文献类型：期刊文献

中文题名：2000~2018年京津风沙源沙化土地时空动态与归因分析

英文题名：Dynamic and attribution analysis of sandy lands in the Beijing-Tianjin sandstorm source region during 2000–2018

作者：李晓松[1,2] 张磊[1,2] 姬翠翠[3] 赵立成[1,2] 杨子玉[1,2] 吴波[4]

第一作者：李晓松

通信作者：Li, XS[1];Li, XS[2]

机构：[1]可持续发展大数据国际研究中心,北京100094;[2]中国科学院空天信息创新研究院,数字地球重点实验室,北京100094;[3]重庆交通大学智慧城市学院,重庆400074;[4]中国林业科学研究院生态保护与修复研究所,荒漠生态系统与全球变化国家林业和草原局重点实验室,北京100091

年份：2023

卷号：68

期号：11

起止页码：1343-1355

中文期刊名：科学通报

外文期刊名：Chinese Science Bulletin

收录：CSTPCD;;EI(收录号：20231814046119);Scopus;WOS:【ESCI(收录号:WOS:000975776900007)】；北大核心:【北大核心2020】；CSCD:【CSCD2023_2024】；

基金：国家重点研发计划(2016YFC0500806)资助。

语种：中文

中文关键词：京津风沙源;沙化土地;遥感监测;驱动因素;沙地开垦

外文关键词：Beijing-Tianjin sandstorm source;sandy land;remote sensing monitoring;driving factors;sandy land reclamation

分类号：P931.3;X171.4

摘要：沙地土地变化信息是京津风沙源治理工程成效评估核心指标之一.为深入调查京津风沙源沙化土地变化状况及驱动因素,本研究首先利用Landsat系列影像,构建了基于面向对象、决策树与变化检测的沙化土地动态监测体系,实现了京津风沙源2000、2010、2015及2018年4期沙化土地分布制图.在此基础上,开展了京津风沙源沙化土地时空动态变化及驱动因素分析.结果表明,(1)京津风沙源工程区沙化土地持续减少,2000~2018年间年均减少323.94 km^(2),2015年后减少速率明显加快;(2)沙化土地总体向沙化程度降低方向转换,流动沙地减少和沙化耕地增加最为明显,相对变化比例分别为18.52%和47.47%,其中流动沙地变化中有56.89%转换为半固定沙地,固定沙地变化中有27.57%转为沙化耕地,较为突出;(3)人类活动及其与自然因素共同作用是沙化土地变化的主要驱动因素,占比分别为52.49%与29.74%,自然因素主导的变化不到1%,京津风沙源治理工程在沙化土地治理上成效明显;(4)沙化土地明显减少的同时,流动沙地仍有较大面积分布,仍有一定比例的退化发生,也存在大量的沙地开垦现象,未来应予以高度关注.
China launched the“Beijing-Tianjin Sandstorm Source Control Project”(BTSSCP)in 2000 to alleviate the negative effects of sandstorms in Beijing,Tianjin,and their surrounding areas.Restoration of sandy lands was a key priority of this project.Since the project would end in 2022,assessing the spatiotemporal changes of sandy lands and their main driving factors since 2000 is of great significance for evaluating the benefits of the project and guiding future scientific planning and governance.In this study,we employed object-oriented-based image segmentation,decision tree classification,and change vector analysis to establish a sandy land monitoring framework based on a series of Landsat datasets,and obtained sandy land distribution maps for 2000,2010,2015,and 2018.Moreover,the dynamics of the different sandy land types in different periods were studied and analyzed,and their characteristics were summarized.Finally,the driving factors were obtained by integrating land use changes and actual and potential productivity trends to identify the contribution of human activities and natural factors to sandy land dynamics during 2000–2018.The overall classification accuracy was as high as 89.30%,indicating that our proposed monitoring framework could accurately acquire information on sandy land distribution.The results showed that sandy land areas continued to decrease during 2000–2018 at an average rate of 323.94 km^(2) per year,while the decrease rate was as high as 865.33 km^(2) per year during 2015–2018(i.e.,this decrease rate nearly tripled over the entire study period).Further,the sandy land type generally showed a shift towards the direction of decreasing desertification.The most obvious changes were a decrease in mobile sandy lands and an increase in cultivated sandy lands,with relative change ratios of 18.52%and 47.47%,respectively,from 2000 to 2018.In-depth analysis showed that approximately 56.89%of the mobile sandy land was converted to semi-fixed sandy land,while 27.57%of the fixed sandy land was converted to cultivated sandy land.Human activities along with natural factors were the main driving factors of changes in sandy lands,accounting for 52.49%and 29.74%,respectively,and the changes dominated by natural factors were less than 1%.These results indicated that the BTSSCP was effective in restoring sandy lands from 2000 to 2018.Nonetheless,sandy lands had been significantly reduced and reclaimed;additionally,mobile sandy lands persisted in a large area,and degradation was prevalent to a certain extent.Hence,we strongly recommend the continuation of the project to significantly reduce mobile sandy lands in the future.