Please wait a minute...
img

官方微信

遥感技术与应用  2022, Vol. 37 Issue (4): 878-887    DOI: 10.11873/j.issn.1004-0323.2022.4.0878
蒸散发遥感专栏     
基于深度神经网络联合AMSR2和MODIS数据估算全球蒸散发研究
廖廓1(),彭中2,3(),姜亚珍2,3,党皓飞1
1.福建省气象科学研究所,福建 福州 350001
2.中国科学院地理科学与资源研究所 资源与环境信息系统国家重点实验室,北京 100101
3.中国科学院大学,北京 100049
Global Evapotranspiration Estimation Study based on AMSR2 and MODIS Data
Kuo Liao1(),Zhong Peng2,3(),Yazheng Jiang2,3,Haofei Dang1
1.Fujian Institute of Meteorological Science,Fuzhou 350001,China
2.State Key Laboratory of Resources and Environment Information System,Institute of Geographic Sciences and NaturalResources Research,Chinese Academy of Sciences,Beijing 100101,China
3.University of Chinese Academy of Sciences,Beijing 100049,China
 全文: PDF(4343 KB)   HTML
摘要:

地表蒸散发(ET)是水循环和能量循环的关键组成部分,具有极其重要的应用价值。研究旨在发展一种可靠且高效的深度神经网络(DNN)模型,基于MODIS可见光数据、微波AMSR2亮度温度和数字高程DEM,实现全天候全球高分辨率每日ET的估算。利用FLUXNET和AmeriFlux通量网6种代表性土地覆盖类型的148个站点观测数据来训练和验证DNN模型,结果表明:DNN模型可以有效建立卫星数据(MODIS、AMSR2数据)与ET之间的关系;6种地类的ET估算结果验证的平均绝对误差(MAE)为0.16—0.63 mm/d,均方根误差(RMSE)为0.27—0.89 mm/d,除裸地的决定系数(R2)为0.37以外,其他地类的R2均>0.7。通过对比模型估算的ET与MOD16A2和GLEAM的ET产品,结果表明3种产品的ET空间分布特征相似,ET值非常接近,估算得到的全球2020年日均ET为0—4 mm/d。

关键词: 蒸散发MODISAMSR2深度神经网络    
Abstract:

The surface Evapotranspiration (ET) is a key component of the water and energy cycles and has critical value for applications. This study is aimed to develop a reliable and efficient Deep Neural Network (DNN) model for all-weather global daily ET estimation with high spatial resolution, using remote sensing MODIS datasets, microwave AMSR2 brightness temperature products and digital elevation DEM data as input. The study used 148 site observations over six representative land cover types from FLUXNET and AmeriFlux to train and validate DNN models. The results showed that the DNN model can effectively established the relationship between satellite (MODIS, AMSR2) data and ET, and the Mean Absolute Error (MAE) of ET estimation results for the six land cover types ranged from 0.16 to 0.63 mm/day, and the Root Mean Square Error (RMSE) ranged from 0.27 to 0.89 mm/day, and the coefficient of determination (R2) of all land types were > 0.7, except for bare land, where the R2 was 0.37. By comparing the ET estimation in this study with the ET products of MOD16A2 and GLEAM, the results demonstrated that the spatial distribution characteristics of the three ET products were similar and the ET values were very close with global average daily ET of 0~4 mm/day over 2020.

Key words: Evapotranspiration    MODIS    AMSR2    Deep neural network
收稿日期: 2021-10-28 出版日期: 2022-09-28
:  P333  
基金资助: 福建省海峡气象开放室课题“机器算法与遥感融合对大城市PM2.5浓度预测研究”(2020KX03)
通讯作者: 彭中     E-mail: liaokuo78@163.com;pengzhong19@mails.ucas.edu.cn
作者简介: 廖 廓(1978-),男,湖南长沙人,硕士,高级工程师,主要从事生态气象和卫星遥感应用研究。E?mail: liaokuo78@163.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
廖廓
彭中
姜亚珍
党皓飞

引用本文:

廖廓,彭中,姜亚珍,党皓飞. 基于深度神经网络联合AMSR2和MODIS数据估算全球蒸散发研究[J]. 遥感技术与应用, 2022, 37(4): 878-887.

Kuo Liao,Zhong Peng,Yazheng Jiang,Haofei Dang. Global Evapotranspiration Estimation Study based on AMSR2 and MODIS Data. Remote Sensing Technology and Application, 2022, 37(4): 878-887.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2022.4.0878        http://www.rsta.ac.cn/CN/Y2022/V37/I4/878

图1  FLUXNET和AmeriFlux全球通量观测网站点(148个)空间分布图
模型参数
激活函数'relu'
学习率‘adaptive’
隐藏层128-64-32
批大小128
初始学习率0.000 1
最大迭代10 000
表1  深度神经网络算法(DNN)模型的最优参数
图2  6种土地覆被类型的DNN模型估算蒸散发与站点观测的验证散点密度图
图3  DNN模型估算的蒸散发在148个站点的验证MAE
图4  模型估算的2020年日均蒸散发与同期ET产品比较
1 Zhang K, Kimball John S, Running Steven W. A review of remote sensing based actual evapotranspiration estimation[J]. Wiley Interdisciplinary Reviews. Water, 2016, 3(6): 834-853. DOI: 10.1002/wat2.1168 .
doi: 10.1002/wat2.1168
2 Liang Shunlin, Bai Rui, Chen Xiaona, et al. Review of China's land surface quantitative remote sensing development in 2019[J]. National Remote Sensing Bulletin, 2020, 24(6): 618-671.
2 梁顺林, 白瑞, 陈晓娜, 等. 2019年中国陆表定量遥感发展综述[J]. 遥感学报, 2020, 24(6): 618-671.
3 Xiong Yujiu, Feng Fangguan, Fang Yizhou, et al. Critical problems when applying remotely sensed evapotranspiration products[J]. Remote Sensing Technology and Application, 2021, 36(1): 121-131.
3 熊育久, 冯房观, 方奕舟, 等. 蒸散发遥感反演产品应用关键问题浅议[J]. 遥感技术与应用, 2021, 36(1): 121-131.
4 Zhang Qiang, Zhang Zhixian, Wen Xiaomei, et al. Comparisons of observational methods of land surface evapotranspiration and their influence factors[J]. Advances in Earth Science, 2011, 26(5): 538-547.
4 张强, 张之贤, 问晓梅, 等. 陆面蒸散量观测方法比较分析及其影响因素研究[J]. 地球科学进展, 2011, 26(5): 538-547.
5 Zhang Gong, Zheng Ning, Zhang Jingsong, et al. Advances in the study of regional-averaged evapotranspiration using the scintillation method[J]. Acta Ecologica Sinica, 2018, 38(8): 2625-2635.
5 张功, 郑宁, 张劲松, 等. 光闪烁方法测算区域蒸散研究进展[J]. 生态学报, 2018, 38(8): 2625-2635.
6 Wang K C, Dickinson R E. Global atmospheric downward longwave radiation at the surface from ground-based observations, satellite retrievals, and reanalyses[J]. Reviews of Geophysics, 2013,51:150-185.
7 Jackson R D, Reginato R J, Idso S B. Wheat canopy temperature: A practical tool for evaluating water requirements[J]. Water Resources Research, 1977, 13(3): 651-656. DOI: 10.1029/WR013i003p00651 .
doi: 10.1029/WR013i003p00651
8 Wang K C, Wang P C, Li Z Q, et al. A simple method to estimate actual evapotranspiration from a combination of net radiation, vegetation index, and temperature[J]. Journal of Geophysical Research,2007,112(D15). DOI:10.1029/2006JD 008351 .
doi: 10.1029/2006JD 008351
9 Bastiaanssen Wim G M, Menenti M, Feddes R A, et al. A remote sensing surface energy balance algorithm for land (SEBAL)-1. Formulation[J]. Journal of Hydrology, 1998, 212(1): 198-212.
10 Tang Ronglin, Wang Shengli, Jiang Yazhen, et al. A review of retrieval of land surface evapotranspiration based on remotely sensed surface temperature versus vegetation index triangular/trapezoidal characteristic space[J]. National Remote Sensing Bulletin,2021,25(1): 65-82.
10 唐荣林, 王晟力, 姜亚珍, 等. 基于地表温度—植被指数三角/梯形特征空间的地表蒸散发遥感反演综述[J]. 遥感学报, 2021, 25(1): 65-82.
11 Peters-Lidard Christa D, Kumar Sujay V, Mocko David M, et al. Estimating evapotranspiration with land data assimilation systems[J].Hydrological Processes,2011,25(26):3979-3992. DOI: 10.1002/hyp.8387 .
doi: 10.1002/hyp.8387
12 Li Jia, Xin Xiaozhou, Peng Zhiqing, et al. Remote sensing products of terrestrial evapotranspiration: Comparison and outlook[J].Remote Sensing Technology and Application,2021,36(1):103-120.
12 李佳,辛晓洲,彭志晴,等.地表蒸散发遥感产品比较与分析[J].遥感技术与应用,2021,36(1):103-120.
13 Liu Meng, Tang Ronglin, Li Zhaoliang, et al. Progress of data-driven remotely sensed retrieval methods and products on land surface evapotranspiration[J]. National Remote Sensing Bulletin, 2021, 25(8): 1517-1537.
13 刘萌, 唐荣林, 李召良, 等. 数据驱动的蒸散发遥感反演方法及产品研究进展[J]. 遥感学报, 2021, 25(8): 1517-1537.
14 Zhou Yi, Qin Zhihao, Bao Gang. Progress in retrieving Land Surface Temperature for the cloud-covered pixels from thermal infrared remote sensing data[J]. Spectroscopy and Spectral Analysis, 2014(2): 364-369.
14 周义, 覃志豪, 包刚. 热红外遥感图像中云覆盖像元地表温度估算研究进展[J]. 光谱学与光谱分析, 2014(2): 364-369.
15 Liu Rong, Wen Jun, Wang Xin, et al. Hourly variation of evapotranspiration estimated by visible infrared and microwave data over the Northern Tibetan Plateau[J]. Journal of Infrared and Millimeter Waves, 2015, 34(2): 211-217.
15 刘蓉, 文军, 王欣, 等. 结合可见光近红外和微波遥感估算藏北高原日蒸散发量[J]. 红外与毫米波学报, 2015, 34(2): 211-217.
16 Yi Z Y, Zhao H L, Jiang Y Z, et al. Daily evapotranspiration estimation at the field scale: Using the modified SEBS model and HJ-1 data in a desert-oasis area, Northwestern China[J]. Water (Basel),2018,10(5):640. DOI:10.3390/w10050640 .
doi: 10.3390/w10050640
17 Wang Y P, Li R, Min Q L, et al. A three-source satellite algorithm for retrieving all-sky evapotranspiration rate using combined optical and microwave vegetation index at twenty AsiaFlux sites[J]. Remote Sensing of Environment, 2019, 235: 111463. DOI: 10.1016/j.rse.2019.111463 .
doi: 10.1016/j.rse.2019.111463
18 Man Haoran, Zang Shuying, Li Miao, et al. Surface temperature inversion in Northeast China based on AMSR-E passive microwave data[J]. Science of Surveying and Mapping, 2021, 46(3): 124-132.
18 满浩然, 臧淑英, 李苗, 等. 应用微波遥感数据的东北地区地表温度反演[J]. 测绘科学, 2021, 46(3): 124-132.
19 Cui Lu, Zhang Peng, Che Jin. Overview of deep neural network based classification algorithms for remote sensing images[J]. Computer Science, 2018, 45(z1): 50-53.
19 崔璐, 张鹏, 车进. 基于深度神经网络的遥感图像分类算法综述[J]. 计算机科学, 2018, 45(z1): 50-53.
20 Martin J, Sujan K, Ulrich W, et al. The FLUXCOM ensemble of global land-atmosphere energy fluxes[J]. Scientific Data, 2019, 6(1). DOI: 10.1038/s41597-019-0076-8 .
doi: 10.1038/s41597-019-0076-8
21 Peter B, Dueben Peter D, Torsten H, et al. The digital revolution of Earth-system science[J]. Nature Computational Science,2021,1(2):104-113.DOI:10.1038/s43588-021-00023-0 .
doi: 10.1038/s43588-021-00023-0
22 Fan J L, Zheng J, Wu L F, et al. Estimation of daily maize transpiration using support vector machines, extreme gradient boosting, artificial and deep neural networks models[J]. Agricultural Water Management,2021,245:106547. DOI:10.1016/ j.agwat.2020.106547 .
doi: 10.1016/ j.agwat.2020.106547
23 Zhao Wenli, Qiu Guoyu, Xiong Yujiu, et al. Simulation of sub-daily transpiration characteristics of typical arbor trees in cities based on Deep Neural Network[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2021, 57(2): 322-332.
23 赵文利, 邱国玉, 熊育久, 等. 基于深度神经网络的城市典型乔木日内蒸腾特征模拟研究[J]. 北京大学学报(自然科学版), 2021, 57(2): 322-332.
24 Feng Lili, Zhang Kun, Han Tuo,et al.Scale expansion of eva-potranspiration in different vegetation types based on the artificial neural network[J]. Journal of Lanzhou University(Natural Sciences Edition),2017,53(2):186-193.
24 冯丽丽,张琨,韩拓,等.基于人工神经网络的不同植被类型蒸散量时空尺度扩展[J].兰州大学学报(自然科学版),2017,53(2):186-193.
25 Liu Jing, Ma Hongzhang, Yang Le, et al. A Survey of surface temperature retrieval by passive microwave remote sensing[J]. Remote Sensing Technology and Application, 2012, 27(6): 812-821.
25 刘晶, 马红章, 杨乐, 等. 基于被动微波的地表温度反演研究综述[J]. 遥感技术与应用, 2012, 27(6): 812-821.
26 Jin C, Per Jönsson, Masayuki T, et al. A simple method for reconstructing a high-quality NDVI time-series data set based on the Savitzky–Golay filter[J]. Remote Sensing of Environment,2004,91(3-4):332-344. DOI:10.1016/j.rse.2004. 03.014 .
doi: 10.1016/j.rse.2004. 03.014
27 Qu Di, Fan Wenyi, Yang Jinming, et al. Quantitative estimation of evapotranspiration from Tahe forest ecosystem, Northeast China[J]. Chinese Journal of Applied Ecology,2014,25(6): 1652-1660.
27 曲迪,范文义,杨金明,等.塔河森林生态系统蒸散发的定量估算[J]. 应用生态学报,2014,25(6):1652-1660.
28 Wang Yining, Zhang Xiaomeng, Lu Lu, et al. Estimation of crop coefficient and evapotranspiration of summer maize by path analysis combined with BP neural network[J]. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36(7): 109-116.
28 王怡宁, 张晓萌, 路璐, 等. 通径分析结合BP神经网络方法估算夏玉米作物系数及蒸散量[J]. 农业工程学报, 2020, 36(7): 109-116.
29 Jin Rui, Li Xin, Ma Mingguo, et al. Key methods and experiment verification for the validation of quantitative remote sensing products[J]. Advances in Earth Science, 2017, 32(6): 630-642.
29 晋锐, 李新, 马明国, 等. 陆地定量遥感产品的真实性检验关键技术与试验验证[J]. 地球科学进展, 2017, 32(6): 630-642.
[1] 孟莹,姜鹏,董巍. 基于遥感的地表蒸散发研究进展[J]. 遥感技术与应用, 2022, 37(4): 839-853.
[2] 马启民,龙银平,汤世宇,贾晓鹏. 库布齐沙漠典型沙地人工林蒸散对比分析[J]. 遥感技术与应用, 2022, 37(4): 854-864.
[3] 崔泽鹏,王志慧,肖培青,申震洲,常晓格,石永磊,马力. 2000—2018年黄河上中游地区蒸散发年际时空变化及其影响因素分析[J]. 遥感技术与应用, 2022, 37(4): 865-877.
[4] 朱爽,张锦水. 中低分冬小麦分布提取模型效率的样本特征分析[J]. 遥感技术与应用, 2022, 37(3): 608-619.
[5] 康新礼,张文豪,刘原萍,顾行发,余涛,张丽丽,徐桦昆. 基于随机森林的京津冀地区PM2.5遥感反演及变化分析[J]. 遥感技术与应用, 2022, 37(2): 424-435.
[6] 赵天玮,朱文彬,裴亮,宝康妮. 三江源蒸散发遥感估算及其时空分布特征研究[J]. 遥感技术与应用, 2022, 37(1): 137-147.
[7] 张晨,袁金国. 基于MODIS数据的河北省草地和林地的物候期及其与NPP相关分析[J]. 遥感技术与应用, 2022, 37(1): 205-217.
[8] 徐艳豪,丁忠昊,宋立生. TSEB模型在复杂下垫面下模拟结果比较研究[J]. 遥感技术与应用, 2022, 37(1): 85-93.
[9] 葛强,沈文举,李冉,李莘莘,蔡坤,左宪禹,乔保军,张云舟. 2001~2018年我国热异常点时空分布特征研究[J]. 遥感技术与应用, 2022, 37(1): 73-84.
[10] 胡晓静,郝晓华,王建,戴礼云,赵宏宇,李弘毅. 基于AMSR2和MODIS数据融合的雪深降尺度算法研究—以北疆地区为例[J]. 遥感技术与应用, 2021, 36(6): 1236-1246.
[11] 敏玉芳,张耀南,康建芳,冯克庭. 基于MODIS影像的中巴经济走廊荒漠化程度时空动态监测研究[J]. 遥感技术与应用, 2021, 36(4): 827-837.
[12] 李佳,辛晓洲,彭志晴,李小军. 地表蒸散发遥感产品比较与分析[J]. 遥感技术与应用, 2021, 36(1): 103-120.
[13] 熊育久,冯房观,方奕舟,邱国玉,赵少华,姚云军. 蒸散发遥感反演产品应用关键问题浅议[J]. 遥感技术与应用, 2021, 36(1): 121-131.
[14] 王浩天,汪源,袁强强. 2008~2016年MODIS多角度大气校正气溶胶产品在中国的时空分布及变化趋势[J]. 遥感技术与应用, 2021, 36(1): 217-228.
[15] 雷华锦,李弘毅,王建,郝晓华,赵宏宇,张娟. 基于环境信息和回归模型的青藏高原MODIS积雪面积比例产品制备[J]. 遥感技术与应用, 2020, 35(6): 1303-1311.