| Natural regeneration in forested systems provides a crucial mechanism for maintaining the physical and biological structure of forests,and therefore is fundamental to conserving their value as biodiversity reserves and providers of key ecosystem services.Yet,our understanding of the regeneration processes,and the key role they play in shaping community assembly and dynamics,still has numerous gaps.In this thesis,I focused on seed production(i.e.,seed rain),seed storage in the soil seed bank,and the seedling bank generated by this input of seeds and their germination,considered as three main interconnected pillars of forest regeneration.The general objective was to increase our understanding of the ecological processes operating at the early plant ontogenetic stages,which ultimately shape forest ecosystems.Using three main observational experiments conducted in the subtropical forest belt of Guangxi Province,south China,a severely understudied region as well as subjected to intense fragmentation and land-use change,I demonstrate that both intrinsic and extrinsic biological system properties as well as above-and belowground abiotic conditions have important impacts on recruitment success.1)Reproductive ecology allows us to understand the factors that drive recruitment and assess the health and productivity of forests.Masting is a widespread reproductive strategy that has evolved in many plant species,defined as an inter-annual and synchronous reproductive episode of high seed production.In the first observational field study,I analyzed a masting event from an endemic species-Acer campbellii subs.sinense – in Cenwanglaoshan Nature Reserve and provided useful information on this species autecology.In addition,I showed that A.campbellii seeds do not persist in the soil seed bank(zero seeds found in our seed germination trials).I recorded high seedling mortality,with 93% of the recruited individuals not surviving beyond the first year.I also demonstrated that greater seedling size,measured as height and number of leaves,increased the probability of survival,while microsite conditions such as canopy openness and slope were of secondary importance.These results suggest that intraspecific variation in seed size can have important impacts on recruitment success in species with masting reproduction,which incurs significant energy costs for the mother plant in producing copious amounts of seeds.2)Abiotic conditions,especially soil nutrients,are considered important factors that limit plant recruitment.However,their spatial distribution and information on how they affect growth and survival of seedlings in naturally regenerating communities in combination with other abiotic factors are not well understood.In my second observational field study,I focused on the response of seedling growth and survival to individual soil chemical properties,light and water availability,three of the major resource axes impacting forest regeneration.To evaluate the individual effects of these properties,I used growth and survival data of over 2000 seedlings belonging to 167 morpho-species and distributed in four subtropical forests monitored over two years.I showed that nutrients such as calcium(Ca),magnesium(Mg),and manganese(Mn)strongly influence seedling growth(Ca: β = 0.15,CI = 0.07 –0.22;Mg: β =-0.27,CI =-0.45 –-0.08;Mn: β = 0.16,CI = 0.07 – 0.24)and survival(Ca: β =-0.94,CI =-1.46 –-0.41;Mg: β =-0.95,CI =-1.51 –-0.38;Mn: β =-1.16,CI =-1.74 –-0.59),more than major nutrients that are usually recognized as limiting factors such as nitrogen and phosphorus.In addition,I showed that these patterns are not consistent from the landscape to forest to individual species level.This is the first time that such relationships are reported from natural forests.These findings are important because they bring attention to understudied nutrients,which affect forest regeneration and are subjected to significant changes.3)Herbivory by insects in natural forests plays a central role in plant productivity,ecology,and evolution,and is of critical importance in forest regeneration and dynamics.In the third observational field study,I measured the extent of damage on seedling leaves by insect consumption in four subtropical forests,and tested five hypotheses have been proposed to explain the patterns in leaf herbivory: JanzenConnell,Resource-concentration,Dietary-mixing,Plant apparency,and Resource availability.I assessed the direct and indirect effects of biotic and abiotic interactions,both within site to test whether patterns are site-specific,and in the pooled data.Using data from 1377 seedlings and over 8000 scored leaves analyzed through model averaging and structural equation modeling I showed that taller seedlings experienced a higher level of herbivore damage.No other biotic variable explained the degree of herbivory directly,but conspecific density showed an indirect effect on leaf damage by negatively mediating seedling height.In addition,abiotic factors such as canopy openness and soil nitrogen indirectly affected herbivory,in particular by influencing seedling height and density of conspecifics.This study suggests that the abiotic environment,which shapes individual size,is more important than the neighborhood composition as a determinant of plant herbivory at the seedling stage.In summary,I analyzed several ecological interactions of the seedling phase,widely considered as a critical bottleneck in plant regeneration,with the biotic and abiotic environment,providing evidence for the interconnectivity between these two components that govern successful recruitment.In addition to seedling intrinsic characteristics,I showed that abiotic environmental factors exerts a significant role in the processes happening on the forest floor,not only in terms of their effects on seedling performance(here measured as height growth and survival),but also as indirect drivers of biotic interactions between the seedling community and its consumers.Specifically,this work brought evidence of the clear relationship between soil nutrients,including base cations,manganese,and sulfur,and seedling regeneration success,as well as the indirect effects of light and nitrogen on biotic interactions.This project took place in an understudied yet biodiverse and at the same time threatened region,contributing to filling and urgent and important geographical knowledge gap.The findings presented here have the potential to be used as a foundation for future studies in forest regeneration dynamics in the subtropics as well as in developing policy instruments for reforestation,conservation and natural resource education. |