气候变化下三裂叶豚草在新疆的潜在地理分布

明确有害入侵种在区域尺度上的潜在分布及其对气候变化的响应对入侵种的预警和具体防控意义重大。三裂叶豚草是全球公认的恶性入侵杂草，目前已经大面积入侵“一带一路”中亚枢纽—新疆伊犁河谷。为有效防控三裂叶豚草在新疆的扩散蔓延，本研究基于最大熵模型，预测了当前及未来两种气候情景（RCP4.5， RCP8.5）下2050s和2070s时期三裂叶豚草在新疆的潜在分布及变化趋势。结果表明：当前气候下三裂叶豚草在新疆的总适生面积达24.01万km²，约占全疆面积的14%，在RCP4.5情景下，至2050s、2070s时期将分别增至37.36万和39.23万km²；在RCP8.5情景下，将分别增至39.45万和42.94万km²。在未来两种气候情景下，随着时间的推移，潜在适生区总体均呈现向北增加转移的趋势，减少区域主要集中在准噶尔盆地。所有环境因子中，与降水相关的因子总贡献率为40.1%，与温度相关的因子总贡献率56.0%，其中最干月降水（36.2%）、温度季节性变化标准差（29.1%）是对三裂叶豚草分布贡献率较高的环境因子。耕地和建设用地是三裂叶豚草入侵风险最高的区域。建议管理的重点除了放在已预测到的适生区内，还应特别关注农田、草场、道路两侧这些人畜扰动大、水分充足的区域。

Potential geographical distribution of Ambrosia trifida in Xinjiang under climate change

Determining the potential distribution of invasive species and their responses to climate change on a regional scale is of great significance for the early warning， specific prevention， and control of invasive species. Ambrosia trifida （giant ragweed） is a globally recognized invasive weed. This plant is now widely distributed in the Yili Valley in Xinjiang which locaded in Central Asian hub of the“Belt and Road”. To predict the potential spread of A. trifida in Xinjiang， a Maxent niche model was used to project its potential distribution under the current climate and the climate in the 2050s and 2070s under two climate change scenarios （RCP4.5 and RCP8.5）. The results indicated that the suitable habitat area of giant ragweed in Xinjiang could extend to 24.01×10⁴ km²， accounting for 14% of the total area of Xinjiang under the current climate conditions. Under the RCP4.5 scenario， its suitable habitat area was predicted to increase to 37.36×10⁴ km² and 39.23×10⁴ km² by the 2050s and 2070s， respectively. Under the RCP8.5 scenario， its suitable habitat area was predicted to increase to 39.45×10⁴ km² and 42.94×10⁴ km² by the 2050s and 2070s， respectively. Under the two future climate scenarios， the potential suitable habitat area will tend to increase and shift northwards， and the decreasing areas will be mainly concentrated in the Junggar Basin. Among all the environmental factors， precipitation and temperature made the strongest contributions to distribution， with total contribution rates of 40.1% and 56.0%， respectively. The individual factors making the strongest contributions to giant ragweed distribution were precipitation in the driest month （36.2%） and temperature seasonality （29.1%）. Farmland and construction land were identified as the areas with the highest risk of giant ragweed invasion. On the basis of these results， we suggest that management strategies should focus not only on predicted habitats， but also on farmland， grassland， and roadsides where there is sufficient water and strong disturbance by humans or animals.