基于新型最大熵模型预测刺槐叶瘿蚊(双翅目:瘿蚊科)在中国的适生区
|
摘要
【目的】基于对刺槐叶瘿蚊在全国的普查情况,利用最大熵模型MaxEnt软件的互补双对数输出方式对刺槐叶瘿蚊在中国当前和未来(2050年)的适生区进行预测,为林业和海关检疫部门对刺槐叶瘿蚊当前与未来的防控与检疫工作提供重要参考依据。【方法】使用MaxEnt、ArcGIS、R软件对刺槐叶瘿蚊危害点,气候图层,模型参数这3方面进行科学的优化选择,确保模型的科学性、有效性。当前气候适生区的预测使用WorldClim网站全球气候数据Version 1.4,未来数据则采用通用气候系统模型CCSM4下3种外排情景(RCP26、RCP45、RCP85)。【结果】最终确定52个危害点,7个主导气候图层,运用互补双对数输出方式对适生区进行预测。模拟结果的测试遗漏率与理论遗漏率基本吻合,ROC曲线即AUC值为0.919,标准差为0.023,表明所使用的数据无空间自相关,构建的模型达到"极好"的标准。通过刀切图分析,对刺槐叶瘿蚊分布影响最大的3个气候图层分别为Bio1(年平均气温)、Bio12(年降水量)、Bio5(最热月的最高温度)。对当前气候刺槐叶瘿蚊适生区进行划分,刺槐叶瘿蚊在中国的适生范围为22.08°—48.42°N,39.39°—135.06°E,达国土面积的31.90%。除西藏、青海、海南、台湾4省区外,其余省份均包含其适生区,其高度适生区以西南(四川、重庆)和华北(北京、天津、河北、山东、陕西)为主。对未来(2050年)适生区的预测,3种外排情景RCP26、RCP45、RCP85的总适生区均比当前气候的总适生范围大,以高度、中度适生区面积的增大为主,新疆和我国北部区域面积显著扩增。RCP85情景下的刺槐叶瘿蚊适生区面积最大,达国土面积的39.71%,比当前预测的多出75万km~2。【结论】结合实际调查情况,新型MaxEnt模型预测结果可信度高,阐明影响刺槐叶瘿蚊分布的主导气候因子,预测出刺槐叶瘿蚊当前与未来的分布范围及适生程度情况,对刺槐叶瘿蚊的防控具有重要意义。
【Objective】Based on the national investigation of Obolodiplosis robiniae, the current and future(2050) sustainable habitats of O. robiniae in China were predicted by using the complementary log-log(Cloglog) of maximum entropy model. This study would provide an important reference for forestry and quarantine authorities to control and quarantine of O. robiniae at present and in the future. 【Method】MaxEnt, ArcGIS, and R software were used to scientifically optimize the three aspects of the O. robiniae distribution points, bio-climatic variables and model parameter setting, to ensure the scientificity and validity of the model. The current bio-climatic variables for the suitable area were predicted by using WorldClim Website Global Weather Database Version 1.4, with5 min of spatial resolution. The future data were derived from three representative concentration pathways RCP26, RCP45 and RCP85 under CCSM4.【Result】A total of 52 distribution points and 7 bio-climatic variables were finally determined. The complementary log-log output format was used to predict the potential geographical distribution. The test omission rate of simulation results is basically consistent with the theoretical omission rate. The ROC(receiver operating characteristic curve), which has an AUC(area under the ROC curve) of 0.919 and a standard deviation of 0.023, indicating that the test and training data have no spatial autocorrelation and the constructed model meets the "excellent" standard. Through Jackknife test analysis, the three bio-climatic variables that have the greatest influence on O. robiniae distribution are Bio1(Annual mean temperature), Bio12(Annual precipitation) and Bio5(Max temperature of warmest month). For the current classification of the potential geographical distribution for O. robiniae, the suitable range is in the range of 22.08°—48.42°N, 39.39°—135.06°E, accounting for 31.90% of the national area. With the exception of Tibet, Qinghai, Hainan, and Taiwan province, the other provinces have their own potential geographical distribution, and the highly potential geographical distribution regions are mainly in the southwest(Sichuan, Chongqing) and Northern China(Beijing, Tianjin, Hebei, Shandong, and Shaanxi). For the prediction of the future(2050), the total susceptibility zones of the three representative concentration pathways RCP26, RCP45, and RCP85 are larger than the total habitats in the current climate, especially for high and moderate suitable habitats. The areas in Xinjiang and northern China will expand significantly. The risk area under the RCP85 scenario is the largest, reaching to 39.71% of the national area, 750 000 km~2 more than the current prediction. 【Conclusion】Combined with the actual investigation, the novel MaxEnt model has high reliability to clarify the dominant climatic factors affecting the distribution of O. robiniae, and predict the distribution and fitness of O. robiniae current and future(2050), which is of important value for the prevention and control of O. robiniae.
【Objective】Based on the national investigation of Obolodiplosis robiniae, the current and future(2050) sustainable habitats of O. robiniae in China were predicted by using the complementary log-log(Cloglog) of maximum entropy model. This study would provide an important reference for forestry and quarantine authorities to control and quarantine of O. robiniae at present and in the future. 【Method】MaxEnt, ArcGIS, and R software were used to scientifically optimize the three aspects of the O. robiniae distribution points, bio-climatic variables and model parameter setting, to ensure the scientificity and validity of the model. The current bio-climatic variables for the suitable area were predicted by using WorldClim Website Global Weather Database Version 1.4, with5 min of spatial resolution. The future data were derived from three representative concentration pathways RCP26, RCP45 and RCP85 under CCSM4.【Result】A total of 52 distribution points and 7 bio-climatic variables were finally determined. The complementary log-log output format was used to predict the potential geographical distribution. The test omission rate of simulation results is basically consistent with the theoretical omission rate. The ROC(receiver operating characteristic curve), which has an AUC(area under the ROC curve) of 0.919 and a standard deviation of 0.023, indicating that the test and training data have no spatial autocorrelation and the constructed model meets the "excellent" standard. Through Jackknife test analysis, the three bio-climatic variables that have the greatest influence on O. robiniae distribution are Bio1(Annual mean temperature), Bio12(Annual precipitation) and Bio5(Max temperature of warmest month). For the current classification of the potential geographical distribution for O. robiniae, the suitable range is in the range of 22.08°—48.42°N, 39.39°—135.06°E, accounting for 31.90% of the national area. With the exception of Tibet, Qinghai, Hainan, and Taiwan province, the other provinces have their own potential geographical distribution, and the highly potential geographical distribution regions are mainly in the southwest(Sichuan, Chongqing) and Northern China(Beijing, Tianjin, Hebei, Shandong, and Shaanxi). For the prediction of the future(2050), the total susceptibility zones of the three representative concentration pathways RCP26, RCP45, and RCP85 are larger than the total habitats in the current climate, especially for high and moderate suitable habitats. The areas in Xinjiang and northern China will expand significantly. The risk area under the RCP85 scenario is the largest, reaching to 39.71% of the national area, 750 000 km~2 more than the current prediction. 【Conclusion】Combined with the actual investigation, the novel MaxEnt model has high reliability to clarify the dominant climatic factors affecting the distribution of O. robiniae, and predict the distribution and fitness of O. robiniae current and future(2050), which is of important value for the prevention and control of O. robiniae.
引文
韩阳阳, 王焱, 项杨, 等. 2015. 基于Maxent生态位模型的松材线虫在中国的适生区预测分析. 南京林业大学学报:自然科学版, 39(1): 6-10.(Han Y Y, Wang Y, Xiang Y, et al. 2015.Prediction of potential distribution of Bursaphelenchus xylophilus in China based on Maxent ecological niche mode. Journal of Nanjing Forestry University:Natural Science Edition, 39(1):6-10. [in Chinese])
雷军成, 徐海根. 2010. 基于Maxent的加拿大一枝黄花在中国的潜在分布区预测. 生态与农村环境学报, 26(2): 137-141.(Lei C J,Xu H G. 2010. MaxEnt-based prediction of potential distribution of Solidago canadensis in China. Journal of Ecology and Rural Environment,26(2): 137-141.[in Chinese])
林保民, 乔秀荣, 徐登华, 等. 2007. 秦皇岛市刺槐叶瘿蚊发生情况初报. 林业科技, 32(4): 30-31.(Lin B M, Qiao X R, Xu D H, et al. 2007.Investigation and control on Robinia pseudoacacia in Qinhuangdao City. Forestry Science & Technology,32(4):30-31. [in Chinese])
穆希凤, 孙静双, 卢文锋, 等. 2010. 北京地区刺槐叶瘿蚊生物学特性及防治. 中国森林病虫, 29(5): 15-18.(Mu X F, Sun J S, Lu W F, et al. 2010.Bionomics and control of Obolodiplosis robiniae in Beijing. Forest Pest And Disease,29(5):15-18. [in Chinese])
尚兴朴, 姚艳霞, 赵文霞. 2015. 外来有害生物刺槐叶瘿蚊在中国的地理分布. 中国森林病虫, 34(4): 33-36.(Shang X P, Yao Y X, Zhao W X. 2015.Geographic distribution of an invasive insect pest, Obolodiplosis robiniae (Diptera: Cecidomyiidae) in China. Forest Pest and Disease,34(4):33-36. [in Chinese])
王梦琳, 范靖宇, 李敏, 等. 2017. 入侵害虫蔗扁蛾在我国的潜在分布区. 生物安全学报, 26(2): 129-133.(Wang M L, Fan J Y, Li M, et al. 2017.Potential geographical distribution of the introduced banana moth,Opogona sacchari (Lepidoptera: Tineidae) in China. Journal of Biosafety,26(2):129-133. [in Chinese])
王茹琳, 高晓清, 石朝鹏. 2018. 危险性入侵昆虫梣粉虱对中国的风险分析. 中国农学通报, 34(1): 129-133.(Wang R L, Gao X Q, Shi C P.2018.Risk analysis of Siphoninus phillyreae (Haliday): A dangerous exotic pest in China. Chinese Agricultural Science Bulletin, 34(1):129-133. [in Chinese])
王运生, 谢丙炎, 万方浩, 等. 2007. Roc曲线分析在评价入侵物种分布模型中的应用. 生物多样性, 15(4): 365-372.(Wang Y S, Xie B Y, Wan F H, et al. 2007. Application of ROC curve analysis in evaluating the performance of alien species’ potential distribution models. Biodiversity Science, 15(4):365-372. [in Chinese])
奚国华, 齐国君, 王书平, 等. 2017. 基于Maxent的截获秀粉蚧潜在地理分布预测. 植物检疫, 31(2): 30-34.(Xi G H, Qi G J, Wang S P, et al. 2017.Potential geographic distribution of Paracoccus interceptus based on MaxEnt. Plant Quarantine,31(2):30-34. [in Chinese])
杨忠岐, 乔秀荣, 卜文俊, 等. 2006. 我国新发现一种重要外来入侵害虫——刺槐叶瘿蚊. 昆虫学报, 49(6): 1050-1053.(Yang Z Q, Qiao X R, Pu W J, et al. 2006. First discovery of an important invasive insect pest, Obolodiplosis robiniae(Diptera: Cecidomyiidae)in China. Acta Entomologica Sinica,49(6):1050-1053. [in Chinese])
张东风, 路常宽, 王晓勤, 等. 2009. 刺槐叶瘿蚊在中国的危险性评估. 生态学报, 29(4): 2155-2161.(Zhang D F, Lu C K, Wang X Q, et al. 2009.Potential risk assessment of Obolodiplosis robiniae(Haldemann)in China. Acta Ecologica Sinica,29(4):2155-2161. [in Chinese])
徐秀琴, 杨敏生. 2006. 刺槐资源的利用现状. 河北林业科技,(S1): 54-57.(Xu X Q, Yang M S.2006. Status of utilization of Robinia pseudoacacia resources. Journal of Hebei Forestry,(S1): 54-57. [in Chinese])
朱耿平, 刘强, 高玉葆. 2014. 提高生态位模型转移能力来模拟入侵物种的潜在分布. 生物多样性, 22(2): 223-230.(Zhu G P, Liu Q, Gao Y B.2014.Improving ecological niche model transferability to predict the potential distribution of invasive exotic species. Biodiversity Science, 22(2):223-230. [in Chinese])
Anderson R P, Gonzalez I J. 2011. Species-specific tuning increases robustness to sampling bias in models of species distributions: An implementation with Maxent. Ecol Model, 222(15): 2796-2811.
Garrote G, Fernandez-López J, López G, et al. 2018. Prediction of Iberian Lynx road-mortality in southern Spain: A new approach using the Maxent Algorithm. Animal Biodiversity & Conservation, 412(2):217-225.
Guevara L, Gerstner B E, Kass J M, et al. 2017. Toward ecologically realistic predictions of species distributions: A cross-time example from tropical montane cloud forests. Global Change Biology,24(4):1511-1522.
Guisan A, Jr T C E, Hastie T. 2002. Generalized linear and generalized additive models in studies of species distributions: Setting the scene. Ecol Model, 157(2): 89-100.
Haldeman S. 1847. Description of several new and interesting animals. American Journal of Agriculture & Science, 6:191-194.
Harris R M B, Grose M R, Lee G, et al. 2014. Climate Projections for Ecologists. Wiley Interdisciplinary Reviews Climate Change, 5(5): 621-637.
Kalboussi M, Achour H. 2018. Modelling the spatial distribution of snake species in northwestern Tunisia using Maximum Entropy (Maxent) and Geographic Information System (Gis). Journal of Forestry Research,29(1):233-245.
Kramer-Schadt S, Niedballa J, Pilgrim J D, et al. 2013. The importance of correcting for sampling bias in maxent species distribution models. Diversity & Distributions, 19(11): 1366-1379.
Kumar R S, Kala R H, Kumar G S, et al. 2017. Predicting the impact of climate change on the distribution of two threatened Himalayan Medicinal Plants of Liliaceae in Nepal. J Mt Sci-Engl, 14(3): 558-570.
Merow C, Smith M J, Silander J A. 2013. A practical guide to maxent for modeling species’ distributions: What it does, and why inputs and settings matter. Ecography, 36(10): 1058-1069.
Miller R G. 1968. Jackknifing variances. Annals of Mathematical Statistics, 39(2): 567-582.
Phillips S J. 2005. A Brief Tutorial On Maxent.[2018-05-16].http://biodiversity in formatics.amnh.org/open_source/maxent/Maxen_tutorial2017.pdf.
Phillips S J, Anderson R P, Dudík M, et al. 2017. Opening the black box: An open-source release of Maxent. Ecography, 40(7):887-893.
Phillips S J, Anderson R P, Schapire R E. 2006. Maximum entropy modeling of species geographic distributions. Ecol Model, 190(3): 231-259.
Phillips S J, Dudík M. 2008. Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography, 31(2): 161-175.
Phillips S J, Schapire R E. 2004. A maximum entropy approach to species distribution modeling. Procceedings of the International Conference on Machine Learning,Banff,canada,89-90.
Radosavljevic A, Anderson R P, Araújo M. 2014. Making better Maxent models of species distributions: Complexity, overfitting and evaluation. J Biogeogr, 41(4): 629-643.
Shcheglovitova M, Anderson R P. 2013. Estimating optimal complexity for ecological niche models: a jackknife approach for species with small sample sizes. Ecol Model, 269(1771): 9-17.
Sillero N. 2011. What does ecological modelling model? a proposed classification of ecological niche models based on their underlying methods. Ecol Model, 222(8): 1343-1346.
Verbruggen H, Tyberghein L, Belton G S, et al. 2013. Improving transferability of introduced species’ distribution models: new tools to forecast the spread of a highly invasive seaweed. Plos One, 8(6): e68337.
Yang X Q, Kushwaha S P S, Saran S, et al. 2013. Maxent modeling for predicting the potential distribution of medicinal plant, Justicia adhatoda L. in Lesser Himalayan Foothills. Ecol Eng, 51(1): 83-87.
雷军成, 徐海根. 2010. 基于Maxent的加拿大一枝黄花在中国的潜在分布区预测. 生态与农村环境学报, 26(2): 137-141.(Lei C J,Xu H G. 2010. MaxEnt-based prediction of potential distribution of Solidago canadensis in China. Journal of Ecology and Rural Environment,26(2): 137-141.[in Chinese])
林保民, 乔秀荣, 徐登华, 等. 2007. 秦皇岛市刺槐叶瘿蚊发生情况初报. 林业科技, 32(4): 30-31.(Lin B M, Qiao X R, Xu D H, et al. 2007.Investigation and control on Robinia pseudoacacia in Qinhuangdao City. Forestry Science & Technology,32(4):30-31. [in Chinese])
穆希凤, 孙静双, 卢文锋, 等. 2010. 北京地区刺槐叶瘿蚊生物学特性及防治. 中国森林病虫, 29(5): 15-18.(Mu X F, Sun J S, Lu W F, et al. 2010.Bionomics and control of Obolodiplosis robiniae in Beijing. Forest Pest And Disease,29(5):15-18. [in Chinese])
尚兴朴, 姚艳霞, 赵文霞. 2015. 外来有害生物刺槐叶瘿蚊在中国的地理分布. 中国森林病虫, 34(4): 33-36.(Shang X P, Yao Y X, Zhao W X. 2015.Geographic distribution of an invasive insect pest, Obolodiplosis robiniae (Diptera: Cecidomyiidae) in China. Forest Pest and Disease,34(4):33-36. [in Chinese])
王梦琳, 范靖宇, 李敏, 等. 2017. 入侵害虫蔗扁蛾在我国的潜在分布区. 生物安全学报, 26(2): 129-133.(Wang M L, Fan J Y, Li M, et al. 2017.Potential geographical distribution of the introduced banana moth,Opogona sacchari (Lepidoptera: Tineidae) in China. Journal of Biosafety,26(2):129-133. [in Chinese])
王茹琳, 高晓清, 石朝鹏. 2018. 危险性入侵昆虫梣粉虱对中国的风险分析. 中国农学通报, 34(1): 129-133.(Wang R L, Gao X Q, Shi C P.2018.Risk analysis of Siphoninus phillyreae (Haliday): A dangerous exotic pest in China. Chinese Agricultural Science Bulletin, 34(1):129-133. [in Chinese])
王运生, 谢丙炎, 万方浩, 等. 2007. Roc曲线分析在评价入侵物种分布模型中的应用. 生物多样性, 15(4): 365-372.(Wang Y S, Xie B Y, Wan F H, et al. 2007. Application of ROC curve analysis in evaluating the performance of alien species’ potential distribution models. Biodiversity Science, 15(4):365-372. [in Chinese])
奚国华, 齐国君, 王书平, 等. 2017. 基于Maxent的截获秀粉蚧潜在地理分布预测. 植物检疫, 31(2): 30-34.(Xi G H, Qi G J, Wang S P, et al. 2017.Potential geographic distribution of Paracoccus interceptus based on MaxEnt. Plant Quarantine,31(2):30-34. [in Chinese])
杨忠岐, 乔秀荣, 卜文俊, 等. 2006. 我国新发现一种重要外来入侵害虫——刺槐叶瘿蚊. 昆虫学报, 49(6): 1050-1053.(Yang Z Q, Qiao X R, Pu W J, et al. 2006. First discovery of an important invasive insect pest, Obolodiplosis robiniae(Diptera: Cecidomyiidae)in China. Acta Entomologica Sinica,49(6):1050-1053. [in Chinese])
张东风, 路常宽, 王晓勤, 等. 2009. 刺槐叶瘿蚊在中国的危险性评估. 生态学报, 29(4): 2155-2161.(Zhang D F, Lu C K, Wang X Q, et al. 2009.Potential risk assessment of Obolodiplosis robiniae(Haldemann)in China. Acta Ecologica Sinica,29(4):2155-2161. [in Chinese])
徐秀琴, 杨敏生. 2006. 刺槐资源的利用现状. 河北林业科技,(S1): 54-57.(Xu X Q, Yang M S.2006. Status of utilization of Robinia pseudoacacia resources. Journal of Hebei Forestry,(S1): 54-57. [in Chinese])
朱耿平, 刘强, 高玉葆. 2014. 提高生态位模型转移能力来模拟入侵物种的潜在分布. 生物多样性, 22(2): 223-230.(Zhu G P, Liu Q, Gao Y B.2014.Improving ecological niche model transferability to predict the potential distribution of invasive exotic species. Biodiversity Science, 22(2):223-230. [in Chinese])
Anderson R P, Gonzalez I J. 2011. Species-specific tuning increases robustness to sampling bias in models of species distributions: An implementation with Maxent. Ecol Model, 222(15): 2796-2811.
Garrote G, Fernandez-López J, López G, et al. 2018. Prediction of Iberian Lynx road-mortality in southern Spain: A new approach using the Maxent Algorithm. Animal Biodiversity & Conservation, 412(2):217-225.
Guevara L, Gerstner B E, Kass J M, et al. 2017. Toward ecologically realistic predictions of species distributions: A cross-time example from tropical montane cloud forests. Global Change Biology,24(4):1511-1522.
Guisan A, Jr T C E, Hastie T. 2002. Generalized linear and generalized additive models in studies of species distributions: Setting the scene. Ecol Model, 157(2): 89-100.
Haldeman S. 1847. Description of several new and interesting animals. American Journal of Agriculture & Science, 6:191-194.
Harris R M B, Grose M R, Lee G, et al. 2014. Climate Projections for Ecologists. Wiley Interdisciplinary Reviews Climate Change, 5(5): 621-637.
Kalboussi M, Achour H. 2018. Modelling the spatial distribution of snake species in northwestern Tunisia using Maximum Entropy (Maxent) and Geographic Information System (Gis). Journal of Forestry Research,29(1):233-245.
Kramer-Schadt S, Niedballa J, Pilgrim J D, et al. 2013. The importance of correcting for sampling bias in maxent species distribution models. Diversity & Distributions, 19(11): 1366-1379.
Kumar R S, Kala R H, Kumar G S, et al. 2017. Predicting the impact of climate change on the distribution of two threatened Himalayan Medicinal Plants of Liliaceae in Nepal. J Mt Sci-Engl, 14(3): 558-570.
Merow C, Smith M J, Silander J A. 2013. A practical guide to maxent for modeling species’ distributions: What it does, and why inputs and settings matter. Ecography, 36(10): 1058-1069.
Miller R G. 1968. Jackknifing variances. Annals of Mathematical Statistics, 39(2): 567-582.
Phillips S J. 2005. A Brief Tutorial On Maxent.[2018-05-16].http://biodiversity in formatics.amnh.org/open_source/maxent/Maxen_tutorial2017.pdf.
Phillips S J, Anderson R P, Dudík M, et al. 2017. Opening the black box: An open-source release of Maxent. Ecography, 40(7):887-893.
Phillips S J, Anderson R P, Schapire R E. 2006. Maximum entropy modeling of species geographic distributions. Ecol Model, 190(3): 231-259.
Phillips S J, Dudík M. 2008. Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography, 31(2): 161-175.
Phillips S J, Schapire R E. 2004. A maximum entropy approach to species distribution modeling. Procceedings of the International Conference on Machine Learning,Banff,canada,89-90.
Radosavljevic A, Anderson R P, Araújo M. 2014. Making better Maxent models of species distributions: Complexity, overfitting and evaluation. J Biogeogr, 41(4): 629-643.
Shcheglovitova M, Anderson R P. 2013. Estimating optimal complexity for ecological niche models: a jackknife approach for species with small sample sizes. Ecol Model, 269(1771): 9-17.
Sillero N. 2011. What does ecological modelling model? a proposed classification of ecological niche models based on their underlying methods. Ecol Model, 222(8): 1343-1346.
Verbruggen H, Tyberghein L, Belton G S, et al. 2013. Improving transferability of introduced species’ distribution models: new tools to forecast the spread of a highly invasive seaweed. Plos One, 8(6): e68337.
Yang X Q, Kushwaha S P S, Saran S, et al. 2013. Maxent modeling for predicting the potential distribution of medicinal plant, Justicia adhatoda L. in Lesser Himalayan Foothills. Ecol Eng, 51(1): 83-87.