Are patterns of tree density and species diversity along topographic gradients in Neotropical savannas controlled by water table depth? Do soil and groundwater nutrient concentrations differ along topographic gradients in Neotropical savannas? Location Fire‐protected savannas of central Brazil. Methods Over 16 mo, we monitored temporal and spatial variations in groundwater levels using automated submersible pressure transducers installed in ten wells located along two topographic gradients (five wells per gradient) of 950 m and 1703 m in length, representing elevations of 47 and 37 m a.s.l., respectively. We located the wells according to changes in vegetation physiognomies from woody savannas at high elevations, to open shrubby grasslands at low elevations. Along each topographic gradient we determined soil and groundwater nutrient concentrations as well as richness, density, basal diameter and height of trees within two plots of 14 × 14 m (392 m2) adjacent to each well. Results Along the two gradients, groundwater levels exhibited larger fluctuations at lower than at higher elevations where the water table was deeper. Richness, density and diversity of trees decreased significantly at lower elevations where soils were waterlogged during the wet season. Soil pH and soil concentrations of carbon, nitrogen and manganese decreased significantly as elevation increased along the topographic gradients, but soil nutrient concentrations of phosphorus, aluminium and iron did not change with elevation. Groundwater samples contained only trace amounts of nutrients and were poorly correlated with elevation along the topographic gradients. Conclusions In Neotropical savannas, the minimum distance between the soil surface and water table depth (reached during the wet season) and the relatively large fluctuations in groundwater limit tree density and diversity at low elevations as savanna trees cannot cope with extended waterlogging during the wet season and with low soil water availability during the dry season. Thus, variations of tree density and diversity along topographic gradients are more related to spatial and temporal variations in water table depth than to soil and groundwater nutrient variations in Neotropical savannas.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.