Community Assembly During Secondary Forest Succession On The Loess Plateau,China

The study of plant community assembly is integral for understanding species coexistence and biodiversity maintenance, and it has long been a central issue in community ecology. Although validation studies of niche theory and neutral theory have achieved great advances in recent years, some difficulty remains in understanding the mechanisms of local community assembly. Succession is community assembly in action, and has served as a lens to understand how ecological communities are assembled.Exploring the mechanism of plant community assembly across secondary succession sireis and the effect of environment factors on the change of community assembly are of important in ecological theory and restoration and reconstruction of degraded ecosystem. Based on multi-dimensionality of plant fuctional traits, community weight traits, species diversity, functional diversity, phylogenetic diversity, soil factor and topoghrphy characteristics, we tested the relative importance of deterministic and stochastic process, as well as environmental filtering and similarity constraint, across whole 6 successional stages from abandoned field to climax forest on the Loess Plateau of northern Shaanxi. By means of null model, community trait gradient analysis, we explored the response of traits to environment at community level, elucidated the effect of environment on community assembly of different successional stages. Our results indicated that community assembly wasgoverned by stochasticity at the species level, but deterministicselectionwas dominant at the functional trait level and phylogenetically level.(1) Species of late forest succession stages were closely related in phylogeny and the late communities exhibited low turnover among plots, but at early herbaceous succession, distantly related species co-occur and there is a high phylogenetic turnover among plots. Shrub stage is showed as a transition from phylogenetic overdispersion to clustering. In addition, assemblages of woody plant differ from herbaceous plant assemblages during forest succession. Phylogenetic clustering dominated in woody plant assemblages, whereas multiple patterns work together in herbaceous plant assemblages. Hence, our results indicate the long-lasting evolutionary imprint on the present-day composition of plant assemblages arrayed along the succession gradient in our study region.(2) There was also a trade-off between commnities traits during succession on the Loess Plateau. Early succession communities have traits related to effcient nutrient acquisition and were located at the quick investment-return end of economics spectrum. As the succession, more woody and evergreen species occured, late succession communities increased the investment in structural support and reduced investment of nutrient resources in plant growth and respiration, were located at the slow investment-return end of economics spectrum.These results are sinigicant for up-scaleing of economics spectrum from organ, individual, species, to community level.(3) Based on fifteen functional traits from different organs, we constrcted a plant economics spectrum for all these species of Loess Plateau. This spectrum were determined by structural and stoichiometry traits from different organs and showed a turn-over from quick investment-return species to slow investment-return species.Moreover, different succession stages showed different plant economics spectrum because of special selection of combination of traits along the succession.Plant traits allow the investigation into the functions of different tissues and thus into the functional trade-offs that influence species distributions along succession gradients.(4) Relationships between leaf traits and other traits showed different patternsforintra-and inter-plots. Therefore, there were strong environmental or/and biotic filtering, these prcessions affected the trade-off of intra-and inter-organs. There was really a non-neutrality process (environmental or/and biotic filtering) related to use of nutrient, these prcessions affected the mechanisms of community assembly along succession by mediating the relationships among traits for intra- and inter- organs.(5) As succession, all measured traits exhibited significant phylogenetic signals and phylogenetically only mediated the a competents(intra-plot variation) of traits rather than the (3 component (inter-plot variation). This indicated that phylogenetically play an important effect in species composition within community and did not affected the turn over of species among plots. Our results also showed that niche breadth based on traits did not exhibited significant phylogenetic signals which indicates that phylogenetically did not affected the distribution of species.(6) Different traits showed different community assembly rules along succession. For these stoichiometric traits (leaf/root/stem nitrogen/carbon/phosphorus content), later community were assembled predominantly through environmental filtering, in contrast they were assembled predominantly through competitive exclusion for plant height and seed mass. Results of redundancy analysis showed that soil NO3 content and pH drived the over-dispersion of stoichiometric traits in early communities, and soil NH4 drived the over-dispersion seed mass in later communities.After all, Community assembly is stochastic at the species level, but deterministic at the functional trait level and phylogenetically level. The fulfillment of this study provided a new evidence of transitional assembly process in warm temperate forest zone and has a bearing on community assembly theory, and will simultaneously provide application basis to guide vegetation restoration and reconstruction in region of the Loess Plateau.