Qiongzhuea tumidinoda stands out as an endemic bamboo species of significant conservation importance in Southwest China, particularly in the upper reaches of the Yangtze River. It holds a pivotal role in poverty alleviation through the commercialization of its wood and bamboo shoots. However, the suitable area of this species is undergoing rapid changes due to climate change, resulting in species redistribution and potential losses for bamboo farmers. We utilized 209 presence records and 25 environmental variables from 1987 to 2012 to predict the potentially suitable habitats for Q. tumidinoda using MaxEnt (version 3.4.1), ArcGIS (version 10.8.2), and R (4.3.3). We rigorously screened the recorded data for reliability and accuracy through expert consultations and observer interviews. We performed pre-processing to select the variables with high contributions for modeling, and 11 variables were selected for the final modeling. Our findings reveal that the top three most influential variables associated with Q. tumidinoda’s distribution were the mean monthly potential evapotranspiration (Pet), annual range of air temperatures (Bio7), and mean diurnal air temperature range (Bio2), and the rates of contributions from 1987 to 2012 were 4.8333, 3.5833, and 1.7000. There was a southeastward shift and an elevation increase in the potentially suitable habitats for Q. tumidinoda. The area of potentially suitable habitats in the study region exhibited fluctuating growth, expanding from 3063.42 km² to 7054.38 km². The mean monthly potential evapotranspiration (Pet) emerged as a critical determinant shaping the distribution of potentially suitable habitats for Q. tumidinoda. Our study sheds light on the response of Q. tumidinoda to climate change, offering valuable insights for the development and management of plantation industries associated with this species. In the future, to enhance prediction accuracy, researchers could equally consider both organic and inorganic environmental variables. For better preservation of environment and development, Q. tumidinoda could be introduced into nature restoration projects in areas with a suitable habitat or as a commodity that participates in forest carbon sink trading.

Currently, natural restoration has been widely proposed as the primary method of ecological restoration and has been studied for a long time. However, research on how to quantify the progress of natural restoration in different climate conditions, especially using long-term succession monitoring data combined with habitat quality data across various succession stages, has been scarce. Our study aims to address this issue in Yunnan, southwest China. To quantify the progress of natural restoration under different climates in Yunnan, we introduced an index, the Natural Succession Index. Utilizing topography and meteorological data, we divided the study area into different climate sub-areas using the Two-stage clustering algorithm. We then combined 1703 sets of 30-year succession monitoring data, each with six observations taken at five-year intervals from 1987 to 2017, with habitat quality data from different succession stages (grassland, shrub, and forest) to quantify the Natural Succession Index. Yunnan province was divided into 14 sub-areas, namely C(I to II), M(I to III), W(I to IV), and H(I to IV), each possessing a unique environment. The indices in each sub-area were calculated, with the results showing a specific order: H-I (0.7812) > H-IV (0.7739) > W-I (0.6498) > M-III (0.6356) > H-III (0.6316) > M-II (0.5735) > W-III (0.5644) > W-IV (0.5571) > C-II (0.4778) > W-II (0.3980) > M-I (0.3624) > H-II (0.3375) > C-I (0.2943). The times for natural succession to reach the forest stage vary from 5 to 19 years, which aligns with the order of indices. The stand volumes of vegetation in the forest stage range from 5 m³ to 110 m³, with a higher Natural Succession Index value corresponding to a higher stand volume of vegetation. In the future, the index could be utilized to reallocate investments in natural restoration projects for better returns. Constant vigilance is required in the first five years following the implementation of restoration actions to avoid failure due to calculation errors.