Accuracy evaluation and consistency analysis of multi-source remote sensing land cover data in the Yellow River Basin
-
摘要: 开源、多分辨率、及时的土地覆盖产品为了解全球地表覆盖状况、陆面过程模型模拟以及社会经济发展决策等提供了丰富的数据支撑, 但多源的数据存在不同程度的不确定性, 在区域尺度如何选择合适的土地覆被产品成为应用中的难题。本研究以黄河流域为例, 对分辨率从30 m到1000 m的CLCD_v01_2020、GLOBELAND30、GLC_FCS30_2020、LANDCOVER (300 m)、MCD12Q1 (500 m)和CNLUCC1000 (1000 m)等6种2020年土地覆被产品进行区域尺度精度评价和一致性分析。基于Google Earth采集的1540个样本点分析6种数据在黄河流域的总体精度, 以最高精度的数据为参考对其他数据进行面积一致性分析, 并对6种数据进行类别混淆分析和混淆图谱分析。结果表明, 6种数据中分类精度最高的为CLCD_v01_2020, 总体精度(overall accuracy, OA)达88.12%; 其次是GLOBELAND30 (OA=85.32%)、GLC_FCS30_2020 (OA=84.09%)、LANDCOVER300 (OA=77.79%)、MCD12Q1 (OA=73.38%)、CNLUCC1000 (OA=71.82%), 30 m土地覆被产品的KAPPA系数均在0.8以上, 随着空间分辨率的下降, 分类精度下降。 6种数据的土地覆被类别组成的相对比例总体上趋于一致, 但在耕地和草地两类土地覆被类别上仍存在较大差异, GLC_FCS30_2020与参考数据CLCD_v01_2020的相关性最高, R2达到0.9976。通过类别混淆分析可知6种数据普遍对耕地、林地和草地的混淆较为严重。类别混淆空间分析表明, 验证数据与参考数据在黄河上游的草地、中下游部分耕地和建设用地等类型较为单一的区域一致性较高, 而在陕西北部、山西西部的一致性较差, 主要表现为草地和林地的混淆。针对黄河流域土地覆被数据一级分类, 本研究建议, 30 m分辨率的数据中选择CLCD_v01_2020, 百米级分辨率数据中选择LANDCOVER300, 二级分类则可以根据所需的分类体系选择合适的数据。Abstract: With the development of multi-source remote-sensing platforms and technologies, various land cover datasets have been developed that provide a wealth of data to support the understanding of global land cover conditions, land surface process model simulations, and socioeconomic development decisions. However, selecting appropriate data for different regions from nationally or globally available land cover datasets is challenging. In this study, six land cover products in 2020 over the Yellow River Basin, including CLCD_v01_2020, GLOBELAND30, GLC_FCS30_2020, LANDCOVER (300 m), MCD12Q1 (500 m), and CNLUCC1000 (1000 m), with resolutions ranging from 30 to 1000 m, were evaluated for regional-scale accuracy and consistency analysis. Accuracy analyses were performed on six products based on 1540 samples for seven land cover types collected by Google Earth. Data with the highest overall accuracy (OA) were used as a reference for the area consistency analysis of the other five products. Category confusion and confusion mapping analyses were performed on six types of data. Hopefully, this study will provide a scientific reference for users to select appropriate land cover data for the Yellow River Basin. The results showed that the highest classification accuracy was for CLCD_v01_2020, with an OA of 88.12%, followed by GLOBELAND30 (OA=85.32%), GLC_FCS30_2020 (OA=84.09%), LANDCOVER300 (OA=77.79%), MCD12Q1 (OA=73.38%), and CNLUCC1000 (OA=71.82%). The KAPPA coefficients of the land cover products with a resolution of 30 m were all above 0.8, and the classification accuracy decreased as the spatial resolution decreased. CLCD_v01_2020, with the highest OA, was used as the reference dataset, and the area correlations and confusion mapping were calculated separately for the remaining five validation product datasets. The relative proportions of different land cover types were generally consistent across the six products; however, there were still large differences between croplands and grasslands. GLC_FCS30_2020 had the highest correlation with the reference data CLCD_v01_2020, with an R2 value of 0.9976. Category confusion analysis showed that the six data types were generally confused between croplands, forests, and grasslands. There was good consistency in the grasslands of eastern Qinghai in the upper reaches of the Yellow River and the cropland and construction land of the middle and lower reaches. The areas of poor consistency were mainly in the middle reaches of the Yellow River in northern Shaanxi and western Shanxi, which were mainly confused grasslands with forests. For the primary classification of land cover data in the Yellow River Basin, it is recommended that CLCD_v01_2020 data be selected for 30 m resolution and LANDCOVER300 for 100-m scale resolution data. In contrast, secondary classification can be chosen according to the desired classification system.
-
![]()
图 2 基于Google Earth的黄河流域样本点分布
Figure 2. Sample points distribution of the Yellow River Basin based on Google Earth
![]()
图 3 不同遥感产品的黄河流域土地覆被类别面积组成(a)与面积偏差(b)
Figure 3. Area composition (a) and area deviation (b) of different land cover types in the Yellow River basin of different remote-sensing products
![]()
图 4 多源遥感数据的土地覆被类型混淆程度
Figure 4. Degrees of confusion between land cover types of different remote-sensing products
![]()
图 5 多源遥感产品与参考数据的类别混淆空间图谱
R表示参考数据, V代表验证数据。R represents the reference data and V represents the verification data.
Figure 5. Confuse spatial atlases of multi-source remote-sensing land cover products with the categories of reference data
表 1 研究所用多源遥感土地覆被产品的参数表
Table 1 Parameters of multi-source remote-sensing land cover products used in the study
产品名称
Product name来源
Source分辨率
Resolution (m)区域
Region制图单位
Cartographic organization分类方法
Classification method传感器
Sensor分类体系
Classification systemCLCD_v01_2020 https://zenodo.org./ 30 中国
China武汉大学
Wuhan University随机森林
Random forestLandsat LCCS(9) GLOBELAND30 https://GlobeLand30/ 30 全球
Global国家基础地理
信息中心等
National Basic Geographic Information Center, etc监督分类
Supervise classificationLandsat LCCS(10) GLC_FCS30_2020 https://data.casearth.cn/ 30 全球
Global中国科学院空天信息创新研究院
Institute of Aerospace Information Innovation, Chinese Academy of Sciences监督分类
Supervise classificationLandsat IGBP(29) LANDCOVER https://cds.climate.copernicus.eu/ 300 全球
Global欧洲航天局
European Space Agency非监督分类
Unsupervised classificationMERIS
PROBA-V
Sentinel-3LCCS(36) MCD12Q1 https://www.earthdata.nasa.gov/ 500 全球
Global波士顿大学
Boston University监督分类
Supervise classificationMODIS IGBP(17) CNLUCC1000 https://www.resdc.cn/ 1000 中国
China中国科学院地理科学与资源研究所
Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences目视解译
Visual interpretationLandsat LCCS(25) 
下载: 导出CSV
表 2 研究所用土地覆被产品类别聚合表
Table 2 Aggregation table of land cover types of multi-source remote-sensing products used by the study
· 产品名称
Product name耕地
Cropland林地
Forest草地
Grassland水域
Waters冰川
Glacier裸地
Barren建设用地
Construction landCLCD_v01_2020 1 2, 3 4 5, 9 6 7 8 GLOBELAND30 10 20, 40 30, 70 50, 60 100 90 80 GLC_FCS30_2020 10, 20 12~122 11, 130~153 180, 210 220 200~202 190 CNLUCC1000 11, 12 21~24 31~33 41~43, 45, 46, 64, 99 44 61~63, 65~67 51~53 MCD12Q1 12, 14 1~7 8~10 11, 17 15 16 13 LANDCOVER 10~30 40~122, 170, 180 130, 150 210 220 200~202 190 表中的数字代表不同遥感产品原始分类体系类别代码。The numbers in the table represent the original classification system category codes for different remote sensing products.
下载: 导出CSV
表 3 不同遥感产品的土地覆被精度
Table 3 Accuracy results of land cover types of different remote-sensing products
% 土地覆被类型
Land cover typeCLCD_v01_2020 GLOBELAND30 GLC_FCS30_2020 LANDCOVER300 MCD12Q1 CNLUCC1000 UA PA UA PA UA PA UA PA UA PA UA PA 耕地 Cropland 92.38 96.48 87.03 99.44 93.28 82.22 80.74 89.26 84.99 93.33 79.40 83.52 林地 Forest 97.42 90.10 95.56 80.89 95.99 89.76 93.88 89.08 99.48 65.19 93.45 73.04 草地 Grassland 64.97 93.63 60.32 73.04 48.78 88.24 43.60 73.53 40.09 91.18 45.57 73.04 水域 Waters 100.00 73.63 96.21 69.78 97.45 84.07 100.00 46.70 91.43 17.58 70.45 34.07 冰川 Glacier 100.00 56.90 100.00 67.24 100.00 53.45 100.00 50.00 100.00 37.93 100.00 44.83 裸地 Barren 81.61 67.62 71.13 65.71 90.36 71.43 76.00 36.19 62.79 51.43 55.74 64.76 建设用地 Construction land 91.08 90.51 97.50 98.73 99.33 94.30 97.45 96.84 95.27 89.24 75.56 86.08 OA 88.12 85.32 84.09 77.79 73.38 71.82 KAPPA 84.91 81.22 80.10 71.54 66.06 64.30 UA为使用者精度, PA为生产者精度, OA为总体精度。UA is the consumer accuracy, PA is the producer accuracy, and OA is the overall accuracy.
下载: 导出CSV
-
[1] 张镱锂, 刘林山, 王兆锋, 等. 青藏高原土地利用与覆被变化的时空特征[J]. 科学通报, 2019, 64(27): 2865−2875 doi: 10.1360/TB-2019-0046 ZHANG Y L, LIU L S, WANG Z F, et al. Spatial and temporal characteristics of land use and cover changes in the Tibetan Plateau[J]. Chinese Science Bulletin, 2019, 64(27): 2865−2875 doi: 10.1360/TB-2019-0046
[2] 李广东. 全球土地覆被时空变化与中国贡献[J]. 地理学报, 2022, 77(2): 353−368 LI G D. Spatio-temporal change of global land cover and China’s contribution[J]. Acta Geographica Sinica, 2022, 77(2): 353−368
[3] 胡乔利, 齐永青, 胡引翠, 等. 京津冀地区土地利用/覆被与景观格局变化及驱动力分析[J]. 中国生态农业学报, 2011, 19(5): 1182−1189 HU Q L, QI Y Q, HU Y C, et al. Changes and driving forces of land use/cover and landscape patterns in Beijing-Tianjin-Hebei region[J]. Chinese Journal of Eco-Agriculture, 2011, 19(5): 1182−1189
[4] 白燕, 冯敏. 全球尺度多源土地覆被数据融合与评价研究[J]. 地理学报, 2018, 73(11): 2223−2235 BAI Y, FENG M. Data fusion and accuracy evaluation of multi-source global land cover datasets[J]. Acta Geographica Sinica, 2018, 73(11): 2223−2235
[5] 廖顺宝, 葛乐玮, 王艳萍, 等. 利用地形参数提高土地覆被分类精度方法的改进[J]. 遥感信息, 2021, 36(3): 10−16 LIAO S B, GE L W, WANG Y P, et al. Improvement of method of enhancing classification accuracy of land cover based on terrain factors[J]. Remote Sensing Information, 2021, 36(3): 10−16
[6] 马红梅, 王苗苗, 刘勇. 多源遥感数据土地覆被空间尺度效应探讨[J]. 遥感信息, 2017, 32(2): 149−155 MA H M, WANG M M, LIU Y. Spatial scale effect of land cover based on multi-source remote sensing data[J]. Remote Sensing Information, 2017, 32(2): 149−155
[7] 胡云锋, 张千力, 戴昭鑫, 等. 多源遥感土地覆被产品在欧洲地区的一致性分析[J]. 地理研究, 2015, 34(10): 1839−1852 HU Y F, ZHANG Q L, DAI Z X, et al. Agreement analysis of multi-sensor satellite remote sensing derived land cover products in the Europe continent[J]. Geographical Research, 2015, 34(10): 1839−1852
[8] 宋宏利, 张晓楠. 中国区域多源土地覆被遥感产品精度分析与验证[J]. 农业工程学报, 2012, 28(22): 207−214, 296 SONG H L, ZHANG X N. Precision analysis and validation of multi-sources landcover products derived from remote sensing in China[J]. Transactions of the Chinese Society of Agricultural Engineering, 2012, 28(22): 207−214, 296
[9] GAO Y, LIU L Y, ZHANG X, et al. Consistency analysis and accuracy assessment of three global 30-m land-cover products over the European union using the LUCAS dataset[J]. Remote Sensing, 2020, 12(21): 3479 doi: 10.3390/rs12213479
[10] LIANG L, LIU Q S, LIU G H, et al. Accuracy evaluation and consistency analysis of four global land cover products in the Arctic region[J]. Remote Sensing, 2019, 11(12): 1396 doi: 10.3390/rs11121396
[11] PÉREZ-HOYOS A, GARCÍA-HARO F J, VALCÁRCEL N. Incorporating sub-dominant classes in the accuracy assessment of large-area land cover products: application to GlobCover, MODISLC, GLC2000 and CORINE in Spain[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(1): 187−205 doi: 10.1109/JSTARS.2013.2258659
[12] 戴昭鑫, 胡云锋, 张千力. 多源卫星遥感土地覆被产品在南美洲的一致性分析[J]. 遥感信息, 2017, 32(2): 137−148 DAI Z X, HU Y F, ZHANG Q L. Agreement analysis of multi-source land cover products derived from remote sensing in South America[J]. Remote Sensing Information, 2017, 32(2): 137−148
[13] 赵凌美, 张时煌. 2种常用的全球土地利用/覆被历史数据集在中国区域的精度评价[J]. 西北农林科技大学学报(自然科学版), 2013, 41(8): 133−140, 148 ZHAO L M, ZHANG S H. The accuracy evaluation of two common global historical land use/cover datasets in China[J]. Journal of Northwest A & F University (Natural Science Edition), 2013, 41(8): 133−140, 148
[14] 陈逸聪, 邵华, 李杨. 多源土地覆被产品在长三角地区的一致性分析与精度评价[J]. 农业工程学报, 2021, 37(6): 142−150 CHEN Y C, SHAO H, LI Y. Consistency analysis and accuracy assessment of multi-source land cover products in the Yangtze River Delta[J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(6): 142−150
[15] 徐泽源, 罗庆辉, 许仲林. 新疆地区土地覆被遥感数据的一致性研究[J]. 地球信息科学学报, 2019, 21(3): 427−436 XU Z Y, LUO Q H, XU Z L. Consistency of land cover data derived from remote sensing in Xinjiang[J]. Journal of Geo-Information Science, 2019, 21(3): 427−436
[16] 左玉珊, 王卫, 郝彦莉, 等. 基于MODIS影像的土地覆被分类研究−以京津冀地区为例[J]. 地理科学进展, 2014, 33(11): 1556−1565 doi: 10.11820/dlkxjz.2014.11.012 ZUO Y S, WANG W, HAO Y L, et al. Land cover classification based on MODIS images: taking the Beijing-Tianjin-Hebei region as an example[J]. Progress in Geography, 2014, 33(11): 1556−1565 doi: 10.11820/dlkxjz.2014.11.012
[17] 邵全琴, 赵志平, 刘纪远, 等. 近30年来三江源地区土地覆被与宏观生态变化特征[J]. 地理研究, 2010, 29(8): 1439−1451 SHAO Q Q, ZHAO Z P, LIU J Y, et al. The characteristics of land cover and macroscopical ecology changes in the source region of three rivers on Qinghai-Tibet Plateau during last 30 years[J]. Geographical Research, 2010, 29(8): 1439−1451
[18] 侯婉, 侯西勇. 全球海岸带多源土地利用/覆盖遥感分类产品一致性分析[J]. 地球信息科学学报, 2019, 21(7): 1061−1073 HOU W, HOU X Y. Consistency of the multiple remote sensing-based land use and land cover classification products in the global coastal zones[J]. Journal of Geo-Information Science, 2019, 21(7): 1061−1073
[19] 黄亚博, 廖顺宝. 首套全球30 m分辨率土地覆被产品区域尺度精度评价−以河南省为例[J]. 地理研究, 2016, 35(8): 1433−1446 HUANG Y B, LIAO S B. Regional accuracy assessments of the first global land cover dataset at 30-meter resolution: a case study of Henan Province[J]. Geographical Research, 2016, 35(8): 1433−1446
[20] 王冰泉, 冉有华. 土地覆被遥感产品真实性检验方法对比[J]. 遥感技术与应用, 2022, 37(1): 196−204 WANG B Q, RAN Y H. Comparison of accuracy assessment methods of remote sensing based land cover products[J]. Remote Sensing Technology and Application, 2022, 37(1): 196−204
[21] 朱筠, 孙九林, 秦奋, 等. 2015年中国1∶10万土地覆被数据河南地区精度评价[J]. 中国土地科学, 2019, 33(3): 59−67 ZHU J, SUN J L, QIN F, et al. Accuracy assessment of the 1∶100 000 land cover data of Henan Province in 2015[J]. China Land Science, 2019, 33(3): 59−67
[22] 2020中国统计年鉴[J]. 统计理论与实践, 2021, 501(1): 2 2020 China Statistical Yearbook[J]. Statistical Theory and Practice, 2021, 501(1): 2
[23] 杨洁, 谢保鹏, 张德罡. 基于InVEST和CA-Markov模型的黄河流域碳储量时空变化研究[J]. 中国生态农业学报(中英文), 2021, 29(6): 1018−1029 YANG J, XIE B P, ZHANG D G. Spatio-temporal evolution of carbon stocks in the Yellow River Basin based on InVEST and CA-Markov models[J]. Chinese Journal of Eco-Agriculture, 2021, 29(6): 1018−1029
[24] LIU B, PAN L B, QI Y, et al. Land use and land cover change in the Yellow River Basin from 1980 to 2015 and its impact on the ecosystem services[J]. Land, 2021, 10(10): 1080 doi: 10.3390/land10101080
[25] YANG J, HUANG X. The 30 m annual land cover dataset and its dynamics in China from 1990 to 2019[J]. Earth System Science Data, 2021, 13(8): 3907−3925 doi: 10.5194/essd-13-3907-2021
[26] JUN C, BAN Y F, LI S N. Open access to Earth land-cover map[J]. Nature, 2014, 514(7523): 434
[27] ZHANG X, LIU L Y, CHEN X D, et al. GLC_FCS30: global land-cover product with fine classification system at 30 m using time-series Landsat imagery[J]. Earth System Science Data, 2021, 13(6): 2753–2776
[28] FRIEDL M, SULLA-MENASHE D. MCD12Q1 MODIS/Terra+Aqua Land Cover Type Yearly L3 Global 500 m SIN Grid V006[DB/OL]. NASA EOSDIS Land Processes DAAC. [2023-2-10]. https://doi.org/10.5067/MODIS/MCD12Q1.006
[29] 刘纪远. 中国资源环境遥感宏观调查与动态研究[M]. 北京: 中国科学技术出版社, 1996 LIU J Y. Macro-scale Survey and Dynamic Study of Natural Resources and Environment of China by Remote Sensing[M]. Beijing: China Science and Technology Press, 1996
[30] NWILO P C, OKOLIE C J, ONYEGBULA J C, et al. Positional accuracy assessment of historical Google Earth imagery in Lagos State, Nigeria[J]. Applied Geomatics, 2022, 14(3): 545−568 doi: 10.1007/s12518-022-00449-9
[31] PADMA S P, VIDHYA L, SIVAKUMAR P, et al. Simulation of land use/land cover dynamics using Google Earth data and QGIS: a case study on outer ring road, Southern India[J]. Sustainability, 2022, 14(24): 16373−16373 doi: 10.3390/su142416373
[32] 仝冉, 杨雅萍, 陈晓娜. 多源30 m分辨率土地覆被数据在蒙古高原的一致性分析和精度评价[J]. 地球信息科学学报, 2022, 24(12): 2420−2434 doi: 10.12082/dqxxkx.2022.220578 TONG R, YANG Y P, CHEN X N. Consistent analysis and accuracy evaluation of multisource land cover datasets in 30 m spatial resolution over the Mongolian Plateau[J]. Journal of Geo-Information Science, 2022, 24(12): 2420−2434 doi: 10.12082/dqxxkx.2022.220578
[33] 宋金超, 李新虎, 吝涛, 等. 基于夜晚灯光数据和Google Earth的城市建成区提取分析[J]. 地球信息科学学报, 2015, 17(6): 750−756 SONG J C, LI X H, XIAO T, et al. Extraction and analysis of urban built-up area based on night lighting data and Google Earth[J]. Journal of Geo-Information Science, 2015, 17(6): 750−756
[34] 刘琼欢, 张镱锂, 刘林山, 等. 七套土地覆被数据在羌塘高原的精度评价[J]. 地理研究, 2017, 36(11): 2061−2074 LIU Q H, ZHANG Y L, LIU L S, et al. Accuracy evaluation of the seven land cover data in Qiangtang Plateau[J]. Geographical Research, 2017, 36(11): 2061−2074
[35] 康军梅. 多源遥感土地覆被产品一致性评价及要素提取分析应用研究[D]. 西安: 长安大学, 2020 KANG J M. Research on consistency assessment of multi-source landcover products and application of element extraction analysis[D]. Xi’an: Chang’an University, 2020
计量
- 文章访问数:
- HTML全文浏览量:
- PDF下载量:
