云南广西蒜头果适生区预测及环境影响因子

【目的】蒜头果(Malania oleifera)是一种兼有很高经济和生态价值的濒危植物。本研究旨在揭示云南、广西蒜头果的潜在适生区的空间分布格局,并且明确其主要环境影响因子,为蒜头果的保护与开发利用提供理论基础。【方法】通过野外调查、数字标本植物馆、全球生物多样性信息网络等方式获取了蒜头果样本136个。选取常见的20个主要环境因子作为参数,基于最大熵模型(MaxEnt)和ArcGIS地理信息系统平台构建了蒜头果潜在适生区预测模型,模拟蒜头果在云南和广西壮族自治区的潜在适生区。【结果】笔者构建的MaxEnt模型预测得到蒜头果的潜在适生区分布范围为104°~107°E及22°~26°N,预测适生区验证结果的受试者工作特征曲线下方的面积(AUC)均超过0.9。MaxEnt预测的蒜头果潜在适生区前4个环境影响因子及贡献率依次为:气温季节性变动系数因子(39.6%)、等温性因子(16.7%)、平均气温因子(13.7%)、气温年较差因子(11.5%)。【结论】云南文山州东南部以及广西西部是蒜头果的集中分布区,温度是影响蒜头果分布的主要因素,MaxEnt模型在蒜头果适生区预测中表现出极高的精度和可靠性。该研究将为蒜头果资源保护利用和人工繁育选址提供重要依据。

Suitable regions forecasting and environmental influencing factors of Malania oleifera in Yunnan and Guangxi

【Objective】 Malania oleifera is an endangered plant with high economic and ecological value. This study focuses on discovering the spatial distribution pattern of potential suitable areas of M. oleifera, and finding its main environmental affecting factors, laying a solid theory foundation for its conservation and utilization.【Method】 This study got 136 sampling points of M. oleifera by field investigations, specimens of the digital library, and the global biodiversity information network. Based on the ArcGIS geographic information system platform and the maximum entropy model (MaxEnt) with parameters of 20 common main environmental factors, the prediction model of M. oleifera potential suitable areas was built to simulate the M. oleifera distribution in Yunnan Province and Guangxi Zhuang Nationality Autonomous Prefecture (Guangxi).【Result】 The results showed that of M. oleifera is mainly distributed in the longitude of 104°-107°E and the latitude of 22°-26°N. The area under the curve (AUC) of potential suitable areas predicted by the MaxEnt model in this study were all over than 0.9. The top 4 environmental affecting factors and their contribution rates predicted by the MaxEnt model of potential suitable areas of M. oleifera were orderly listed: seasonal variation factors of temperature (contribution rate 39.6%), isothermal factors (contribution rate 16.7%), average daily temperature factors (contribution rate 13.7%), and annual temperature difference factors (contribution rate 11.5%).【Conclusion】 The southeast of Wenshan Zhuang and Miao Nationality Autonomous Prefecture of Yunnan and the west of Guangxi are the concentrated potential distribution suitable areas of M. oleifera, and temperature is the dominated affecting factor. Additionally, the MaxEnt model performs well for predicting potential suitable areas of M. oleifera both in accuracy and reliability. This study will provide an important basis for the conservation and utilization of M. oleifera resources, and an artificial breeding site selection.