The dynamic environments in riparian zones support a variety of life-history strategies, which constitute a fundamental mechanism for development and maintenance of biodiversity in riparian forest ecosystems. To demonstrate the effect of life-history strategies on biodiversity, we investigated community-level diversity and its relationship to environmental variability in the riparian Populus euphratica forests of the Ejina Oasis. Communities were divided into 14 associations on the basis of their species impor- tance values. The Simpson＇s index, Shannon-Wiener index, Pielou＇s evenness index and the regional Whittaker＇s index were applied to calculate community diversity. An ordination of the sample plots was carried out by correspondence analysis （CA）. Biodiversity was relatively low across the entire study area, but there was high community diversity （flw = 8.09） due to the spatial heterogeneity of habitats in different plots. In addition, we investigated the relationship between biodiversity and several environmental factors, such as water availability, community components and soil conditions. We conclude that the heterogeneity of soil and water availability drives community diversity patterns in riparian zones and that community-level diversity favors the maintenance of species diversity in the P euphratica forests in the Ejina Oasis.
We conducted an investigation into the germination of seeds from individual Populus euphratica trees of different ages and growing conditions in order to discover the effect of intraspecific factors on their sexual reproduction in the Ejina Oasis, Inner Mongolia of China. By carrying out germination experiments of seeds collected at various seed dispersal periods, we found that P. euphratica seeds could be germinated intensively within 8 h from the start of the experiments, that the germination percentage decreased with the time of seed dispersal and that the germination percentage of seeds collected at the early stage of seed dispersal was 1.86 times greater than those collected at the final stage. There was no significant difference in the germination percentage or the germination index between seeds collected at the early and peak stages. The vitality of seeds from the mature forest was clearly higher than that of the half-mature and near-mature forests. The rate of branch dieback only affected seed germination at the final stage. Therefore, we conclude that the mature P. euphratica forest was the main contributor to reproduction. The results show that both the time of seed harvest and stand age were the main factors affecting the germination percentage of P. euphratica seeds.
LIU Qian-wenHAO PengLI Jing-wenTAN Zhi-gangAmugulangLUO Jing-yu
Seed dispersal is a fundamental process affecting destinies of seeds and seedlings, as well as regeneration dynamics and distribution patterns of communities. Recently however, declining Populus euphratica forests have failed to regenerate by seeding, while the spatial pattern and utility of their seed rains are unclear. In this study, we designed a model to simulate seed rains based on field investigations by fitting our observations to a normal distribution and testing the model with data from three years, with ac- ceptable results. We examined the simulated results of dispersed seeds patterns on the classification of three factors, i.e., density of Sophora alopecuroides, surface soil moisture and surface soil salinity. The results of seed rain simulation show that over 70% of seeds were dispersed and confined to each of the three plots where their mother trees located. The proportion of 3：7 seeds dispersed inside and outside each plots remained largely unaltered. The differences in the amounts of dispersed seed among the different levels of each of the factors were not significant, although the distributing pattern of P. euphratica in each plot varied. Therefore, in P. eu- phratica communities, the amount of seed does not become a constraint in subsequent germination, although the surface environment does. We conclude that successful P. euphratica seed regeneration relies on less than 30% of seeds dispersed over longer distance to colonize favorable growth habitats.
