基于GEE的地表覆盖与气候状态关联研究

doi:10.13474/j.cnki.11-2246.2021.381

摘要/Abstract

摘要： 为探究地表覆盖与气候状态间的关联性,本文选取2019年的Landsat影像数据,结合温度、降水量、PM_2.5浓度3种气候指标,利用GEE平台,结合NDVI、MNDWI、NDBI,采用SVM、RF、CART方法进行地表覆盖分类,探究气候指标与地表覆盖类型分布的关联性;提出了使用3种气候指标构建分类特征进行地表覆盖分类的方法,并通过消融试验分析了气候指标对地表覆盖分类精度的影响。结果表明:①RF有较好的分类结果,总体精度为96.0%;②3种气候指标均能提高地表覆盖分类精度,其中PM_2.5浓度效果最好;③温度与植被、水体关联性较大,PM_2.5浓度与城区、植被关联性较大,降水量与耕地关联性较大。

关键词: 地表覆盖, 气候, PM_2.5, GEE, 关联性, 分类

Abstract: In order to explore the correlation between surface cover and climate status, this paper selects Landsat image data in 2019 and three climate indicators using temperature, precipitation and PM_2.5 concentration, by GEE platform, combined with NDVI, MNDWI and NDBI, and uses SVM, RF and cart methods to classify surface cover, so as to explore the correlation between climate indicators and surface cover type distribution. The method of using three climate indicators to construct classification features for land cover classification is proposed, and the influence of climate indicators on the accuracy of land cover classification is analyzed through ablation experiments. The results show that:① RF has good classification results, and the overall accuracy is 96.0%; ② The three climate indexes can improve the accuracy of surface cover classification, and the concentration of PM_2.5 is the best; ③ Temperature is closely related to vegetation and water body, PM_2.5 concentration is highly correlated with urban area and vegetation, precipitation is closely related to cultivated land.

Key words: land cover, climate, PM_2.5, GEE, correlation, classification

中图分类号:

P237

陈伟, 刘湘媛, 曾圣, 张华平, 毛志芳. 基于GEE的地表覆盖与气候状态关联研究[J]. 测绘通报, 2021, 0(12): 99-104.

CHEN Wei, LIU Xiangyuan, ZENG Sheng, ZHANG Huaping, MAO Zhifang. Research on the correlation between land cover and climatic status based on Google Earth Engine[J]. Bulletin of Surveying and Mapping, 2021, 0(12): 99-104.

参考文献

[1] 中华人民共和国国家质量监督检验检疫总局. 中华人民共和国推荐性国家标准:土地基本术语GB/T 19231-2003[S]. 北京:中国标准出版社, 2003.
[2] 王乐. 陆面过程及地表覆盖对中国地区区域气候变化模拟的影响评估[D]. 南京:南京大学, 2018.
[3] 陈羽阳, 王婧, 赵聆言, 等. 城市公园绿地对周边环境空气PM10和PM2.5的影响及效应场特征——以武汉市中山公园为例[J]. 生态学杂志, 2021, 40(7):2263-2276.
[4] 陈泮勤,孙成权. 国际全球变化研究核心计划(二)[M]. 北京:气象出版社, 1994.
[5] 宋帅, 鞠永茂, 王汉杰. 有序人类活动造成的土地利用变化对区域降水的可能影响[J]. 气候与环境研究, 2008, 13(6):759-774.
[6] 田育红, 纪中奎, 刘鸿雁. 内蒙古高原中部主要气候因子及地表覆盖对沙尘暴影响分析[J]. 应用气象学报, 2005, 16(4):476-483.
[7] 宋春桥, 游松财, 柯灵红, 等. 藏北高原地表覆盖时空动态及其对气候变化的响应[J]. 应用生态学报, 2011, 22(8):2091-2097.
[8] 冀彩星, 延军平. 西安城市气候对地表覆盖变化的响应[J]. 干旱区资源与环境, 2011, 25(7):117-123.
[9] GORELICK N, HANCHER M, DIXON M, et al. Google Earth Engine:planetary-scale geospatial analysis for everyone[J]. Remote Sensing of Environment, 2017, 202:18-27.
[10] YANG Guofu, XU Ronghua, CHEN Yi, et al. Identifying the greenhouses by Google Earth Engine to promote the reuse of fragmented land in urban fringe[J]. Sustainable Cities and Society, 2021, 67:102743.
[11] 湖南统计局.2020湖南统计年鉴[M].北京:中国统计出版,2020.
[12] 付甜梦, 张丽, 陈博伟, 等. 基于GEE平台的海岛地表覆盖提取及变化监测——以苏拉威西岛为例[J]. 遥感技术与应用, 2021, 36(1):55-64.
[13] 徐涵秋. 利用改进的归一化差异水体指数(MNDWI)提取水体信息的研究[J]. 遥感学报, 2005, 9(5):589-595.
[14] 史泽鹏, 马友华, 王玉佳, 等. 遥感影像土地利用/覆盖分类方法研究进展[J]. 中国农学通报, 2012, 28(12):273-278.
[15] CORTES C, VAPNIK V. Support-vector networks[J]. Machine Learning, 1995, 20(3):273-297.
[16] RASMUSSEN C E. Gaussian processes in machine learning[M]//Advanced Lectures on Machine Learning. Berlin, Heidelberg:Springer Berlin Heidelberg, 2004:63-71.
[17] MOISEN G G. Classification and regression trees[M]//Encyclopedia of Ecology. Amsterdam:Elsevier, 2008:582-588.
[18] 周志华. 机器学习[M]. 北京:清华大学出版社, 2016.
[19] BREIMAN L.Random forests[J].Machine learning,2001,45(1):5-32.
[20] GISLASON P O, BENEDIKTSSON J A, SVEINSSON J R. Random forests for land cover classification[J]. Pattern Recognition Letters, 2006, 27(4):294-300.
[21] 周珂, 柳乐, 程承旗, 等. 基于谷歌地球引擎的开封城区2010-2019年水体分布变化研究[J]. 科学技术与工程, 2021, 21(6):2397-2404.
[22] ZURQANI H A, POST C J, MIKHAILOVA E A, et al. Geospatial analysis of land use change in the Savannah River Basin using Google Earth Engine[J]. International Journal of Applied Earth Observation and Geoinforma-tion, 2018, 69:175-185.