TROPICAL GEOGRAPHY ›› 2017, Vol. 37 ›› Issue (2): 218-225.doi: 10.13284/j.cnki.rddl.002926

Previous Articles     Next Articles

Prediction of Potential Geographic Distribution of Actinidia chinensis in China Based on Maximum Entropy Niche Model and ArcGIS

ZHANG Jie1a,b,AO Ziqiang1c,WU Yongming1a,YANG Chunyan1a,LI Min2   

  1. (1.a.Poyang Lake Research Center,Jiangxi Academy of Sciences;b.Key Lab of Poyang Lake,Jiangxi Academy of Sciences;c.Energy Research Institute,Jiangxi Academy of Sciences,Nanchang 330096,China;2.Nanchang First Polytechnic School,Nanchang 330013,China)
  • Online:2017-03-05 Published:2017-03-05


Actinidia chinensis, a large deciduous climbing vine, is native to China and listed as an endangered plant of the national second-grade protection. The origin of Actinidia chinensis is supposed to be the Yangtze River valley, and Actinidia chinensis is known as the widely introduced and cultivated kiwi fruit. The kiwi fruit has both edible and medicinal value. In recent years, because of rapid urbanization and ecological environment change, its wild resources were seriously damaged and its habitat fragmented, which need to be protected, reasonably developed and utilized. Mapping its current distribution is useful for conservation planning,and understanding the change in distribution impacted by climate change is important for mitigation of the impact from future climate change. In this study, 181 historical investigation records were used to predict the potential suitable distribution and current distribution space reconstruction of Chinese kiwi fruit based on ArcGIS and Maximum Entropy niche model. We modeled the potential geographical distribution of the kiwi fruit in China under current and future climate change, and analyzed the key factors determining such distribution areas in China. Results show that the main suitable areas for Chinese kiwi were current distributed between 23° N and 35° N latitude and in the subtropical hilly areas to the east of 102° E longitude, such as the Qinling-Dabashan Mountains, Yun-Gui plateau, Luoxiao Mountains, Nanling Mountains and the Wuyi Mountains. The Maximum Entropy niche model is evaluated to be highly reliable by Receiver operating characteristic (ROC).Training data and Testing data of AUC were both above 0.950, indicating a better forecast. Response curves created by Jackknife method displayed that mean temperature of coldest quarter, min temperature of coldest month, mean diurnal range and temperature seasonality, which are all related to low temperature and their variation range, were the dominant environmental variables that mainly contributed to prediction of suitable distribution potential. The total area of suitable region under current climate condition is 2.38×106 km2, accounting for 25.19% of China's total land area. Under moderate greenhouse gas emissions in 2050 (RCP4.5) scenario, the area of suitable region would change weakly, but the spatial distribution pattern would shift rapidly. Results showed that under the influence of climate warming in the future, the area of suitable region would increase and migrate northwards. These results would provide valuable reference for conservation of the wild resources of Actinidia chinensis as well as regional planting design, and might contribute to the establishment of climate change adaptation policies for considering various adaption options.

Key words: Actinidia chinensis, potential distribution, suitability, Maximum Entropy model, climate change