基于Google Earth Engine与机器学习的大尺度30m分辨率沙地灌木覆盖度估算
|
摘要
相较于降雨充沛的南方,中国北方沙地植被呈现覆盖整体偏低、空间异质性强的特点。灌木作为该区域的优势植被,对于风沙固定、食品/木材供给起着极为重要的作用。针对当前大尺度、中高分辨率干旱地区灌木覆盖度遥感产品缺失的现状,研究提出了一套通过Collect Earth样本收集器进行样本采集、利用Google Earth Engine遥感云平台的数据与计算优势开展大尺度灌木覆盖度估算的方法,并选取中国北方四大沙地之一的毛乌素沙地开展了示范应用。研究结果表明:(1)Collect Earth样本收集器可以有效地获取地面灌木覆盖度样本数据集,可以将灌木与高大乔木与草本植被进行有效区分,为灌木覆盖度估算样本集的建立打下了基础;(2)利用Landsat数据与其他辅助数据,机器学习算法可以较好地实现灌木覆盖度的估算,CART模型确定性系数R~2为0.73,均方根误差(Root Mean Square Error, RMSE)为13.66%,预测精度(Estimated Accuracy, EA)为61.8%,SVM模型R~2为0.72,RMSE为13.73%,EA为61.6%;(3)提出的基于GEE的灌木覆盖度估算体系可为我国乃至全球尺度干旱地区沙地灌木覆盖度信息提取提供有效支撑,具有较大的应用潜力。
Compared with the humid region in southern China, the sandy land vegetation is characterized by low coverage and a scattered spatial distribution in the north. Shrub is a dominant vegetation type in this region and plays an important role in sand control, food/timber product provision, etc. In view of the current lack of the medium-and high-resolution remote sensing products for shrub coverage at large-scale in arid regions, we proposed a new approach to estimate shrub coverage at large scale based on Collect Earth sample collector and Google Earth Engine(GEE) platform. Then this approach was applied to Mu Us sandy land, one of the four major sandy lands in Northern China. The results showed that:(1) Collect Earth sample collector could effectively obtain the ground shrub coverage sample data set for distinguishing shrubs from tall trees and herbaceous vegetation, which laid a foundation for the establishment of shrub coverage estimation;(2) Using Landsat data, other ancillary data, and the machine learning algorithm in GEE, the shrub coverage could be estimated effectively. The CART model had a deterministic coefficient R~2 of 0.73 and a Root Mean Square Error(RMSE) of 13.66% with the estimated Accuracy(EA) of 61.8%. For SVM model, R~2, RMSE, and EA were 0.72, 13.73%, and 61.6%, respectively.(3) The GEE-based approach proposed in this study could provide support to shrub coverage estimation in sandy land at the national and even the global scale with a potential application.
Compared with the humid region in southern China, the sandy land vegetation is characterized by low coverage and a scattered spatial distribution in the north. Shrub is a dominant vegetation type in this region and plays an important role in sand control, food/timber product provision, etc. In view of the current lack of the medium-and high-resolution remote sensing products for shrub coverage at large-scale in arid regions, we proposed a new approach to estimate shrub coverage at large scale based on Collect Earth sample collector and Google Earth Engine(GEE) platform. Then this approach was applied to Mu Us sandy land, one of the four major sandy lands in Northern China. The results showed that:(1) Collect Earth sample collector could effectively obtain the ground shrub coverage sample data set for distinguishing shrubs from tall trees and herbaceous vegetation, which laid a foundation for the establishment of shrub coverage estimation;(2) Using Landsat data, other ancillary data, and the machine learning algorithm in GEE, the shrub coverage could be estimated effectively. The CART model had a deterministic coefficient R~2 of 0.73 and a Root Mean Square Error(RMSE) of 13.66% with the estimated Accuracy(EA) of 61.8%. For SVM model, R~2, RMSE, and EA were 0.72, 13.73%, and 61.6%, respectively.(3) The GEE-based approach proposed in this study could provide support to shrub coverage estimation in sandy land at the national and even the global scale with a potential application.
引文
[1] 王兵,魏江生,胡文.中国灌木林-经济林-竹林的生态系统服务功能评估.生态学报,2011,31(7):1936-1945.
[2] 姚雪玲,姜丽娜,李龙,王锋,吴波,郭秀江.浑善达克沙地6种灌木生物量模拟.生态学报,2019,39(3):905-912.
[3] 李苗苗.植被覆盖度的遥感估算方法研究[D].北京:中国科学院研究生院(遥感应用研究所),2003.
[4] Naidoo L,Mathieu R,Main R,Wessels K,Asner G P.L-band synthetic aperture radar imagery performs better than optical datasets at retrieving woody fractional cover in deciduous,dry savannahs.International Journal of Applied Earth Observation and Geoinformation,2016,52:54-64.
[5] Liu X,Liu H Y,Qiu S,Wu X C,Tian Y H,Hao Q.An improved estimation of regional fractional woody/herbaceous cover using combined satellite data and high-quality training samples.Remote Sensing,2017,9(1):32.
[6] Suess S,van der Linden S,Okujeni A,Griffiths P,Leit?o P,Schwieder M,Hostert P.Characterizing 32 years of shrub cover dynamics in southern Portugal using annual Landsat composites and machine learning regression modeling.Remote Sensing of Environment,2018,219:353-364.
[7] Baumann M,Levers C,Macchi L,Bluhm H,Waske B,Gasparri N I,Kuemmerle T.Mapping continuous fields of tree and shrub cover across the Gran Chaco using Landsat 8 and sentinel-1 data.Remote Sensing of Environment,2018,216:201-211.
[8] Ge J,Meng B P,Liang T G,Feng Q S,Gao J L,Yang S X,Huang X D,Xie H J.Modeling alpine grassland cover based on MODIS data and support vector machine regression in the headwater region of the Huanghe River,China.Remote Sensing of Environment,2018,218:162-173.
[9] 张瑾,李晓松,吴炳方.基于分类回归树的密云水库上游森林覆盖度遥感估算.遥感技术与应用,2014,29(3):394-400.
[10] Huang H B,Chen Y L,Clinton N,Wang J,Wang X Y,Liu C X,Gong P,Yang J,Bai Y Q,Zheng Y M,Zhu Y L.Mapping major land cover dynamics in Beijing using all Landsat images in google earth engine.Remote Sensing of Environment,2017,202:166-176.
[11] 高志海,李增元,魏怀东,丁锋,丁国栋.干旱地区植被指数(VI)的适宜性研究.中国沙漠,2006,26(2):243-248.
[12] Li X S,Zheng G X,Wang J Y,Ji C C,Sun B,Gao Z H.Comparison of methods for estimating fractional cover of photosynthetic and non-photosynthetic vegetation in the Otindag sandy land using GF-1 wide-field view data.Remote Sensing,2016,8(10):800.
[13] Bastin J,Berrahmouni N,Grainger A,Maniatis D,Mollicone D,Moore R,Patriarca C,Picard N,Sparrow B,Maria Abraham E,Aloui K,Atesoglu A,Attore F,Bassüllü C,Bey A,Garzuglia M,G.García-Montero L,Groot N,Guerin G,Laestadius L,Lowe A J,Mamane B,Marchi G,Patterson P,Rezende M,Ricci S,Salcedo I,Sanchez-Paus Diaz A,Stolle F,Surappaeva V,Castro R.The extent of forest in dryland biomes.Science,2017,356(6338):635-638.
[14] Sexton J O,Song X P,Feng M,Noojipady P,Anand A,Huang C Q,Kim D H,Collins K,Channan S,DiMiceli C,Townshend J.Global,30-m resolution continuous fields of tree cover:Landsat-based rescaling of MODIS vegetation continuous fields with Lidar-based estimates of error.International Journal of Digital Earth,2013,6(5):427-448.
[15] Herrmann S M,Wickhorst A J,Marsh S E.Estimation of tree cover in an agricultural parkland of Senegal using rule-based regression tree modeling.Remote Sensing,2013,5(10):4900-4918.
[16] Brandt M,Hiernaux P,Tagesson T,Verger A,Rasmussen K,Diouf A A,Mbow C,Mougin E,Fensholt R.Woody plant cover estimation in drylands from earth observation based seasonal metrics.Remote Sensing of Environment,2016,172:28-38.
[17] Higginbottom T P,Symeonakis E,Meyer H,Van der Linden S.Mapping fractional woody cover in semi-arid savannahs using multi-seasonal composites from Landsat data.ISPRS Journal of Photogrammetry and Remote Sensing,2018,139:88-102.
[18] 李晓松,李增元,吴波,高志海,白黎娜,王琫瑜.基于光谱混合分析的毛乌素沙地油蒿群落覆盖度提取.遥感学报,2007,11(6):923-930.
[19] 李晓松,高志海,李增元,白黎娜,王琫瑜.基于高光谱混合像元分解的干旱地区稀疏植被覆盖度估测.应用生态学报,2010,21(1):152-158.
[20] 吴俊君,高志海,李增元,王红岩,庞勇,孙斌,李长龙,李绪志,张九星.基于天宫一号高光谱数据的荒漠化地区稀疏植被参量估测(英文).光谱学与光谱分析,2014,34(3):751-756.
[21] Gorelick N,Hancher M,Dixon M,Ilyushchenko S,Thau D,Moore D.Google earth engine:planetary-scale geospatial analysis for everyone.Remote Sensing of Environment,2017,202:18-27.
[22] Langner A,Miettinen J,Kukkonen M,Vancutsem C,Simonetti D,Vieilledent G,Verhegghen A,Gallego J,Stibig H J.Towards operational monitoring of forest canopy disturbance in evergreen rain forests:a test case in continental Southeast Asia.Remote Sensing,2018,10(4):544.
[23] Dong J W,Xiao X G,Menarguez M A,Zhang G L,Qin Y W,Thau D,Biradar C,Moore III B.Mapping paddy rice planting area in Northeastern Asia with Landsat 8 images,phenology-based algorithm and google earth engine.Remote Sensing of Environment,2016,185:142-154.
[24] Xiong J,Thenkabail P S,Tilton J C,Gumma M K,Teluguntla P,Oliphant A,Congalton R G,Yadav K,Gorelick N.Nominal 30-m cropland extent map of continental Africa by integrating pixel-based and object-based algorithms using sentinel-2 and landsat-8 data on google earth engine.Remote Sensing,2017,9(10):1065.
[25] Parastatidis D,Mitraka Z,Chrysoulakis N,Abrams M.Online global land surface temperature estimation from Landsat.Remote Sensing,2017,9(12):1208.
[26] Zhou D J,Zhao X,Hu H F,Shen H H,Fang J Y.Long-term vegetation changes in the four mega-sandy lands in Inner Mongolia,China.Landscape Ecology,2015,30(9):1613-1626.
[27] 吴波,慈龙骏.毛乌素沙地景观格局变化研究.生态学报,2001,21(2):191-196.
[28] 冯颖.毛乌素沙地植被盖度变化及其对气候变化的响应[D].北京:北京林业大学,2015.
[29] 张新时.毛乌素沙地的生态背景及其草地建设的原则与优化模式.植物生态学报,1994,18(1):1-16.
[30] 屠志方,李梦先,孙涛.第五次全国荒漠化和沙化监测结果及分析.林业资源管理,2016,(1):1-5,13-13.
[31] Abatzoglou J T,Dobrowski S Z,Parks S A,Hegewisch K C.TerraClimate,a high-resolution global dataset of monthly climate and climatic water balance from 1958-2015.Scientific Data,2018,5:170191.
[32] 孙斌,李增元,郭中,高志海,王琫瑜.高分一号与Landsat TM数据估算稀疏植被信息对比.遥感信息,2015,30(5):48-56.
[33] Huete A R.A soil-adjusted vegetation index (SAVI).Remote Sensing of Environment,1988,25(3):295-309.
[34] Bey A,Sánchez-Paus Díaz A,Maniatis D,Marchi G,Mollicone D,Ricci S,Bastin J F,Moore R,Federici S,Rezende M,Patriarca,C,Turia R,Gamoga G,Abe H,Kaidong E,Micel G.Collect earth:land use and land cover assessment through augmented visual interpretation.Remote Sensing,2016,8(10):807.
[35] Breiman L,Friedman J H,Olshen R A and Stone C J.Classification and Regression Trees.Belmont,Calif.:Wadsworth International Group,1984:1-358.
[36] 赵萍,傅云飞,郑刘根,冯学智,Satyanarayana B.基于分类回归树分析的遥感影像土地利用/覆被分类研究.遥感学报,2005,9(6):708-716.
[37] Cortes C,Vapnik V.Support-vector networks.Machine Learning,1995,20(3):273-297.
[38] 周绍磊,廖剑,史贤俊.RBF-SVM的核参数选择方法及其在故障诊断中的应用.电子测量与仪器学报,2014,28(3):240-246.
[39] 张睿,马建文.支持向量机在遥感数据分类中的应用新进展.地球科学进展,2009,24(5):555-562.
[40] Chang C C,Lin J C.LIBSVM:a library for support vector machines.ACM Transactions on Intelligent Systems and Technology,2011,2(3):27.
[41] 刘艳慧,蔡宗磊,包妮沙,刘善军.基于无人机大样方草地植被覆盖度及生物量估算方法研究.生态环境学报,2018,27(11):2023-2032.
[42] 宋军伟,张友静,李鑫川,杨文治.基于GF-1与Landsat-8影像的土地覆盖分类比较.地理科学进展,2016,35(2):255-263.
[43] Hansen M C,Potapov P V,Moore R,Hancher M,Turubanova S A,Tyukavina A,Thau D,Stehman S V,Goetz S J,Loveland T R,Kommareddy A,Egorov A,Chini L,Justice C O,Townshend J R G.High-resolution global maps of 21st-century forest cover change.Science,2013,342(6160):850-853.
[2] 姚雪玲,姜丽娜,李龙,王锋,吴波,郭秀江.浑善达克沙地6种灌木生物量模拟.生态学报,2019,39(3):905-912.
[3] 李苗苗.植被覆盖度的遥感估算方法研究[D].北京:中国科学院研究生院(遥感应用研究所),2003.
[4] Naidoo L,Mathieu R,Main R,Wessels K,Asner G P.L-band synthetic aperture radar imagery performs better than optical datasets at retrieving woody fractional cover in deciduous,dry savannahs.International Journal of Applied Earth Observation and Geoinformation,2016,52:54-64.
[5] Liu X,Liu H Y,Qiu S,Wu X C,Tian Y H,Hao Q.An improved estimation of regional fractional woody/herbaceous cover using combined satellite data and high-quality training samples.Remote Sensing,2017,9(1):32.
[6] Suess S,van der Linden S,Okujeni A,Griffiths P,Leit?o P,Schwieder M,Hostert P.Characterizing 32 years of shrub cover dynamics in southern Portugal using annual Landsat composites and machine learning regression modeling.Remote Sensing of Environment,2018,219:353-364.
[7] Baumann M,Levers C,Macchi L,Bluhm H,Waske B,Gasparri N I,Kuemmerle T.Mapping continuous fields of tree and shrub cover across the Gran Chaco using Landsat 8 and sentinel-1 data.Remote Sensing of Environment,2018,216:201-211.
[8] Ge J,Meng B P,Liang T G,Feng Q S,Gao J L,Yang S X,Huang X D,Xie H J.Modeling alpine grassland cover based on MODIS data and support vector machine regression in the headwater region of the Huanghe River,China.Remote Sensing of Environment,2018,218:162-173.
[9] 张瑾,李晓松,吴炳方.基于分类回归树的密云水库上游森林覆盖度遥感估算.遥感技术与应用,2014,29(3):394-400.
[10] Huang H B,Chen Y L,Clinton N,Wang J,Wang X Y,Liu C X,Gong P,Yang J,Bai Y Q,Zheng Y M,Zhu Y L.Mapping major land cover dynamics in Beijing using all Landsat images in google earth engine.Remote Sensing of Environment,2017,202:166-176.
[11] 高志海,李增元,魏怀东,丁锋,丁国栋.干旱地区植被指数(VI)的适宜性研究.中国沙漠,2006,26(2):243-248.
[12] Li X S,Zheng G X,Wang J Y,Ji C C,Sun B,Gao Z H.Comparison of methods for estimating fractional cover of photosynthetic and non-photosynthetic vegetation in the Otindag sandy land using GF-1 wide-field view data.Remote Sensing,2016,8(10):800.
[13] Bastin J,Berrahmouni N,Grainger A,Maniatis D,Mollicone D,Moore R,Patriarca C,Picard N,Sparrow B,Maria Abraham E,Aloui K,Atesoglu A,Attore F,Bassüllü C,Bey A,Garzuglia M,G.García-Montero L,Groot N,Guerin G,Laestadius L,Lowe A J,Mamane B,Marchi G,Patterson P,Rezende M,Ricci S,Salcedo I,Sanchez-Paus Diaz A,Stolle F,Surappaeva V,Castro R.The extent of forest in dryland biomes.Science,2017,356(6338):635-638.
[14] Sexton J O,Song X P,Feng M,Noojipady P,Anand A,Huang C Q,Kim D H,Collins K,Channan S,DiMiceli C,Townshend J.Global,30-m resolution continuous fields of tree cover:Landsat-based rescaling of MODIS vegetation continuous fields with Lidar-based estimates of error.International Journal of Digital Earth,2013,6(5):427-448.
[15] Herrmann S M,Wickhorst A J,Marsh S E.Estimation of tree cover in an agricultural parkland of Senegal using rule-based regression tree modeling.Remote Sensing,2013,5(10):4900-4918.
[16] Brandt M,Hiernaux P,Tagesson T,Verger A,Rasmussen K,Diouf A A,Mbow C,Mougin E,Fensholt R.Woody plant cover estimation in drylands from earth observation based seasonal metrics.Remote Sensing of Environment,2016,172:28-38.
[17] Higginbottom T P,Symeonakis E,Meyer H,Van der Linden S.Mapping fractional woody cover in semi-arid savannahs using multi-seasonal composites from Landsat data.ISPRS Journal of Photogrammetry and Remote Sensing,2018,139:88-102.
[18] 李晓松,李增元,吴波,高志海,白黎娜,王琫瑜.基于光谱混合分析的毛乌素沙地油蒿群落覆盖度提取.遥感学报,2007,11(6):923-930.
[19] 李晓松,高志海,李增元,白黎娜,王琫瑜.基于高光谱混合像元分解的干旱地区稀疏植被覆盖度估测.应用生态学报,2010,21(1):152-158.
[20] 吴俊君,高志海,李增元,王红岩,庞勇,孙斌,李长龙,李绪志,张九星.基于天宫一号高光谱数据的荒漠化地区稀疏植被参量估测(英文).光谱学与光谱分析,2014,34(3):751-756.
[21] Gorelick N,Hancher M,Dixon M,Ilyushchenko S,Thau D,Moore D.Google earth engine:planetary-scale geospatial analysis for everyone.Remote Sensing of Environment,2017,202:18-27.
[22] Langner A,Miettinen J,Kukkonen M,Vancutsem C,Simonetti D,Vieilledent G,Verhegghen A,Gallego J,Stibig H J.Towards operational monitoring of forest canopy disturbance in evergreen rain forests:a test case in continental Southeast Asia.Remote Sensing,2018,10(4):544.
[23] Dong J W,Xiao X G,Menarguez M A,Zhang G L,Qin Y W,Thau D,Biradar C,Moore III B.Mapping paddy rice planting area in Northeastern Asia with Landsat 8 images,phenology-based algorithm and google earth engine.Remote Sensing of Environment,2016,185:142-154.
[24] Xiong J,Thenkabail P S,Tilton J C,Gumma M K,Teluguntla P,Oliphant A,Congalton R G,Yadav K,Gorelick N.Nominal 30-m cropland extent map of continental Africa by integrating pixel-based and object-based algorithms using sentinel-2 and landsat-8 data on google earth engine.Remote Sensing,2017,9(10):1065.
[25] Parastatidis D,Mitraka Z,Chrysoulakis N,Abrams M.Online global land surface temperature estimation from Landsat.Remote Sensing,2017,9(12):1208.
[26] Zhou D J,Zhao X,Hu H F,Shen H H,Fang J Y.Long-term vegetation changes in the four mega-sandy lands in Inner Mongolia,China.Landscape Ecology,2015,30(9):1613-1626.
[27] 吴波,慈龙骏.毛乌素沙地景观格局变化研究.生态学报,2001,21(2):191-196.
[28] 冯颖.毛乌素沙地植被盖度变化及其对气候变化的响应[D].北京:北京林业大学,2015.
[29] 张新时.毛乌素沙地的生态背景及其草地建设的原则与优化模式.植物生态学报,1994,18(1):1-16.
[30] 屠志方,李梦先,孙涛.第五次全国荒漠化和沙化监测结果及分析.林业资源管理,2016,(1):1-5,13-13.
[31] Abatzoglou J T,Dobrowski S Z,Parks S A,Hegewisch K C.TerraClimate,a high-resolution global dataset of monthly climate and climatic water balance from 1958-2015.Scientific Data,2018,5:170191.
[32] 孙斌,李增元,郭中,高志海,王琫瑜.高分一号与Landsat TM数据估算稀疏植被信息对比.遥感信息,2015,30(5):48-56.
[33] Huete A R.A soil-adjusted vegetation index (SAVI).Remote Sensing of Environment,1988,25(3):295-309.
[34] Bey A,Sánchez-Paus Díaz A,Maniatis D,Marchi G,Mollicone D,Ricci S,Bastin J F,Moore R,Federici S,Rezende M,Patriarca,C,Turia R,Gamoga G,Abe H,Kaidong E,Micel G.Collect earth:land use and land cover assessment through augmented visual interpretation.Remote Sensing,2016,8(10):807.
[35] Breiman L,Friedman J H,Olshen R A and Stone C J.Classification and Regression Trees.Belmont,Calif.:Wadsworth International Group,1984:1-358.
[36] 赵萍,傅云飞,郑刘根,冯学智,Satyanarayana B.基于分类回归树分析的遥感影像土地利用/覆被分类研究.遥感学报,2005,9(6):708-716.
[37] Cortes C,Vapnik V.Support-vector networks.Machine Learning,1995,20(3):273-297.
[38] 周绍磊,廖剑,史贤俊.RBF-SVM的核参数选择方法及其在故障诊断中的应用.电子测量与仪器学报,2014,28(3):240-246.
[39] 张睿,马建文.支持向量机在遥感数据分类中的应用新进展.地球科学进展,2009,24(5):555-562.
[40] Chang C C,Lin J C.LIBSVM:a library for support vector machines.ACM Transactions on Intelligent Systems and Technology,2011,2(3):27.
[41] 刘艳慧,蔡宗磊,包妮沙,刘善军.基于无人机大样方草地植被覆盖度及生物量估算方法研究.生态环境学报,2018,27(11):2023-2032.
[42] 宋军伟,张友静,李鑫川,杨文治.基于GF-1与Landsat-8影像的土地覆盖分类比较.地理科学进展,2016,35(2):255-263.
[43] Hansen M C,Potapov P V,Moore R,Hancher M,Turubanova S A,Tyukavina A,Thau D,Stehman S V,Goetz S J,Loveland T R,Kommareddy A,Egorov A,Chini L,Justice C O,Townshend J R G.High-resolution global maps of 21st-century forest cover change.Science,2013,342(6160):850-853.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |