基于地理探测器的浑善达克沙地植被变化定量归因

开展区域植被覆盖变化及驱动机制研究具有重要的意义。选取浑善达克沙地为研究区，基于地理探测器模型，定量研究了7个驱动因子对区域植被变化的单独影响及交互影响。结果表明：（1）2000—2018年，浑善达克沙地68.83%的区域植被覆盖在增加，其中显著增加的区域集中在研究区的中部和南部，即阿巴嘎旗、锡林浩特市、正蓝旗、正镶白旗和多伦县。（2）相对湿度和降水量是影响区域植被覆盖的主要自然因素，解释力分别达到了56.6%和49.5%。累计造林面积密度和年末牲畜头数密度是影响区域植被覆盖的主要人类因素，解释力分别达到了49.1%和46.4%，表明生态工程的实施对区域植被覆盖变化产生了重要的影响。（3）不同因子之间的交互作用决定值q均大于每个单独因子的决定值q，表明驱动因子的两两交互作用均会增加对区域植被变化的解释力。其中，相对湿度和人类活动因子（累计造林面积密度、年末牲畜头数密度和人口密度），降水量和年末牲畜头数密度均以双协同为主，解释力均超过了60%。而平均气温和年末牲畜头数密度以非线性协同为主，解释力也达到了60%以上。

Vegetation dynamics and its driving force in Otindag Sandy Land based on Geodetector

It is of vital significance to study the vegetation cover dynamics and its driving mechanism. Therefore， this research was carried out in Otindag Sandy Land （OSL） area and the individual and interactive effects of seven driving factors on regional vegetation changes were studied using the Geodetector model. The results showed that： （1） The vegetation cover increased in 68.83% area of the OSL from 2000 to 2018， and the significant regions were mainly in the middle and south of the study area （i.e. Abag banner， Xilinhot， Zhenglan banner， Zhengxiangbai Banner and Duolun County）. （2） The relative humidity （RH） and precipitation were the main natural factors affecting the spatial distribution of vegetation cover in the OSL from 2000 to 2018， with explanatory powers of 56.6% and 49.5%， respectively. While the density of total afforestation area and the density of livestock were the main human activities factors， with explanatory powers of 49.1% and 46.4%， respectively. These results indicated that the implementation of ecological engineering had important effects on regional vegetation cover dynamics. （3） The interaction between natural factors and human activities had a stronger impact on vegetation changes than single factors. The interaction between RH and human activities and the interaction between precipitation and the density of livestock were dominated by double synergy. The explanatory power of these interaction impacts were over 60%. However， the interaction between temperature and the density count of large livestock at the end of the year was dominated by non-linear synergy， the explanatory power also reached over 60%.