Dynamic monitoring of eco-environmental quality in Xi'an based on GEE and adjusted RSEI

PDF HTML阅读 XML下载 导出引用 引用提醒 基于谷歌地球引警和改进型遥感生态指数的西安市生态环境质量动态监测 DOI: 10.5846/stxb202112103510 作者: 作者单位: 作者简介: 通讯作者: 中图分类号: 基金项目: 国家自然科学基金项目（41371220，42071345）；陕西省重点研发项目（2020ZDLSF06-07） Dynamic monitoring of eco-environmental quality in Xi'an based on GEE and adjusted RSEI Author: Affiliation: Fund Project: National?Natural?Science?Foundation?of?China,No.41371220,No.42071345；Key Research Development Program Shaanxi Province,No.2020ZDLSF06-07 摘要 | 图/表 访问统计 参考文献 相似文献 引证文献 资源附件 文章评论 摘要:近年来，在经济全球化的背景下，西安市经济迅速增长，生态环境问题日益突出，快速全面地定量监测生态环境质量的时空变化，对指导生态环境保护具有重要意义。基于谷歌地球引警（GEE）平台，筛选2000、2004、2010、2015、2020年及其前后各一年的四季Landsat影像，利用主成分分析基于绿度（NDVI）、热度（LST）、干度（NDSI）、湿度（Wet）和气溶胶光学厚度（AOD）构建改进型遥感生态指数（ARSEI），实现西安市2000-2020年生态环境质量的动态监测，并运用莫兰指数（Moran's I）探讨西安市生态环境质量的空间自相关。以主成分效果最好的夏季为例，结果表明：（1）构建的ARSEI将大气污染因素考虑在内，第1主成分（PC1）贡献度在83%以上，能较好地集中各指标特征，有助于更加全面地评价研究区生态环境质量；（2）西安市2000-2020年平均ARSEI分别为0.565、0.521、0.572、0.644、0.695，生态环境质量总体呈现先退化后转好的趋势。20年来，生态环境质量较差和极差等级的面积减少了1339.08 km2，主要分布在秦岭以北地区，良好和极好等级区面积增加了2241.80 km2，主要位于南部的秦岭地区；（3）西安市生态环境质量改善区面积大于退化区，改善和退化情况在各辖区均有发生。2000-2004年退化情况最为严重，占比29.41%，而2010-2015年改善区占比最大，达31.62%；（4）5个年度的全局莫兰指数（Global Moran's I）均在0.627以上，表明西安市生态环境质量的空间分布具有较强的正相关，呈聚集分布，且以高-高、低-低分布为主。研究基于GEE平台实现了西安市生态环境质量的快速监测，可为生态环境监测与治理保护提供方法借鉴与数据支撑。 Abstract:As the capital city Province, has experienced rapid economic growth urbanization under globalization background recent decades still besets by increasingly prominent ecological problems. is also one typical cities suffering from severe aerosol pollution China. Rapidly comprehensively quantitative spatio-temporal variations crucial for regional guidance protection. This study was carried out Google Earth Engine (GEE), a new cloud-computing platform with merits easy access to tremendous public resources convenient processing substantially geospatial data. Four-season Landsat images 2000, 2004, 2010, 2015, 2020 those previous-successive target years were selected firstly. Then, principal component analysis (PCA) used improve previously developed remote sensing index (RSEI) adding optical depth (AOD) it, an (ARSEI) including greenness (NDVI), heat (LST), dryness (NDSI), humidity (Wet) AOD proposed dynamically monitor 2000 2020. Moreover, Moran adopted explore spatial autocorrelation Xi'an. Using summer, season best effect, as example, results showed that:(1) taking effect air into account, first (PC1) contribution ARSEI this paper more than 83%. could better concentrate characteristics each provide comprehensive evaluation (2) The average ARSEIs 0.565, 0.521,0.572, 0.644 0.695, respectively, indicating that urban degraded 2004 promoted In past 20 years, areas poor very decreased 1339.08 km2, which mainly located north Qinling Mountains. good excellent increased 2241.80 km2 concentrated areas, southern part city. (3) improved larger over district undergone both improvement degradation processes. It worth noting worst accounting 29.41%, greatest between 2010 31.62%. Generally speaking, been years. (4) All five Global I values above 0.627 2015 2020, indicated distribution strongly positive correlation. LISA cluster map aggregation dominated high-high low-low patterns. northern Mountains, while Based cloud platform, accomplished fast will method reference data support monitoring, management restoration.