基于MaxEnt模型预测冬虫夏草在青海省的潜在适生区

Predicting the potential habitat of Chinese Cordyceps in Qinghai Province based on MaxEnt model

摘要:

【目的】 预测冬虫夏草在青海省的潜在适生区分布，对冬虫夏草资源可持续利用及稳定产区农牧民经济收入具有重要意义。【方法】 利用62个冬虫夏草地理分布点和22个环境变量数据，采用最大熵模型MaxEnt与地理信息系统ArcGIS模拟冬虫夏草的地理分布，分析环境变量对冬虫夏草分布的影响，并划分适生等级，分析适生区环境特征。【结果】 AUC（曲线下面积）值高于0.9，模型具有较高的准确度，适宜于冬虫夏草的生境适宜性预测。利用Jackknife刀切法计算得到影响冬虫夏草分布的主要环境变量为年降水量（613~727 mm）、年平均气温（-2~3 ℃）、海拔（3 800~4 500 m）和最冷月最低温 （-13~ -11 ℃），上述环境变量的累计贡献率为88.5%。适生区分析结果显示，冬虫夏草主要集中于玉树藏族自治州的玉树市、杂多县、称多县、囊谦县、曲麻莱县，果洛藏族自治州的玛多县、玛沁县、甘德县、达日县、久治县、班玛县，海北藏族自治州的门源县，海西藏族自治州的都兰县。【结论】 MaxEnt模型能很好地模拟冬虫夏草的地理分布，本次预测的潜在适生区分布范围大于已知分布区。

Abstract:

【Objective】 With human activities and global climate change，the provinces Chinese Cordyceps resources are facing a risk reduction. It is important to predict the potential distribution of Chinese Cordyceps in Qinghai province for the sustainable use of Chinese Cordyceps resources and the stabilization of the economic income of farmers and herdsmen.【Method】 Based on 62 geographical distribution points of Chinese Cordyceps and 22 environmental variables，the maximum entropy model MaxEnt and the geographic information system ArcGIS were used to simulate the geographical distribution of Chinese Cordyceps，analyze the impact of environmental variables on the distribution of Chinese Cordyceps，and divide the suitable growth grade and analyze the environmental characteristics of the suitable areas.【Result】 The results showed that the AUC （area under the curve） value was higher than 0. 9 and the model had high accuracy and was suitable for habitat suitability prediction of Chinese Cordyceps. The main environmental variables affecting the distribution of Chinese Cordyceps were calculated using the Jackknife cut method. The calculation demonstrated that the main environmental variables affecting the distribution of cordyceps sinensis were annual precipitation （613~727 mm），annual average temperature （-2~3 ℃），altitude （3 800~4 500 m） and the lowest temperature in the coldest month （-13~-11 ℃）. The cumulative contribution rate of the above environmental variables is 88. 5%. The analysis of suitable areas showed that Chinese Cordyceps was mainly concentrated in Yushu City，Zadoi County，Chindu County，Nangqen County and Qumarleb County in Yushu Tibetan Autonomous Prefecture，Madoi County，Maqen County，Gade County，Darlag County，Jiuzhi County and Banma County in Guoluo Tibetan Autonomous Prefecture，Menyuan County in Haibei Tibetan Autonomous Prefecture，and Dulan County in Haixi Tibetan Autonomous Prefecture.【Conclusion】 The results show that the MaxEnt model can well simulate the geographical distribution of Chinese Cordyceps，clarify the influence of environmental variables on the distribution of Chinese Cordyceps，and provide reference for the conservation of Chinese Cordyceps resources.