不同尺度的土壤含水量主被动微波联合反演方法研究

doi:10.13745/j.esf.sf.2023.12.32

地学前缘 ›› 2024, Vol. 31 ›› Issue (2): 42-53.DOI: 10.13745/j.esf.sf.2023.12.32

• 场地土壤污染机制与风险管控 • 上一篇下一篇

不同尺度的土壤含水量主被动微波联合反演方法研究

刘奇鑫¹^,²(), 顾行发¹^,²^,³, 王春梅¹, 杨健¹, 占玉林¹^,^*()

1.中国科学院空天信息创新研究院, 北京 100094
2.中国科学院大学, 北京 100049
3.北华航天工业学院遥感信息工程学院, 河北廊坊 065000

收稿日期:2023-12-13 修回日期:2024-01-02 出版日期:2024-03-25 发布日期:2024-04-18
通讯作者: *占玉林(1975—),男,研究员,主要从事生态环境遥感、遥感产品时空一致性研究工作。E-mail: zhanyl@aircas.ac.cn
作者简介:刘奇鑫(1993—),男,博士研究生,研究方向为微波遥感土壤含水量反演。E-mail: liuqx@radi.ac.cn
基金资助:
国家重点研发计划项目(2021YFE0117300);国家民用空间基础设施陆地观测卫星共性应用支撑平台项目(2017-000052-73-01-001735);国家国防科技工业局高分辨率对地观测系统重大专项(30-Y60B01-9003-22/23)

Soil moisture retrieval on both active and passive microwave data scales

LIU Qixin¹^,²(), GU Xingfa¹^,²^,³, WANG Chunmei¹, YANG Jian¹, ZHAN Yulin¹^,^*()

1. Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China
2. University of Chinese Academy of Sciences, Beijing 100049, China
3. School of Remote Sensing and Information Engineering, North China Institute of Aerospace Engineering, Langfang 065000, China

Received:2023-12-13 Revised:2024-01-02 Online:2024-03-25 Published:2024-04-18

摘要/Abstract

摘要：

土壤含水量是水文、农业和气象等领域的关键参数,而微波遥感是目前监测土壤含水量最有效的手段之一。本文利用主动微波与被动微波数据,结合其他多源遥感数据,运用随机森林算法分别在主动微波数据分辨率尺度和被动微波数据分辨率尺度下完成主被动微波数据的土壤含水量联合反演。首先对被动微波尺度的地表覆盖类型与归一化植被指数(NDVI)参数进行空间分辨率优化,再利用回归ReliefF方法对两种尺度所用的输入变量的重要性进行评估,并对输入变量进行优选,最后对比主被动微波数据土壤含水量联合反演和单独利用主动/被动微波数据进行反演的精度,分析主被动微波联合反演方法的有效性。结果表明:在主动微波尺度,主被动微波联合反演的精度相比单独利用主动微波数据反演的精度有所提升,相关系数r由0.691升至0.744,RMSE由0.084 8 cm³/cm³降至0.079 6 cm³/cm³;在被动微波尺度,主被动微波联合反演的精度反而比单独利用被动微波数据反演的精度更低,相关系数r由0.944变为0.939,RMSE由0.043 5 cm³/cm³变为0.045 1 cm³/cm³。因此在主动微波尺度更适合进行主被动微波的联合反演。

关键词: 土壤含水量, 微波遥感, 联合反演, 随机森林, 空间优化, 重要性评估

Abstract:

Soil moisture is a critical parameter in the field of hydrology, agriculture and meteorology, and microwave remote sensing is one of the most effective methods for soil moisture detection. This study uses active and passive microwave data and other multi-source remote sensing data and applies Random forest algorithm to perform soil moisture retrieval on both active and passive microwave data scales. First, on passive microwave data scale, the parameters for land cover and normalized difference vegetation index (NDVI) are spatially optimized. Second, ReliefF method is used for evaluating the importance of input parameters and parameter selection. Last, results by using active and passive microwave data jointly or separately are compared to assess the retrial accuracy and effectiveness of the former method. It was found that on the scale of active microwave data, the joint retrieval method yielded results with higher accuracy (r=0.691, RMSE=0.0796) compared to using only active microwave data (r=0.744, RMSE=0.0848); however, on the scale of passive microwave data, the opposite was true (r=0.939, RMSE=0.0451 using joint data; r=0.944, RMSE=0.0435 using only passive microwave data). The joint retrieval method proved to be applicable on the scale of active microwave data.

Key words: soil moisture, microwave remote sensing, joint retrieval, random forest, spatial optimization, importance evaluation

中图分类号:

刘奇鑫, 顾行发, 王春梅, 杨健, 占玉林. 不同尺度的土壤含水量主被动微波联合反演方法研究[J]. 地学前缘, 2024, 31(2): 42-53.

LIU Qixin, GU Xingfa, WANG Chunmei, YANG Jian, ZHAN Yulin. Soil moisture retrieval on both active and passive microwave data scales[J]. Earth Science Frontiers, 2024, 31(2): 42-53.

图/表 13

图1 研究区与土壤含水量地面监测站点示意图

Fig.1 Schematic diagram of the study area and ISMN monitoring sites

表1 各土壤数据观测站网的站点数量与所得的日平均土壤含水量观测值总数

Table 1 The numbers of monitoring sites and values of daily averaged soil moisture observation in each soil moisture network

站网名称	站点数量	日平均土壤含水量观测值总数
ARM	14	25
COSMOS	15	28
FLUXNET-AMERIFLUX	5	29
iRON	7	19
PBO_H2O	104	270
RISMA	17	54
SCAN	143	454
SNOTEL	357	1 135
SOILSCAPE	15	26
USCRN	83	266

图2 随机森林结构示意图(据文献[38])

Fig.2 Schematic diagram of the structure of random forest. Adapted from [38].

表2 MODIS LC图层分类方案与重分类方案的对比

Table 2 Comparison between MODIS land cover classification scheme and the reclassification scheme

MODIS LC图层分类方案	重分类方案
Evergreen Needleleaf Forests(常绿针叶林)	森林
Evergreen Broadleaf Forests(常绿阔叶林)
Deciduous Needleleaf Forests(落叶针叶林)
Deciduous Broadleaf Forests(落叶阔叶林)
Mixed Forests(混交林)
Closed Shrublands(封闭灌丛)	灌木
Open Shrublands(开放灌丛)	灌木
Woody Savannas(热带稀树草原)	草地
Savannas(热带草原)
Grasslands(草地)
Croplands(农田)	农田
Cropland/Natural Vegetation Mosaics (农田/自然植被)	农田
Barren(裸土)	裸土
Permanent Wetlands(永久湿地,去除)	其他(去除)
Urban and Built-up Lands(城市与建筑物,去除)
Water Bodies(水体,去除)
Permanent Snow and Ice(永久冰雪,去除)

图3 “SMAP像元中典型地类的占比”和“SMAP像元中典型地类对应的NDVI平均值”计算方法图示

Fig.3 Illustration of calculating “percentages of typical land cover classes” and “average NDVIs corresponding to typical land cover classes”

表3 两种微波数据尺度下初步选择作为输入变量与输出变量的参数

Table 3 Input and output parameters initially selected in two different microwave data scale

数据尺度	输入变量	输出变量
主动微波尺度	σ_VH/VV,θ,NDVI,LST,elevation,slope,DOY	ISMN实测数据
被动微波尺度	T_bH/V,Perc_X,NDVI_X,LST,elevation,slope,DOY	ERA5-Land模拟数据

表4 主动微波尺度输入变量组合中各变量的重要性排序

Table 4 Ranking of the importance of input parameters in active microwave data scale

排序	输入变量
1	slope
2	elevation
3	NDVI
4	LST
5	$σ V H$
6	$σ V V$
7	DOY
8	θ

表4 主动微波尺度输入变量组合中各变量的重要性排序

Table 4 Ranking of the importance of input parameters in active microwave data scale

排序	输入变量
1	slope
2	elevation
3	NDVI
4	LST
5	$σ V H$
6	$σ V V$
7	DOY
8	θ

表5 被动微波尺度输入变量组合中各变量的重要性排序

Table 5 Ranking of the importance of input parameters in passive microwave data scale

排序	输入变量
1	elevation
2	DOY
3	LST
4	NDVI_G
5	T_bH
6	T_bV
7	Perc_F
8	slope
9	Perc_G
10	NDVI_F
11	Perc_C
12	NDVI_S
13	Perc_S
14	NDVI_B
15	NDVI_C
16	Perc_B

图4 将主动微波尺度各变量按重要性顺序依次加入随机森林模型得到的反演精度指标

Fig.4 Statistical metrics of soil moisture retrieval by adding parameters into the random forest model according to the importance ranking of active microwave data scale

图5 将被动微波尺度各变量按重要性顺序依次加入随机森林模型得到的反演精度指标

Fig.5 Statistical metrics of soil moisture retrieval by adding parameters into the random forest model according to the importance ranking of passive microwave data scale

表6 两种数据尺度主被动微波联合反演输入变量优选结果

Table 6 Optimized input parameters for joint retrieval of soil moisture on two microwave data scale

数据尺度	优选输入变量
主动微波尺度	slope,elevation,NDVI,LST,σ_VH,σ_VV,DOY
被动微波尺度	elevation,DOY,LST,NDVI_G,T_bH,T_bV,Perc_F,slope,Perc_G,NDVI_F

表7 主被动微波联合反演与单独利用主动微波/被动微波数据反演的精度评价指标对比

Table 7 Comparison of statistical metrics of joint retrieval of soil moisture using both active and passive microwave data and statistical metrics of soil moisture retrieval respectively using active/passive microwave data

数据尺度	输入变量组合	r	RMSE/(cm³·cm^-3)
主动微波尺度	slope,elevation,NDVI,LST,σ_VH,σ_VV,DOY	0.691	0.084 8
主动微波尺度	slope,elevation,NDVI,LST,σ_VH,σ_VV,DOY,T_bH/V	0.744	0.079 6
被动微波尺度	elevation,DOY,LST,NDVI_G,T_bH,T_bV,Perc_F,slope,Perc_G,NDVI_F	0.944	0.043 5
被动微波尺度	elevation,DOY,LST,NDVI_G,T_bH,T_bV,Perc_F,slope,Perc_G,NDVI_F,σ_VH/VV	0.939	0.045 1

表8 将输入变量组合中的新参数去掉后被动微波尺度下的联合反演与单独利用被动微波数据的反演精度

Table 8 Statistical metrics of joint retrieval of soil moisture using both active and passive microwave data and statistical metrics of soil moisture retrieval using passive microwave data after removing the new parameters from the inputs

数据尺度	输入变量组合	r	RMSE/(cm³·cm^-3)
被动微波尺度	elevation,DOY,LST,T_bH,T_bV,slope	0.929	0.048 3
被动微波尺度	elevation,DOY,LST,T_bH,T_bV,slope,σ_VH/VV	0.926	0.049 3

参考文献 41

[1]	DAI A G, TRENBERTH K E, QIAN T T. A global dataset of palmer drought severity index for 1870-2002: relationship with soil moisture and effects of surface warming[J]. Journal of Hydrometeorology, 2004, 5(6): 1117-1130.
[2]	LOEW A, STACKE T, DORIGO W, et al. Potential and limitations of multidecadal satellite soil moisture observations for selected climate model evaluation studies[J]. Hydrology and Earth System Sciences, 2013, 17(9): 3523-3542.
[3]	PETROPOULOS G P, IRELAND G, SRIVASTAVA P K, et al. An appraisal of the accuracy of operational soil moisture estimates from SMOS MIRAS using validated in situ observations acquired in a Mediterranean environment[J]. International Journal of Remote Sensing, 2014, 35(13): 5239-5250.
[4]	HIRSCHI M, SENEVIRATNE S I, SCH\AR C. Seasonal variations in terrestrial water storage for major midlatitude river basins[J]. Journal of Hydrometeorology, 2006, 7(1): 39.
[5]	ROBINSON D A, CAMPBELL C S, HOPMANS J W, et al. Soil moisture measurement for ecological and hydrological watershed-scale observatories: a review[J]. Vadose Zone Journal, 2008, 7(1): 358-389.
[6]	逄春浩. 土壤水分测定方法的新进展: TDR测定仪[J]. 干旱区资源与环境, 1994, 8(2): 69-76.
[7]	REYNOLDS S G. The gravimetric method of soil moisture determination Part I: a study of equipment, and methodological problems[J]. Journal of Hydrology, 1970, 11(3): 258-273.
[8]	JACKSON T J, BINDLISH R, COSH M H, et al. Validation of soil moisture and ocean salinity (SMOS) soil moisture over watershed networks in the U.S[J]. IEEE Transactions on Geoscience and Remote Sensing, 2012, 50(5): 1530-1543.
[9]	SENEVIRATNE S I, CORTI T, DAVIN E L, et al. Investigating soil moisture-climate interactions in a changing climate: a review[J]. Earth-Science Reviews, 2010, 99(3): 125-161.
[10]	雷志斌, 孟庆岩, 田淑芳, 等. 基于GF-3和Landsat8遥感数据的土壤水分反演研究[J]. 地球信息科学学报, 2019, 21(12): 1965-1976.
[11]	DE JEU R A M, WAGNER W, HOLMES T R H, et al. Global soil moisture patterns observed by space borne microwave radiometers and scatterometers[J]. Surveys in Geophysics, 2008, 29(4): 399-420.
[42]	MOHANTY B P, SKAGGS T H. Spatio-temporal evolution and time-stable characteristics of soil moisture within remote sensing footprints with varying soil, slope, and vegetation[J]. Advances in Water Resources, 2001, 24(9): 1051-1067.
[43]	JOSHI C, MOHANTY B P, JACOBS J M, et al. Spatiotemporal analyses of soil moisture from point to footprint scale in two different hydroclimatic regions[J]. Water Resources Research, 2011, 47(1): W01508.
[44]	LIN H, BOUMA J, WILDING L P, et al. Advances in hydropedology[J]. Advances in Agronomy, 2005, 85: 1-89.
[45]	宋沛林. 基于AMSR被动微波辐射计的地表土壤含水量估算方法改进及应用研究[D]. 杭州: 浙江大学, 2019.
[46]	李占杰, 陈基培, 刘艳民, 等. 土壤水分遥感反演研究进展[J]. 北京师范大学学报(自然科学版), 2020, 56(3): 474-481.
[12]	MOHANTY B P, COSH M H, LAKSHMI V, et al. Soil moisture remote sensing: state-of-the-science[J]. Vadose Zone Journal, 2017, 16(1): 1-9.
[13]	陈书林, 刘元波, 温作民. 卫星遥感反演土壤水分研究综述[J]. 地球科学进展, 2012, 27(11): 1192-1203.
[14]	COLLIANDER A, JACKSON T J, BINDLISH R, et al. Validation of SMAP surface soil moisture products with core validation sites[J]. Remote Sensing of Environment, 2017, 191: 215-231.
[15]	张滢, 丁建丽, 周鹏. 干旱区土壤水分微波遥感反演算法综述[J]. 干旱区地理, 2011, 34(4): 671-678.
[16]	FUNG A K, PAN G W. A scattering model for perfectly conducting random surfaces I. Model development[J]. International Journal of Remote Sensing, 1987, 8(11): 1579-1593.
[17]	PAN G W, FUNG A K. A scattering model for perfectly conducting random surfaces II. Range of validity[J]. International Journal of Remote Sensing, 1987, 8(11): 1595-1605.
[18]	CHEN K S, WU T D, TSANG L, et al. Emission of rough surfaces calculated by the integral equation method with comparison to three-dimensional moment method simulations[J]. IEEE Transactions on Geoscience and Remote Sensing, 2003, 41(1): 90-101.
[19]	OH Y, SARABANDI K, ULABY F T. An empirical model and an inversion technique for radar scattering from bare soil surfaces[J]. IEEE Transactions on Geoscience and Remote Sensing, 1992, 30(2): 370-381.
[20]	OH Y, SARABANDI K, ULABY F T. Semi-empirical model of the ensemble-averaged differential Mueller matrix for microwave backscattering from bare soil surfaces[J]. IEEE Transactions on Geoscience and Remote Sensing, 2002, 40(6): 1348-1355.
[21]	DUBOIS P C, VAN ZYL J, ENGMAN T. Measuring soil moisture with imaging radars[J]. IEEE Transactions on Geoscience and Remote Sensing, 1995, 33(4): 915-926.
[22]	SHI J C, WANG J, HSU A Y, et al. Estimation of bare surface soil moisture and surface roughness parameter using L-band SAR image data[J]. IEEE Transactions on Geoscience and Remote Sensing, 1997, 35(5): 1254-1266.
[23]	BAGHDADI N, EL HAJJ M, ZRIBI M. Coupling SAR C-band and optical data for soil moisture and leaf area index retrieval over irrigated grasslands[C]// 2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS). July 10-15, 2016, Beijing. IEEE, 2016: 3551-3554.
[24]	EL HAJJ M, BAGHDADI N, ZRIBI M, et al. Soil moisture retrieval over irrigated grassland using X-band SAR data[J]. Remote Sensing of Environment, 2016, 176: 202-218.
[25]	YAO P P, SHI J C, ZHAO T J, et al. Rebuilding long time series global soil moisture products using the neural network adopting the microwave vegetation index[J]. Remote Sensing, 2017, 9(1): 35.
[26]	LU Z, CHAI L N, LIU S M, et al. Estimating time series soil moisture by applying recurrent nonlinear autoregressive neural networks to passive microwave data over the Heihe River Basin, China[J]. Remote Sensing, 2017, 9(6): 574.
[27]	赵天杰, 张立新, 蒋玲梅, 等. 利用主被动微波数据联合反演土壤水分[J]. 地球科学进展, 2009, 24(7): 769-775.
[28]	RODRÍGUEZ-FERNÁNDEZ N, AIRES F, RICHAUME P, et al. Soil moisture retrieval using neural networks: application to SMOS[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(11): 5991-6007.
[29]	GE L L, HANG R L, LIU Y, et al. Comparing the performance of neural network and deep convolutional neural network in estimating soil moisture from satellite observations[J]. Remote Sensing, 2018, 10(9): 1327.
[30]	SANTI E, PALOSCIA S, PETTINATO S, et al. On the synergy of SMAP, AMSR2 and SENTINEL-1 for retrieving soil moisture[J]. International Journal of Applied Earth Observation and Geoinformation, 2018, 65: 114-123.
[31]	BAI X J, ZENG J Y, CHEN K S, et al. Parameter optimization of a discrete scattering model by integration of global sensitivity analysis using SMAP active and passive observations[J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(2): 1084-1099.
[32]	ZENG J Y, SHI P F, CHEN K S, et al. On the relationship between radar backscatter and radiometer brightness temperature from SMAP[J]. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 3115140.
[33]	ZHANG L, ZHANG Z X, XUE Z H, et al. Sensitive feature evaluation for soil moisture retrieval based on multi-source remote sensing data with few in-situ measurements: a case study of the continental U.S[J]. Water, 2021, 13(15): 2003.
[34]	KOLASSA J, GENTINE P, PRIGENT C, et al. Soil moisture retrieval from AMSR-E and ASCAT microwave observation synergy. Part 2: product evaluation[J]. Remote Sensing of Environment, 2017, 195: 202-217.
[35]	唐豪, 王晓云, 陈伏龙, 等. 基于ERA5-Land数据集的玛纳斯河径流模拟研究[J]. 地学前缘, 2022, 29(3): 271-283.
[36]	BREIMAN L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32.
[37]	MIN M, LI J, WANG F, et al. Retrieval of cloud top properties from advanced geostationary satellite imager measurements based on machine learning algorithms[J]. Remote Sensing of Environment, 2020, 239: 111616.
[38]	王思楠. 融合多源遥感和机器学习算法的土壤水分反演和干旱监测及植被响应研究[D]. 呼和浩特: 内蒙古农业大学, 2022.
[39]	LIU Q X, GU X F, CHEN X R, et al. Soil moisture content retrieval from remote sensing data by artificial neural network based on sample optimization[J]. Sensors, 2022, 22(4): 1611.
[40]	梁鸿, 魏倩, 孙运雷. 基于改进的RReliefF算法的超参数重要性分析[J]. 计算机与数字工程, 2020, 48(8): 1840-1845, 2045.
[41]	KIRA K, RENDELL L A. The feature selection problem: traditional methods and a new algorithm[C]// Proceedings of the tenth national conference on artificial intelligence. July 12-16, 1992, San Jose, California. ACM, 1992: 129-134.

不同尺度的土壤含水量主被动微波联合反演方法研究

Soil moisture retrieval on both active and passive microwave data scales

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 41

相关文章 3

编辑推荐

Metrics

本文评价

[1]	胡义明, 陈腾, 罗序义, 唐超, 梁忠民. 基于机器学习模型的淮河流域中长期径流预报研究[J]. 地学前缘, 2022, 29(3): 284-291.
[2]	洪双, 左仁广, 胡浩, 熊义辉, 王子烨. 磁铁矿元素地球化学大数据构建及其在矿床成因分类中的应用[J]. 地学前缘, 2021, 28(3): 87-96.
[3]	洪瑾，甘成势，刘洁. 基于机器学习的洋岛玄武岩主量元素预测稀土元素[J]. 地学前缘, 2019, 26(4): 45-54.