Abstract: Aims Rhododendron xiaoxidongense is endemic, rare and endangered species in China, with a narrow distribution in the central part of the Luoxiao Mountains. The R. xiaoxidongense was once assessed as extinct (EX) because its habitat and individual numbers are so rare that it is difficult to find in the wild. However, as details on the resource status, population structure and dynamics of the species are lacking, the conservation of this species is severely constrained. This study aims to clarify the survival status and future development trend of the population, and key factors to affect population regeneration through to analyses structure and dynamics characteristics of the existing populations of R. xiaoxidongense, which will provide scientific basis for the conservation, management, and revitalization of wild populations. Methods Based on the investigated parameters, the age structure of the R. xiaoxidongense population was obtained by using a space-for-time substitution method. Subsequently, the dynamic index, static life table, survival curves and survival function curves of the population were determined to analyze the population structure characteristics and survival potential. The time series prediction model was employed to predict the future development trend of the population, and the aggregation degree index was used to determine the spatial distribution pattern of the population. Important findings (1) The age structure of the four populations of R. xiaoxidongense was incomplete, with an overall trend of more middle-aged and mature individuals and a severe lack of low-aged and old individuals, the age structure tended to be a decline type. (2) The dynamic index of adjacent age class (Vn) showed a fluctuating trend of “decline-growth-stability-decline” with increasing age class, the dynamic index without external interference (Vpi) > the dynamic index under external disturbance (V′pi) > 0, V′pi was closed to 0. The maximum risk probability in response to random disturbance (Pmax) under external environmental interference was 11.11%, indicating that the population was high risk probability to external disturbance and had a very weak resistance to disturbances. (3) The life expectancy (ex) was highest at age class I, and the population survival curve tended to be Deevey-II type. (4) The mortality rate (qx) and vanish rate (Kx) curves showed a dynamic change of “increase-decrease-increase”, which indicated the population was in an unstable state. (5) The survival function curve showed the trend of sharp decline in the early stages, relative stability in the middle stages, and slow decline in the later stages, and entered the decline phase at a relatively small age class (2.25 age class), which indicated that the population had weak viability. (6) In the future, after the 2, 4, 6, and 8 age classes, the population number of low and middle-aged individuals would decrease, in contrast, the number of adult and old individuals would increase, indicating that the population will face a risk of decline. (7) The overall spatial pattern of the population was clumped distribution, but the degree of aggregation decreased with increasing age class, transitioning to random distribution in age classes VII-VIII. Conclusion In summary, the small population size, narrow distribution range, low resistance to external disturbance, and difficulty in seedling regeneration are the likely factors contributing to the endangered status of the R. xiaoxidongense. We suggest strengthening the habitat protection and tending management of R. xiaoxidongense population, and achieving population conservation and revitalization through multiple approaches, including in situ and near situ conservation.

Key words: Rhododendron xiaoxidongense, static life table, survival function, time sequence prediction, spatial pattern, conservation strategies