基于广义回归神经网络的京津冀地区土壤湿度遥感逐日估算研究

doi:10.12082/dqxxkx.2021.200149

地球信息科学学报 ›› 2021, Vol. 23 ›› Issue (4): 749-761.doi: 10.12082/dqxxkx.2021.200149

基于广义回归神经网络的京津冀地区土壤湿度遥感逐日估算研究

邓雅文¹(), 凌子燕²^,^*(), 孙娜¹, 吕金霞¹^,³

1.环境遥感与数字城市北京市重点实验室,北京师范大学地理科学学部,北京 100875
2.北部湾环境演变与资源利用教育部重点实验室,南宁师范大学地理科学与规划学院,南宁 530001
3.遥感科学国家重点实验室,北京师范大学地理科学学部,北京 100875

收稿日期:2020-03-27 修回日期:2020-06-21 出版日期:2021-04-25 发布日期:2021-06-25
通讯作者: 凌子燕
作者简介:邓雅文（1999— ）,女,湖南衡阳人,硕士生,研究方向为生态水文遥感。E-mail: 202021051199@mail.bnu.edu.cn
基金资助:
国家自然科学基金项目(41571077);国家重点研发计划项目(2016YFC0503002);环境遥感与数字城市北京重点实验室开放课题(12800-310430001)

Daily Estimation of Soil Moisture over Beijing-Tianjin-Hebei Region based on General Regression Neural Network Model

DENG Yawen¹(), LING Ziyan²^,^*(), SUN Na¹, LV Jinxia¹^,³

1. Beijing Key Laboratory for Remote Sensing of Environment and Digital Cities, Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China
2. Beibu Gulf Key Laboratory of Environment Change and Resources Use, School of Geography and Planning, Nanning Normal University, Nanning 530001, China
3. State Key Laboratory of Remote Sensing Science, Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China

Received:2020-03-27 Revised:2020-06-21 Online:2021-04-25 Published:2021-06-25
Contact: LING Ziyan
Supported by:
National Natural Science Foundation of China(41571077);National Key R&D Program of China(2016YFC0503002);Open subject of Beijing Key Laboratory for Remote Sensing of Environment and Digital Cities(12800-310430001)

摘要/Abstract

摘要：

土壤湿度是地表水热交换过程和水文循环中的一个关键组成部分,获取高时空分辨率的土壤湿度数据一直是当前研究的热点。SMAP（Soil Moisture Passive and Active）主被动微波土壤湿度产品的精度高,但存在着空间分辨率低和时间分辨率缺失的问题,这限制了其在区域尺度上的应用,为解决这一问题得到更高时空分辨率的土壤湿度产品,本文利用广义回归神经网络模型（GRNN）模拟了MODIS地表温度、反射率、植被指数光学/热红外遥感数据以及高程、坡度、坡向、经纬度数据与SMAP土壤湿度的关系,从而将京津冀地区SMAP L2土壤湿度产品的时间分辨率由不连续（4~20 d）提升至1 d,空间分辨率由3 km提升至1 km,并扩展其在京津冀地区的空间覆盖范围。研究发现:① GRNN模型总体验证结果表明土壤湿度估算值与SMAP原始值的相关性较高（r=0.7392）,均方根误差（RMSE）为0.0757 cm3/cm3;② 不同季节典型日期的GRNN模型估算结果精度相差较大,春季处的相关性相比其他季节最低,精度相对较高（r_春=0.6152,RMSE_春=0.0653cm3/cm3）,秋季和夏季土壤湿度估算精度较为接近（r_夏=0.6957,r_秋=0.7053,RMSE_夏=0.0754cm3/cm3,RMSE_秋=0.0694cm3/cm3）,冬季的估算精度最高（r_冬=0.8214,RMSE_冬=0.0367cm3/cm3）;③ 2016年京津冀夏秋季节的土壤湿度较其他季节要显著提高,空间分布上坝上高原区域较低,而沿海地区的土壤湿度明显较高。本研究对京津冀地区的生态水文、气候预测以及干旱监测等应用领域具有重要价值。

关键词: 遥感, SMAP, 土壤湿度, 反演算法, 降尺度, 广义回归神经网络模型（GRNN）, 京津冀, 机器学习

Abstract:

Surface Soil Moisture (SM) plays an important role in the land-atmosphere interaction and hydrological cycle. Low spatiotemporal resolution (i.e., 25~40 km and 2~3 days) microwave-based SM products such as the Soil Moisture and Ocean Salinity (SMOS) and the Advanced Microwave Scanning Radiometer for EOS (AMSR-E) limit their application in regional scale studies. The Soil Moisture Active Passive (SMAP) and Copernicus Sentinel 1A/B microwave active-passive surface soil moisture product (L2_SM_SP) has a higher spatial resolution (3 km), but its temporal resolution is coarse from 4 to 20 days due to the narrow overlapped swath width. In this study, we developed a machine learning algorithm using the General Regression Neural Network (GRNN) to improve the spatiotemporal resolution of the L2_SM_SP product based on multi-source remote sensing data. Land Surface Temperature (LST), Multi-band Reflectance, Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Elevation, Slope, Longitude (Lon), and Latitude (Lat) were selected as input variables to simulate the L2_SM_SP soil moisture in GRNN model. Results show that: (1) GRNN-estimated soil moisture and the original estimates of L2_SM_SP were strongly correlated (r=0.7392, RMSE=0.0757 cm³/cm³); (2) the correlation between GRNN estimates and original L2_SM_SP product at typical dates of different seasons varied a lot. The correlation in spring was the lowest (r_Spr=0.6152, RMSE_Spr=0.0653 cm3/cm3). While the correlation in winter was the strongest (r_Win=0.8214, and RMSE_Win=0.0367 cm³/cm³). The correlation in summer and autumn was close to each other (r_Sum=0.6957, r_Aut=0.7053, RMSE_Sum=0.0754 cm3/cm3, and RMSE_Aut=0.0694 cm3/cm3); and (3) in 2016, the soil moisture in summer and autumn of the study area was significantly higher than that that in other seasons. In terms of spatial distribution, the soil moisture in the Bashang plateau area was low, while the soil moisture along coastal areas was obviously higher. In this study, we successfully improved the spatiotemporal resolution of L2_SM_SP product over Beijing-Tianjin-Hebei region from 3 km, and 4~20 days to 1 km, and 1 day. Its spatial coverage was also extended. The improved soil moisture product is of great significance for future eco-hydrological assessment, climate prediction, and drought monitoring in Beijing-Tianjin-Hebei region.

Key words: remote sensing, Soil Moisture Active Passive（SMAP）, soil moisture, retrieval algorithms, downscaling, GRNN, Beijing-Tianjin-Hebei Region, machine learning

邓雅文, 凌子燕, 孙娜, 吕金霞. 基于广义回归神经网络的京津冀地区土壤湿度遥感逐日估算研究[J]. 地球信息科学学报, 2021, 23(4): 749-761.DOI:10.12082/dqxxkx.2021.200149

DENG Yawen, LING Ziyan, SUN Na, LV Jinxia. Daily Estimation of Soil Moisture over Beijing-Tianjin-Hebei Region based on General Regression Neural Network Model[J]. Journal of Geo-information Science, 2021, 23(4): 749-761.DOI:10.12082/dqxxkx.2021.200149

图/表 11

图1

图2

表1

图3

图4

图5

表2

表3

表4

图6

图7

参考文献 34

[1]	Koster R D, Dirmeyer P A, Guo Z, et al. Regions of strong coupling between soil moisture and precipitation[J]. Science, 2004,305(5687):1138-1140. doi: 10.1126/science.1100217 pmid: 15326351
[2]	Chen F, Crow W T, Starks P J, et al. Improving hydrologic predictions of a catchment model via assimilation of surface soil moisture[J]. Advances in Water Resources, 2011,34(4):526-536. doi: 10.1016/j.advwatres.2011.01.011
[3]	Li Y, Grimaldi S, Pauwels V R N, et al. Hydrologic model calibration using remotely sensed soil moisture and discharge measurements: The impact on predictions at gauged and ungauged locations[J]. Journal of Hydrology, 2018,557:897-909. doi: 10.1016/j.jhydrol.2018.01.013
[4]	Das K, Paul P K, Dobesova Z. Present status of soil moisture estimation by microwave remote sensing[J]. Cogent Geoscience. 2015,1(1):1084669. doi: 10.1080/23312041.2015.1084669
[5]	Gallego Elvira B, Taylor C M, Harris P P, et al. Global observational diagnosis of soil moisture control on the land surface energy balance[J]. Geophysical Research Letters, 2016,43(6):2623-2631. doi: 10.1002/grl.v43.6
[6]	丁旭, 赖欣, 范广洲, 等. 中国不同气候区土壤湿度特征及其气候响应[J]. 高原山地气象研究, 2016,36(4):28-35.
	[ Ding X, Lai X, Fan G Z, et al. Soil moisture characteristics and climate response of different climatic regional in China[J]. Plateau and Mountain Meteorology Research, 2016,36(4):28-35.]
[7]	Engda T A, Kelleners T J. Soil moisture-based drought monitoring at different time scales: A case study for the U.S. Great Plains[J]. Jawra Journal of the American Water Resources Association, 2016,52(1):77-88. doi: 10.1111/1752-1688.12367
[8]	Kim S, Paik K, Johnson, et al. Building a flood-warning framework for ungauged locations using low resolution, open-access remotely sensed surface soil moisture, precipitation, soil, and topographic information[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018,11(2):375-387. doi: 10.1109/JSTARS.2018.2790409
[9]	韩帅, 师春香, 林泓锦, 等. CLDAS土壤湿度业务产品的干旱监测应用[J]. 冰川冻土, 2015,37(2):446-453. doi: 10.7522/j.issn.1000-0240.2015.0051
	[ Han S, ShiC X, LinH J, et al. The CLDAS soil moisture operation products applied to monitor soil drought[J]. Journal of Glaciology and Geocryology, 2015,37(2):446-453. ]
[10]	习阿幸, 刘志辉, 卢文君. 干旱区季节性冻土冻融状况及对融雪径流的影响[J]. 水土保持研究, 2016,23(2):333-339.
	[ Xi A X, Liu Z H, Lu W J. Processes of seasonal frozen soil freezing-thawing and impact on snowmelt runoff in arid area[J]. Research of Soil and Water Conservation, 2016,23(2):333-339. ]
[11]	刘荣华, 张珂, 晁丽君, 等. 基于多源卫星观测的中国土壤湿度时空特征分析[J]. 水科学进展, 2017,28(4):479-487.
	[ Liu R H, Zhang K, Chao L J, et al. Analysis of spatiotemporal characteristics of surface soil moisture across China based on multi-satellite observations[J]. Advances in Water Science, 2017,28(4):479-487. ]
[12]	孔金玲, 李菁菁, 甄珮珮, 等. 微波与光学遥感协同反演旱区地表土壤湿度研究[J]. 地球信息科学学报, 2016,18(6):857-863. doi: 10.3724/SP.J.1047.2016.00857
	[ Kong J L, Li J J, Zhen P P, et al. Inversion of soil moisture in arid area based on microwave and optical remote sensing data[J]. Journal of Geo-information Science, 2016,18(6):857-863. ]
[13]	陈书林, 刘元波, 温作民. 卫星遥感反演土壤湿度研究综述[J]. 地球科学进展, 2012(11):1192-1203.
	[ Chen S L, Liu Y B, Wen Z M. Satellite retrieval of soil moisture: An overview[J]. Advances in Earth Science, 2012,27(11):1192-1203. ]
[14]	Petropoulos G P, Ireland G, Barrett B. Surface soil moisture retrievals from remote sensing: current status, products & future trends[J]. Physics and Chemistry of the Earth, Parts A/B/C. 2015,83-84:36-56.
[15]	Entekhabi D, Njoku E G, O'neil P E, et al. The Soil Moisture Active Passive (SMAP) mission[J]. Proceedings of the IEEE, 2010,98(5):704-716. doi: 10.1109/JPROC.2010.2043918
[16]	Njoku E G, Jackson T J, Lakshmi V, et al. Soil moisture retrieval from AMSR-E[J]. IEEE Transactions on Geoscience and Remote Sensing, 2003,41(2):215-229. doi: 10.1109/TGRS.2002.808243
[17]	Wu Q, Liu H, Lei W, et al. Evaluation of AMSR2 soil moisture products over the contiguous United States using in situ data from the International Soil Moisture Network[J]. International Journal of Applied Earth Observation & Geoinformation, 2016,45(Part B):187-199. doi: 10.1016/j.jag.2015.10.011
[18]	Kerr Y H, Waldteufel P, Richaume P, et al. The SMOS soil moisture retrieval algorithm[J]. IEEE Transactions on Geoscience and Remote Sensing, 2012,50(5):1384-1403. doi: 10.1109/TGRS.2012.2184548
[19]	贾炳浩, 师春香, 孙帅, 等. 基于风云卫星的土壤湿度数据同化研究:第35届中国气象学会年会[ C].中国安徽合肥, 2018.
	[ Jia B H, Shi C X, Sun S, et al. Study on soil moisture data assimilation based on FY satellite: The 35^th annual meeting of Chinese Meteorological Society[C]. Hefei, Anhui, China, 2018. ]
[20]	Cui Y, Chen X, Xiong W, et al. A soil moisture spatial and temporal resolution improving algorithm based on multi-Source remote sensing data and GRNN model[J]. Remote Sensing, 2020,12(3):455. doi: 10.3390/rs12030455
[21]	Merlin O, Chehbouni A G, Kerr Y H, et al. A combined modeling and multispectral/multiresolution remote sensing approach for disaggregation of surface soil moisture: application to SMOS configuration[J]. IEEE Transactions on Geoscience and Remote Sensing, 2005,43(9):2036-2050. doi: 10.1109/TGRS.2005.853192
[22]	Wagner W, Dorigo W, De Jeu R, et al. Fusion of active and passive microwave observations to create an essential climate variable data record on soil moisture[J]. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2012,7:315-321.
[23]	贾艳昌, 谢谟文, 姜红涛. 全球36 km格网土壤湿度逐日估算[J]. 地球信息科学学报, 2017,19(6):854-860. doi: 10.3724/SP.J.1047.2017.00854
	[ Jia Y C, Xie M W, Jiang H T. Daily estimate of global 36 km grid soil moisture[J]. Journal of Geo-information Science, 2017,19(6):854-860. ]
[24]	Said S, Kothyari U C, Arora M K. ANN-based soil moisture retrieval over bare and vegetated areas using ERS-2 SAR data[J]. Journal of Hydrologic Engineering, 2008,13(6):461-475. doi: 10.1061/(ASCE)1084-0699(2008)13:6(461)
[25]	Yuan Q, Xu H, Li T, et al. Estimating surface soil moisture from satellite observations using a generalized regression neural network trained on sparse ground-based measurements in the continental U.S[J]. Journal of Hydrology, 2020,580:124351. doi: 10.1016/j.jhydrol.2019.124351
[26]	Das N N, Entekhabi D, Dunbar R S, et al. The SMAP and Copernicus Sentinel 1A/B microwave active-passive high resolution surface soil moisture product[J]. Remote Sensing of Environment, 2019,233:111380. doi: 10.1016/j.rse.2019.111380
[27]	于占江, 杨鹏. 近50年京津冀气候变化及其对土壤湿度的影响[J]. 贵州农业科学, 2019,47(2):144-149.
	[ Yu Z J, Yang P. Effect of climatic change on soil moisture in Beijing-Tianjin-Hebei region in nearly 50 years[J]. Guizhou Agricultural Sciences, 2019,47(2):144-149. ]
[28]	Chan S K, Bindlish R, E. O'neill P, et al. Assessment of the SMAP passive soil moisture product[J]. IEEE Transactionson Geoscience and Remote Sensing, 2016,54(8):4994-5007.
[29]	李鹏飞, 刘文军, 赵昕奕, 等. 京津冀区域近50年气温、降水与潜在蒸散量变化分析[J]. 干旱区资源与环境, 2015,29(3):137-143.
	[ Li P F, Liu W J, Zhao X Y, et al. The changes of atmospheric temperature,precipitation and potential evapotranspiration in Beijing -Tianjin-Hebei region in recent 50 years[J]. Journal of Arid Land Resources and Environment, 2015,29(3):137-143. ]
[30]	张悦, 沈润平, 彭露露, 等. 基于重建MODIS无云数据反演京津冀地区土壤湿度[J]. 江苏农业科学, 2016,44(12):375-378.
	[ Zhang Yue, Shen R P, Peng L L, et al. Soil moisture monitoring based on reconstruction of MODIS data in Beijing-Tianjin-Hebei region[J]. Jiangsu Agricultural Sciences, 2016,44(12):375-378. ]
[31]	Specht D F. A general regression neural network[J]. IEEE Transactions on Neural Networks, 1991,2(6):568-576. doi: 10.1109/72.97934 pmid: 18282872
[32]	Ali I, Greifeneder F, Stamenkovic J, et al. Review of machine learning approaches for biomass and soil moisture retrievals from remote sensing data[J]. Remote Sensing, 2015,7(12):16398-16421. doi: 10.3390/rs71215841
[33]	Li W, Yang X, Li H, et al. Hybrid forecasting approach based on GRNN neural network and SVR machine for electricity demand forecasting[J]. Energies, 2017,10(1):44. doi: 10.3390/en10010044
[34]	Rodriguez J D, Perez A, Lozano J A. Sensitivity analysis of k-fold cross validation in prediction error estimation[J]. IEEE Trans Pattern Anal Mach Intell, 2010,32(3):569-575. doi: 10.1109/TPAMI.2009.187 pmid: 20075479

数据类型	数据名称	时间分辨率/d	空间分辨率	产品类型	时间节点	数据位置
MODIS	MOD13A2	16	1 km	植被指数与反射率	2016-04-07	h26v04, h26v05
	MOD11A1	1	1 km	地表温度	2016-07-05	h27v04, h27v05
SMAP	SPL2SMAP_S	4~20	3 km	土壤湿度	2016-09-28	116°E—118°E, 37°N—44°N
					2016-02-06
高程	DEM	-	30 m	高程	-	-

主成分	特征值	方差贡献率	累计贡献率
1	5.8544	0.488	0.488
2	1.8477	0.154	0.642
3	1.1322	0.094	0.736
4	0.9806	0.082	0.818
5	0.7982	0.067	0.884
6	0.5754	0.048	0.932
7	0.4060	0.034	0.966
8	0.2731	0.023	0.989
9	0.0982	0.008	0.997
10	0.0257	0.002	0.999
11	0.0079	0.001	1.000
12	0.0006	0.000	1.000

		r	RMSE/(cm³/cm³)	bias	ubRMSE	样本量/个	训练时间/s
GRNN降维前扩散因子为0.09	训练精度	0.7566	0.0736	-0.0011	0.0736	48000
	验证精度	0.7322	0.0766	-0.0005	0.0765	12000	1007.75
	总体精度	0.7392	0.0757	-0.0007	0.0757	60000
GRNNPCA降维后扩散因子为0.04	训练精度	0.7032	0.0777	-0.0002	0.0777	48000
	验证精度	0.7140	0.0787	0.0012	0.0787	12000	768.02
	总体精度	0.7110	0.0784	0.0008	0.0784	60000

时间	指标	训练精度	验证精度	总体精度	时间	指标	训练精度	验证精度	总体精度
春季（2016-04-07）	r	0.6300^**	0.5886^**	0.6152^**	秋季（2016-09-28）	r	0.7467^**	0.6900^**	0.7053^**
	RMSE/(cm³/cm³)	0.0636	0.0669	0.0653		RMSE/(cm³/cm³)	0.0658	0.0708	0.0694
	bias	0.0009	0.0003	0.0005		bias	0.0004	0.0008	0.0007
	ubRMSE	0.0636	0.0659	0.0653		ubRMSE	0.0658	0.0708	0.0694
	样本量/个	32 000	8000	40 000		样本量/个	32 000	8000	40 000
夏季（2016-07-05）	r	0.7277^**	0.6843^**	0.6957^**	冬季（2016-12-14）	r	0.8064^**	0.8308^**	0.8214^**
	RMSE/(cm³/cm³)	0.0715	0.0769	0.0754		RMSE/(cm³/cm³)	0.0348	0.0374	0.0367
	bias	0.0018	0.0013	0.0014		bias	-0.0019	-0.0016	-0.0017
	ubRMSE	0.0715	0.0769	0.0754		ubRMSE	0.0348	0.0374	0.0367
	样本量/个	32 000	8000	40 000		样本量/个	16 000	4000	20 000

基于广义回归神经网络的京津冀地区土壤湿度遥感逐日估算研究

Daily Estimation of Soil Moisture over Beijing-Tianjin-Hebei Region based on General Regression Neural Network Model

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 34

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	唐璎, 刘正军, 杨懿, 顾海燕, 杨树文. 基于特征增强和ELU的神经网络建筑物提取研究[J]. 地球信息科学学报, 2021, 23(4): 692-709.
[2]	张志恒, 章超, 孟麟, 唐凯, 朱红春. 改进DINEOF算法在中国渤海叶绿素a遥感缺失数据重构中的应用研究[J]. 地球信息科学学报, 2021, 23(4): 737-748.
[3]	李广洋, 寇卫利, 陈帮乾, 代飞, 强振平, 吴超. 多核学习算法及其在高光谱图像分类中的应用研究进展[J]. 地球信息科学学报, 2021, 23(3): 492-504.
[4]	朱盼盼, 李帅朋, 张立强, 李洋. 基于多任务学习的高分辨率遥感影像建筑提取[J]. 地球信息科学学报, 2021, 23(3): 514-523.
[5]	郭紫甜, 王春梅, 刘欣, 庞国伟, 朱梦阳, 王晋卿. 基于小流域抽样单元的中国FROM-GLC30数据精度评价[J]. 地球信息科学学报, 2021, 23(3): 524-535.
[6]	朱婉雪, 孙志刚, 李彬彬, 杨婷, 刘振, 彭金榜, 朱康莹, 李仕冀, 娄金勇, 侯瑞星, 李静, 于武江, 王永利, 张峰, 刘向冶, 胡华浪, 欧阳竹. 基于无人机遥感的滨海盐碱地土壤空间异质性分析与作物光谱指数响应胁迫诊断[J]. 地球信息科学学报, 2021, 23(3): 536-549.
[7]	帅艳民, 马现伟, 曲歌, 邵聪颖, 刘涛, 刘守民, 黄华兵, 谷玲霄, 拉提帕·吐尔汗江, 梁继, 李玲. 协同多时相波谱特征的不透水面信息级联提取[J]. 地球信息科学学报, 2021, 23(1): 171-186.
[8]	王志华, 杨晓梅, 周成虎. 面向遥感大数据的地学知识图谱构想[J]. 地球信息科学学报, 2021, 23(1): 16-28.
[9]	刘明杰, 徐卓揆, 郜允兵, 杨晶, 潘瑜春, 高秉博, 周艳兵, 周万鹏, 王凌. 基于机器学习的稀疏样本下的土壤有机质估算方法[J]. 地球信息科学学报, 2020, 22(9): 1799-1813.
[10]	王艳杰, 王卷乐, 魏海硕, Altansukh Ochir, Davaadorj Davaasuren, Sonomdagva Chonokhuu. 基于稀疏样点的蒙古国产草量估算方法研究[J]. 地球信息科学学报, 2020, 22(9): 1814-1822.
[11]	陈辉, 厉青, 王中挺, 马鹏飞, 李营, 赵爱梅. 一种基于FY3D/MERSI2的AOD遥感反演方法[J]. 地球信息科学学报, 2020, 22(9): 1887-1896.
[12]	陈昂, 杨秀春, 徐斌, 金云翔, 张文博, 郭剑, 邢晓语, 杨东. 基于面向对象与深度学习的榆树疏林识别方法研究[J]. 地球信息科学学报, 2020, 22(9): 1897-1909.
[13]	蒲东川, 王桂周, 张兆明, 牛雪峰, 何国金, 龙腾飞, 尹然宇, 江威, 孙嘉悦. 基于独立成分分析和随机森林算法的城镇用地提取研究[J]. 地球信息科学学报, 2020, 22(8): 1597-1606.
[14]	刘新, 赵宁, 郭金运, 郭斌. 基于LSTM神经网络的青藏高原月降水量预测[J]. 地球信息科学学报, 2020, 22(8): 1617-1629.
[15]	李婉, 牛陆, 陈虹, 吴骅. 基于随机森林算法的地表温度鲁棒降尺度方法[J]. 地球信息科学学报, 2020, 22(8): 1666-1678.