基于Maxent模型的不同气候变化情景下我国草地螟越冬区预测
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家国际科技合作专项(2014DFR31250);公益性行业(农业)科研专项(201303057,201403031);国家自然科学基金(31301656,31371947);北京市自然科学基金(6142017)


Maxent-based prediction of overwintering areas of Loxostege sticticalis in China under different climate change scenarios
Author:
Affiliation:

Fund 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:

    草地螟(Loxostege sticticalis L.)是我国华北、东北和西北地区农牧业生产的重大害虫,其每年暴发成灾的程度及造成的经济损失与越冬虫源基数或面积密切相关,但我国草地螟主要越冬区的变化规律至今尚未见报道。为了阐明全球气候变化条件下我国草地螟越冬区的变化规律,以1951-2000年我国草地螟越冬场所及面积为基础,结合政府间气候变化专门委员会(IPCC)第五次评估报告发布的BCC-CSM1.1气候模式数据,采用最大熵(Maxent)模型预测了4种气候变化(RCP 2.6, 4.5, 6.0和8.5)情景下2050s及2070s 我国草地螟越冬区的变化情况。研究结果:1)训练数据集和测试数据集的受试者工作特征曲线下的面积(AUC)分别为0.989和0.987,表明模型的模拟精度很好。2)经Jackknife方法检验,降雨量变化方差bio_15是最重要的变量,最湿月份降雨量bio_13次之,最后为10月份降水量prec_10及年温变化范围bio_7。这些环境变量对模型的贡献率均超过10%,是模型构建最重要的环境变量。3)在各种气候变化情景下,到2050s及2070s我国草地螟的越冬区面积和位置相对于当前都有不同程度的扩大和北移。其中高适宜越冬区面积为当前的1.41-2.94倍,其质心位置向北移动78.79-226.97 km。这些结果表明,我国未来草地螟越冬场所将会扩大和北移。

    Abstract:

    The beet webworm, Loxostege sticticalis L. (Lepidoptera: Crambidae) is a destructive insect pest of crops and fodder plants in northern China. The degree of annual outbreaks of the larval population during the breeding season is highly related to the size and location of the overwintering area for hibernating larvae, but how this relationship will change in the future remains unknown. Therefore, variations in the size and location of the overwintering area of L. sticticalis were studied to reveal changes under various scenarios of global climate change. This was accomplished using overwintering data of L. sticticalis in China from 1951 to 2000, combined with the application of the Maxent model and the BCC-CSM1.1 climate change modeling data under four scenarios of RCP, 2.6, 4.5, 6.0 and 8.5, released by IPCC AR5. Our results showed that the AUC (area under the receiver operating characteristic curve) value for the training and test data were 0.989 and 0.987, respectively, indicating the data fit the model very well. In addition, the most important variable in the model as determined by the jackknife test was the variance in the precipitation coefficient (bio_15), followed by precipitation during the wettest month (bio_13), precipitation in October (bio_10), and annual temperature range (bio_7). Their individual contribution to the model was greater than 10% each. Finally, the centroid of the highly suitable overwintering location would be shifted 104.93-207.47 km and 78.79-226.97 km northward, whereas the size of the area would expanded to 1.41-2.00 and 1.94-2.94 fold by the 2050s and 2070s, respectively, compared to those at the present time.

    参考文献
    相似文献
    引证文献
引用本文

唐继洪,程云霞,罗礼智,张蕾,江幸福.基于Maxent模型的不同气候变化情景下我国草地螟越冬区预测.生态学报,2017,37(14):4852~4863

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: