| 878 | 11 | 346 |
| 下载次数 | 被引频次 | 阅读次数 |
土壤水分是陆地大气间水和能量交换的重要变量，特别是半干旱区土壤水分在植被恢复过程中扮演着关键作用。针对黄河流域土壤水分站网稀少、获取高质量时空连续的土壤水分信息较为困难的问题，为了在这种缺资料地区获取高精度的土壤水分信息，从而进一步开展黄河流域生态水文退化恢复机制解析，选取该区域内浑河流域作为试验区，同为我国北方半干旱地区呼伦贝尔草原的土壤水分实测数据集被用来驱动时空融合的卷积-循环神经网络(CNN-RNN)深度学习方法，集成多源遥感数据和表层土壤水分数据预测缺资料区域浑河流域时空连续的深层土壤水分信息，并采用多指标评估该方法的可行性。结果表明：基于深度学习方法预测的浑河流域5～80 cm土壤水分与实测数据相比存在低估现象，总体的相关系数R、均方根误差RMSE和平均绝对误差MAE分别可以达到0.67、0.029 cm3·cm-3和0.025 cm3·cm-3,其中5～10 cm土壤水分预测效果表现最好，即使对于难以预测的深层土壤水分，最低的R、RMSE和MAE也可以达到0.58、0.031 cm3·cm-3和0.027 cm3·cm-3。研究成果表明基于深度学习预测缺资料区域深层土壤水分的方法具有一定的可行性，为缺资料地区获取高精度的土壤水分提供了另一种思路。
Abstract：Soil moisture is an important variable for water and energy exchange between the terrestrial atmosphere, especially in semi-arid areas where soil moisture plays a key role in vegetation recovery. Aiming at the problem that the networks of soil moisture monitoring station within Yellow River Basin are sparse and then more difficult to obtain the relevant high quality spatio-temporal continuous soil moisture information, Hunhe River Watershed within the region is taken as the study area, while the measured soil moisture data set of Hulunbuir Grassland-a similar semi-arid area in northern China is use to drive the spatio-temporally fused convolution cyclic neural network(CNN-RNN)deep learning method for integrating multi-source remote sensing data and surface soil moisture data to predict the spatio-temporal continuous deep soil moisture information within the data-deficient region—Hunhe River Watershed, so as to get highly accurate soil moisture in this kind of data-deficient region and further analyze the restoration mechanism of the eco-hydrological degradation in the Yellow River Basin. Moreover, the feasibility of this method is evaluated by multiple indexes. The result shows that the soil moisture of 5～80 cm in Hunhe River Watershed predicted based on the deep learning method is underestimated if compared with the measured data, while the overall correlation coefficient(R), root mean square error(RMSE) and mean absolute error(MAE) can reach 0.67, 0.029 cm3·cm-3 and 0.025 cm3·cm-3 respectively, in which the predicting effect is the best for the soil moisture of 5～10 cm, while the lowest R, RMSE and MAE can also reach 0.58, 0.031 cm3·cm-3 and 0.027 cm3·cm-3 even for the deep soil moisture that is difficult to be predicted. The study result indicates that the deep learning-based method for predicting the deep soil moisture in the data-deficient region has certain feasibility and provides another idea for obtaining highly accurate soil moisture within data-deficient region.
[1] MCCOLL K A,ALEMOHAMMAD S H,AKBAR R,et al.The global distribution and dynamics of surface soil moisture[J].Nature Geoscience,2017,10(2):100.
[2] 王云强，邵明安，刘志鹏.黄土高原区域尺度土壤水分空间变异性[J].水科学进展，2012,23(3):310-316.WANG Y Q,SHAO M A,LIU Z P.Spatial variability of soil moisture at a regional scale in the Loess Plateau[J].Advances in Water Science,2012,23(3):310-316.
[3] 张翀，雷田旺，宋佃星.黄土高原植被覆盖与土壤湿度的时滞关联及时空特征分析[J].生态学报，2018,38(6):2128-2138.ZHANG C,LEI T W,SONG D X.Analysis of temporal and spatial characteristics of time lag correlation between the vegetation cover and soil moisture in the Loess Plateau[J].Acta Ecologica Sinica,2018,38(6):2128-2138.
[4] GAO H,HRACHOWITZ M,SCHYMANSKI S J,et al.Climate controls how ecosystems size the root zone storage capacity at catchment scale[J].Geophysical Research Letters,2014,41(22):7916-7923.
[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] ZHAO C,JIA X,GONGADZE K,et al.Permanent dry soil layer a critical control on soil desiccation on China′s Loess Plateau[J].Scientific Reports,2019,9(1):3296.
[7] 雷志栋，杨诗秀，谢森传.田间土壤水量平衡与定位通量法的应用[J].水利学报，1988,19(5):1-7.LEI Z D,YANG S X,XIE S C.Fixed plane flux method and its application to soil water balance[J].Journal of Hydraulic Engineering,1988,19(5):1-7.
[8] 潘宁，王帅，刘焱序，等.土壤水分遥感反演研究进展[J].生态学报，2019,39(13):4615-4626.PAN N,WANG S,LIU Y X,et al.Advances in soil moisture retrieval from remote sensing[J].Acta Ecologica Sinica,2019,39(13):4615-4626.
[9] 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.CHEN S L,LIU Y B,WEN Z M.Satellite retrieval of soil moisture:an overview[J].Advance in Earth Sciences,2012,27(11):1192-1203.
[10] SUSHA L S U,SINGH D N,SHOJAEI B M.A critical review of soil moisture measurement[J].Measurement,2014,54:92-105.
[11] CHAN S K,NINDLISH R,et al.Assessment of the SMAP passive soil moisture product[J].IEEE Transactions on Geoscience and Remote Sensing,2016,54(8):4994-5007.
[12] KERR Y H,WALDTEUFEL P,et al.The SMOS soil moisture retrieval algorithm[J].IEEE Transactions on Geoscience and Remote Sensing,2012,50(5):1384-1403.
[13] CAPEHART W J S D,CARLSON T N.Decoupling of surface and near-surface soil water content:a remote sensing perspective[J].Water Resources Research,1997,33(6):1383-1395.
[14] WANG G Q,ZHANG X J,A Y L,et al.A spatio-temporal cross comparison framework for the accuracies of remotely sensed soil moisture products in a climate-sensitive grassland region[J].Journal of Hydrology,2021,597:126089.
[15] ZENG J Y,LI Z,CHEN Q,et al.Evaluation of remotely sensed and reanalysis soil moisture products over the Tibetan Plateau using in-situ observations[J].Remote Sensing of Environment,2015,163:91-110.
[16] MA H L,ZENG J Y,CHEN N C,et al.Satellite surface soil moisture from SMAP,SMOS,AMSR2 and ESA CCI:a comprehensive assessment using global ground-based observations[J].Remote Sensing of Environment,2019,231:111215.
[17] DENG Y,WANG S,BAI X,et al.Comparison of soil moisture products from microwave remote sensing,land model,and reanalysis using global ground observations[J].Hydrological Processes,2020,34(3):836-851.
[18] 王子龙，林百健，姜秋香，等.松嫩平原黑土区冻融期表层土壤含水量对环境因子时空变化的响应分析[J].水利水电技术，2020,51(5):108-117.WANG Zilong,LIN Baijian,JIANG Qiuxiang,et al.Analysis on response from surface soil water content on spatio-temporal variation of environmental factors during freezing-thawing period of black soil region in Songnen Plain[J].Water Resources and Hydropower Engineering,2020,51(5):108-117.
[19] PENG J,LOEW A,MERLIN O,et al.A review of spatial downscaling of satellite remotely sensed soil moisture[J].Reviews of Geophysics,2017,55(2):341-366.
[20] CHEN Y Z,FENG X M,FU B J.An improved global remote-sensing-based surface soil moisture (RSSSM) dataset covering 2003—2018[J].Earth System Science Data,2021,13(1):1-31.
[21] SENANAYAKE I P,YEO I Y,WALKER J P,et al.Estimating catchment scale soil moisture at a high spatial resolution:Integrating remote sensing and machine learning[J].Science of The Total Environment,2021,776.
[22] KOLASSA J,REICHLE R H,LIU Q,et al.Estimating surface soil moisture from SMAP observations using a Neural Network technique[J].Remote Sensing of Environment,2018,204:43-59.
[23] YAO P,LU H,SHI J,et al.A long term global daily soil moisture dataset derived from AMSR-E and AMSR2 (2002—2019)[J].Scientific Data,2021,8(1):143.
[24] HU F,WEI Z,ZHANG W,et al.A spatial downscaling method for SMAP soil moisture through visible and shortwave-infrared remote sensing data[J].Journal of Hydrology,2020,590:125360.
[25] 李元寿，王根绪，丁永建，等.青藏高原高寒草甸区土壤水分的空间异质性[J].水科学进展，2008,19(1):61-67.LI Y S,WANG G X,DING Y J,et al.Spatial heterogeneity of soil moisture in alpine meadow area of the Qinghai-Xizang Plateau[J].Advances in Water Science,2008,19(1):61-67.
[26] 刘川，余晔，解晋，等.多套土壤温湿度资料在青藏高原的适用性[J].高原气象，2015,34(3):653-665.LIU C,YU Y,XIE J,et al.Applicability of soil temperature and moisture in several datasets over Qinghai-Xizang Plateau[J].Plateau Meteorology,2015,34(3):653-665.
[27] 范科科，张强，史培军，等.基于卫星遥感和再分析数据的青藏高原土壤湿度数据评估[J].地理学报，2018,73(9):1778-1791.FAN K K,ZHANG Q,SHI P J,et al.Evaluation of remote sensing and reanalysis soil moisture products on the Tibetan Plateau[J].Acta Geographica Sinica,2018,73(9):1778-1791.
[28] 金凤君，马丽，许堞.黄河流域产业发展对生态环境的胁迫诊断与优化路径识别[J].资源科学，2020,42(1):127-136.JIN F J,MA L,XU D.Environmental stress and optimized path of industrial development in the Yellow River Basin[J].Resources Science,2020,42(1):127-136.
[29] 陈怡平，傅伯杰.黄河流域不同区段生态保护与治理的关键问题[N].中国科学报，2021-03-03(7).CHEN Y P,FU B J.Key Issues in Ecological Protection and Management of Different Sections of the Yellow River Basin[N].China Science News,2021-03-03(7).
[30] SIMUNEK J,HOPMANS J W.Modeling compensated root water and nutrient uptake[J].Ecological Modelling,2009,220(4):505-521.
[31] KOHNE J M,KOHNE S,SIMUNEK J.A review of model applications for structured soils:a) Water flow and tracer transport[J].Journal of Contaminant Hydrology,2009,104(1-4):4-35.
[32] SIMUNEK J,VAN GENUCHTEN M T,SEJNA M.Recent developments and applications of the Hydrus computer software packages[J].Vadose Zone Journal,2016,15(7).
[33] A Y L,WANG G Q,LIU T X,et al.Vertical variations of soil water and its controlling factors based on the structural equation model in a semi-arid grassland[J].Science of The Total Environment,2019,691:1016-1026.
[34] YUAN Q,SHEN H,LI T,et al.Deep learning in environmental remote sensing:a chievements and challenges[J].Remote Sensing of Environment,2020,241.
[35] 周飞燕，金林鹏，董军.卷积神经网络研究综述[J].计算机学报，2017,40(6):1229-1251.ZHOU F Y,JIN L P,DONG J.Review of convolutional neural network[J].Chinese Journal of Computers,2017,40(6):1229-1251.
[36] VAN LINT J W C,HOOGENDOORN S P,VAN ZUYLEN H J.Freeway travel time prediction with state-space neural networks:modeling state-space dynamics with recurrent neural networks[J].Transportation Research Record,2002,1811(1):30-39.
[37] HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
[38] 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.
基本信息:
DOI：10.13928/j.cnki.wrahe.2022.11.003
中图分类号:TP18;S152.7
引用信息:
[1]张自豪,王国强,薛宝林,等.基于深度学习方法预测缺资料区域深层土壤水分[J],2022,53(11):25-36.DOI:10.13928/j.cnki.wrahe.2022.11.003.
基金信息:
内蒙自治区科技重大专项(2020ZD0009);; 国家杰出青年科学基金项目(52125901)