基于MaxEnt模型的三裂叶豚草在新疆草地的潜在入侵区预测
DOI:
作者:
作者单位:

新疆农业大学

作者简介:

通讯作者:

中图分类号:

基金项目:

第三次新疆综合科学考察项目(2022xjkk0401)


Prediction of potential invasion areas of Ambrosia trifida in Xinjiang grasslands based on the MaxEnt model
Author:
Affiliation:

Xinjiang Agricultural University

Fund Project:

The Third Xinjiang Comprehensive Scientific Expedition Project.

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 文章评论
    摘要:

    新疆草地资源丰富,是畜牧业发展的重要支撑。三裂叶豚草(Ambrosia trifida)是我国重点管理的外来入侵物种,研究其潜在适生区分布,对入侵草地的早期监测与防控具有重要意义。基于388个物种分布点及32个环境因子,运用MaxEnt模型预测当前及未来(2030s,2050s,2070s)不同气候情景(SSP126,SSP245,SSP585)下三裂叶豚草在新疆不同草地类的潜在入侵区,结果表明:(1)模型AUC值均大于0.90,预测结果稳定;(2)影响其分布的关键环境因子依次为:温度季节性变化标准差(Bio4)、人类足迹(HF)、最干月降水量(Bio14)和最湿月降水量(Bio13);(3)当前条件下,三裂叶豚草适生区总面积为14.96×104 km2,约占新疆总面积的9.25 %,主要集中于新疆西北部地区;在草地类分布上,除高寒荒漠外,其余10个草地类均存在入侵风险,其潜在入侵总面积为98164.74 km2,约占新疆草地总面积的17.00 %,其中温性荒漠、山地草甸和温性草原的潜在入侵风险较高;(4)在未来条件下,三裂叶豚草的适生区面积略有减少,整体变化率介于0.51 %—15.06 %之间,仍主要集中分布于新疆西北部地区;且大部分区域保持稳定,收缩面积普遍超过扩增面积,其中,2070s-SSP585情景下(温室气体排放高)收缩率最高,表明高温室气体排放量的增加并未导致其迅速蔓延。分布上,未来三裂叶豚草涉及11个草地类,其潜在入侵总面积在91435.70—99075.31 km2之间波动,约占新疆草地总面积的15.83 %—17.16 %,潜在入侵分布主要集中于温性荒漠、山地草甸和温性草原,分布面积较广,且分布中心将分别向东南、东北方向迁移。

    Abstract:

    Xinjiang is rich in grassland resources, which serve as an important support for the development of animal husbandry. Ambrosia trifida is a key managed alien invasive species in China, and studying the distribution of its potential suitable areas is of great significance for the early monitoring, prevention and control of grassland invasion. Based on 388 species distribution points and 32 environmental factors, this study used the MaxEnt model to predict the potential invasion areas of Ambrosia trifida in different grassland types in Xinjiang under current and future (2030s, 2050s, 2070s) climate scenarios (SSP126, SSP245, SSP585). The results showed that: (1) The AUC values of the model were all greater than 0.90, indicating stable prediction results; (2) The key environmental factors affecting its distribution were, in order: standard deviation of temperature seasonality (Bio4), human footprint (HF), precipitation of the driest month (Bio14), and precipitation of the wettest month (Bio13); (3) Under current conditions, the total area of suitable habitats for Ambrosia trifida is 14.96×10? km2, accounting for approximately 9.25 % of Xinjiang’s total area, and is mainly concentrated in the northwestern part of Xinjiang. In terms of distribution across grassland types, except for alpine deserts, the other 10 grassland types are at risk of invasion, with a total potential invasion area of 98,164.74 km2, accounting for roughly 17.00 % of Xinjiang’s total grassland area. Among these, temperate deserts, mountain meadows, and temperate steppes face relatively high potential invasion risks; (4) In the future, the area of suitable habitats for Ambrosia trifida was predicted to decrease slightly, with an overall change rate ranging from 0.51 % to 15.06 %, and it was predicted to still be mainly concentrated in the northwestern part of Xinjiang. Moreover, most areas was predicted to remain stable, and the contracted area was predicted to generally exceed the expanded area. Among all scenarios, the contraction rate was highest under the 2070s-SSP585 scenario (high greenhouse gas emissions), indicating that increased greenhouse gas emissions does not lead to its rapid spread. In terms of distribution, Ambrosia trifida was expected to affect 11 grassland types in the future, with its total potential invasion area fluctuating between 91,435.70 km2 and 99,075.31 km2, accounting for approximately 15.83 % to 17.16 % of Xinjiang’s total grassland area. Its potential invasion distribution is mainly concentrated in temperate deserts, mountain meadows, and temperate steppes, covering a relatively wide area, and its distribution centers were projected to migrate to the southeast and northeast respectively.

    参考文献
    相似文献
    引证文献
引用本文

杜玟霖,靳瑰丽,隋晓青,赵文轩,沈秉娜,陈梦甜,李文雄,李超,郑鼎甲.基于MaxEnt模型的三裂叶豚草在新疆草地的潜在入侵区预测.生态学报,,(). http://dx. doi. org/[doi]

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数: