主动微波数据来源于Sentinel-1卫星,为了使SMAP亮度温度数据与Sentinel-1后向散射系数数据在时间与空间上相匹配,需参照上一步获取的SMAP亮度温度数据,选择成像日期与SMAP数据相同,且成像区域与SMAP数据存在重叠的Sentinel-1数据。本研究选用L1级地距影像(Ground Range Detected,GRD),并经过滤波、辐射定标和地理编码等预处理步骤,最终得到后向散射系数数据。
本研究所用的机器学习算法为随机森林算法。随机森林(Random Forest)最早由Breiman于2001年提出[36],是一系列二元分类回归树(Classification And Regression Tree)的有序集成,能够凭借大量分类回归树对结果的正确预测获得更高的预测精度[37]。图2[38]为随机森林算法的结构示意图。随机森林算法的实现主要包括以下步骤:(1)按照设定的比例,从训练数据集中有放回地抽取N次,组成N个用于训练的数据子集;(2)对每个训练数据子集进行训练,形成由N棵决策树构成的随机森林;(3)综合各决策树的决策结果,对于分类问题则取众数类别,对于回归问题则计算其算术平均值,得到最终的预测值。
DAIA G, TRENBERTHK E, QIANT 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]
LOEWA, STACKET, DORIGOW, 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]
PETROPOULOSG P, IRELANDG, SRIVASTAVAP 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]
HIRSCHIM, SENEVIRATNES I, SCH\ARC. Seasonal variations in terrestrial water storage for major midlatitude river basins[J]. Journal of Hydrometeorology, 2006, 7(1): 39.
[5]
ROBINSOND A, CAMPBELLC S, HOPMANSJ W, et al. Soil moisture measurement for ecological and hydrological watershed-scale observatories: a review[J]. Vadose Zone Journal, 2008, 7(1): 358-389.
REYNOLDSS 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]
JACKSONT J, BINDLISHR, COSHM 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]
SENEVIRATNES I, CORTIT, DAVINE L, et al. Investigating soil moisture-climate interactions in a changing climate: a review[J]. Earth-Science Reviews, 2010, 99(3): 125-161.
DEJEU R A M, WAGNERW, HOLMEST 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.
[12]
MOHANTYB P, COSHM H, LAKSHMIV, et al. Soil moisture remote sensing: state-of-the-science[J]. Vadose Zone Journal, 2017, 16(1): 1-9.
FUNGA K, PANG W. A scattering model for perfectly conducting random surfaces I. Model development[J]. International Journal of Remote Sensing, 1987, 8(11): 1579-1593.
[17]
PANG W, FUNGA 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]
CHENK S, WUT D, TSANGL, 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]
OHY, SARABANDIK, ULABYF 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]
OHY, SARABANDIK, ULABYF 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]
DUBOISP C, VANZYL J, ENGMANT. Measuring soil moisture with imaging radars[J]. IEEE Transactions on Geoscience and Remote Sensing, 1995, 33(4): 915-926.
[22]
SHIJ C, WANGJ, HSUA 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]
BAGHDADIN, ELHAJJ M, ZRIBIM. 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]
ELHAJJ M, BAGHDADIN, ZRIBIM, et al. Soil moisture retrieval over irrigated grassland using X-band SAR data[J]. Remote Sensing of Environment, 2016, 176: 202-218.
[25]
YAOP P, SHIJ C, ZHAOT 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]
LUZ, CHAIL N, LIUS 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.
RODRÍGUEZ-FERNÁNDEZN, AIRESF, RICHAUMEP, 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]
GEL L, HANGR L, LIUY, 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]
SANTIE, PALOSCIAS, PETTINATOS, 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]
BAIX J, ZENGJ Y, CHENK 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]
ZENGJ Y, SHIP F, CHENK 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]
ZHANGL, ZHANGZ X, XUEZ 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]
KOLASSAJ, GENTINEP, PRIGENTC, 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.
BREIMANL. Random forests[J]. Machine Learning, 2001, 45(1): 5-32.
[37]
MINM, LIJ, WANGF, 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.
LIUQ X, GUX F, CHENX 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.
KIRAK, RENDELLL 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.
[42]
MOHANTYB P, SKAGGST 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]
JOSHIC, MOHANTYB P, JACOBSJ 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]
LINH, BOUMAJ, WILDINGL P, et al. Advances in hydropedology[J]. Advances in Agronomy, 2005, 85: 1-89.