Please wait a minute...
img

官方微信

遥感技术与应用  2021, Vol. 36 Issue (4): 898-907    DOI: 10.11873/j.issn.1004-0323.2021.4.0898
遥感应用     
基于Landsat 8与实测数据的水质参数反演研究
吴欢欢(),国巧真(),臧金龙,乔悦,朱丽,何云海
天津城建大学地质与测绘学院,天津 300384
Study on Water Quality Parameter Inversion based on Landsat 8 and Measured Data
Huanhuan Wu(),Qiaozhen Guo(),Jinlong Zang,Yue Qiao,Li Zhu,Yunhai He
School of Geology and Geomatics,Tianjin Chengjian University,Tianjin 300384,China
 全文: PDF(5098 KB)   HTML
摘要:

目前遥感技术已成为监测水质参数的重要手段,精度更高的水质参数反演模型是当前水质监测的重点。但由于水环境的复杂性、遥感数据的局限性等多重原因,水质参数遥感反演精度有限,且多集中于水色水质参数反演。为了得到精度更高的水质参数反演模型,以天津市海河下游段为研究区,对Landsat 8 OLI遥感影像进行大气校正、辐射定标等预处理,通过实验室理化分析测定水体的总磷、氮氨、总氮浓度及电导率,建立实测水质参数与Landsat 8 OLI遥感影像数据的统计回归模型及神经网络模型,采用决定系数(R2)、平均绝对误差(MAE)、均方根误差(RMSE)进行精度检验,神经网络模型反演结果R2均大于0.85,MAE分别为0.019、0.09、0.242、0.411,RMSE分别为0.024、0.118、0.286、0.562,反演精度较好。结果表明:基于神经网络建立的水质参数反演模型精度较高。

关键词: Landsat 8 OLI遥感影像神经网络模型水质参数反演    
Abstract:

At present, remote sensing technology has become an important method for monitoring water quality parameters, and a more accurate water quality parameter inversion model is the focus of current water quality monitoring. However, due to multiple reasons such as the complexity of the water environment and the limitations of remote sensing data, the accuracy of water quality parameter remote sensing inversion is limited, and most of them focus on the inversion of water color water quality parameters. In order to obtain a better accurate water quality parameter inversion model, taking the lower reaches of the Haihe River in Tianjin as the research area, Landsat 8 OLI remote sensing images were subjected to atmospheric correction, radiometric calibration and other pretreatments, and the total phosphorus, ammonia nitrogen, total nitrogen concentration and conductivity of the water body were determined by laboratory physical and chemical analysis. The statistical regression model and neural network model of measured water quality parameters and Landsat 8 OLI remote sensing image data are established. Coefficient of determination (R2), Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) are used to test the accuracy, and the neural network model inversion results R2 is greater than 0.85, MAE is 0.019, 0.09, 0.242, 0.411, RMSE is 0.024, 0.118, 0.286, 0.562, and the inversion accuracy is better. The results show that the water quality parameter inversion model based on neural network has high accuracy.

Key words: Landsat 8 OLI remote sensing images    Neural network model    Water quality parameters    Inversion
收稿日期: 2020-04-25 出版日期: 2021-09-26
ZTFLH:  TP79  
基金资助: 天津市自然科学基金项目“天津滨海新区地表水环境遥感监测与生态风险评价”(18JCYBJC90900);天津市教委科研计划项目“遥感技术视角下的天津市地表温度研究”(2018KJ164)
通讯作者: 国巧真     E-mail: 18822085914@163.com;gqiaozhen@tcu.edu.cn
作者简介: 吴欢欢(1994-),女,山西临汾人,硕士研究生,主要从事遥感反演方面的研究。E?mail:18822085914@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  
吴欢欢
国巧真
臧金龙
乔悦
朱丽
何云海

引用本文:

吴欢欢,国巧真,臧金龙,乔悦,朱丽,何云海. 基于Landsat 8与实测数据的水质参数反演研究[J]. 遥感技术与应用, 2021, 36(4): 898-907.

Huanhuan Wu,Qiaozhen Guo,Jinlong Zang,Yue Qiao,Li Zhu,Yunhai He. Study on Water Quality Parameter Inversion based on Landsat 8 and Measured Data. Remote Sensing Technology and Application, 2021, 36(4): 898-907.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2021.4.0898        http://www.rsta.ac.cn/CN/Y2021/V36/I4/898

图1  采样点地理位置
水体指数公式水体指数公式
归一化 水体指数NDWI=B3-B5B3+B5增强型 水体指数EWI=B3-B5-B6B3+B5+B6
改进归一化 水体指数MNDWI=B3-B6B3+B6新型水 体指数NWI=B1-(B5+B6+B7)B1+B5+B6+B7
表1  水体指数公式[25]
因子相关性因子相关性
B1/B70.672B3/B5-0.581
B2/B70.649B3/B40.645
B3/B70.705NDWI0.590
B3/B60.691EWI0.663
表2  总磷与波段组合间的相关性指数
因子相关性因子相关性
B60.773B6+B70.781
B70.789B5+B70.779
B5+B60.774
表3  氨氮与波段组合间的相关性指数
因子相关性因子相关性
B1-B50.654B4/(B5+B60.672
B4-B50.661EWI0.621
B3-B50.705
表4  总氮与波段组合间的相关性指数
因子相关性因子相关性
NDWI0.684NWI-0.634
EWI0.751B5/B3-0.677
表5  电导率与波段组合间的相关性指数
因子模型R2函数关系式MAD
EWI三次0.595

Y=0.232+0.604X

-0.141X2-6.431X3

0.047
B1/B7三次0.588

Y=0.695-0.660X

+0.254X2-0.028X3

0.049
B3/B6三次0.703

Y=0.958-1.508X

+0.885X2-0.154X3

0.066
B3/B7三次0.681

Y=1.084-1.571X

+0.838X2-0.134X3

0.167
表6  总磷浓度的拟合模型
因子模型R2函数关系式MAD
B6线性0.598Y=0.289+0.002X0.339
对数0.588Y=-5.349+1.079ln(X)0.360
B7线性0.622Y=0.139+0.003X0.246
对数0.603Y=-5.992+1.2ln(X)0.336
B5+B6线性0.600Y=0.251+0.001X0.410
对数0.597Y=-6.237+1.116ln(X)0.343
B5+B7线性0.606Y=0.180+0.001X0.491
对数0.597Y=-6.587+1.176ln(X)0.330
B6+B7线性0.611Y=0.219+0.001X0.404
对数0.597Y=-6.450+1.137ln(X)0.345
表7  氨氮浓度的拟合模型
因子模型R2函数关系式MAD
EWI三次0.519Y=3.425+5.440X-11.517X2-66.953X30.409
B3-B5三次0.542Y=4.254-0.018X+5.425E-5X2-4.228E-8X30.331
表8  总氮浓度的拟合模型
因子模型R2函数关系式MAD
EWI三次0.592

Y=11.753+14.843X

-10.176X2-95.526X3

1.077
NWI线性0.540Y=12.016+10.157X1.156
三次0.572

Y=12.182+12.532X

-13.980X2-60.690X3

0.986
表9  电导率的拟合模型
图2  各水质参数网络回归分析示意图
图3  水体总磷、氨氮、总氮浓度和电导率统计回归模型与神经网络反演值与实测值对比图
水质参数统计回归模型BP神经网络模型
MADRMSEMADRMSE
总磷0.0660.1010.0190.024
氨氮0.2460.3210.090.118
总氮0.4090.5170.2420.286
电导率0.9861.5830.4110.562
表10  统计回归模型与BP神经网络模型反演结果对比
图4  总磷、氨氮、总氮浓度及电导率神经网络反演结果
1 Huang Litong. Retrieval and application of remote sensing water quality of complicated water body in coastal zone: a case study of the Hangzhou Bay area[D]. Hangzhou: Zhejiang University, 2019.
1 黄李童. 海岸带复杂水体水质遥感反演与应用:以杭州湾区域为例[D]. 杭州:浙江大学, 2019.
2 Pu Lingwei. Remote sensing inversion of Chlorophyll-a concentration using spectral smoothing method to improve the validation precision of the Chlorophyll-a estimation models in turbidity waters[D]. Suzhou: Suzhou University of Science and Technology, 2019.
2 浦玲伟. 基于高分影像的太湖东岸带叶绿素a浓度遥感反演研究[D]. 苏州: 苏州科技大学, 2019.
3 Zhao Song. Inversion of water quality parameters of Fuyang River in Handan City based on multi-source remote sensing data[D]. Handan: Hebei University of Engineering, 2019.
3 赵松. 基于多源遥感数据的邯郸市滏阳河水质参数反演[D]. 邯郸: 河北工程大学, 2021.
4 Song K S, Li L, Tedesco Lenore P , et al. Remote quantification of phycocyanin in potable water sources through an adaptive model[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2014,95:68-80. DOI:10.1016/j.isprsjprs. 2014.06.008.
doi: 10.1016/j.isprsjprs. 2014.06.008
5 Yin Ziyao, Li Junsheng, Fan Haisheng, et al. Preliminary study on water quality parameter inversion for the Yuqiao Reservoir based on Zhuhai-1 hyperspectral satellite data[J]. Spectroscopy and Spectral Analysis, 2021, 41(2): 494-498.
5 殷子瑶, 李俊生, 范海生,等. 珠海一号高光谱卫星的于桥水库水质参数反演初步研究[J]. 光谱学与光谱分析, 2021, 41(2): 494-498.
6 Sun Siyang. Inversion of water quality parameters of Miyun Reservoir based on multi-source remote sensing and machine learning[D]. Beijing: Beijing Forestry University, 2019.
6 孙驷阳. 基于多源遥感与机器学习的密云水库水质参数反演研究[D]. 北京:北京林业大学, 2019.
7 Wang Shan. Monitoring and analyzing of Panjiakou-Daheiting Reservoir water quality change based on multi-source remote sensing images[D]. Lanzhou: Lanzhou Jiaotong University, 2019.
7 王珊. 基于多源遥感影像的潘家口—大黑汀水库水质变化监测与分析[D]. 兰州: 兰州交通大学, 2019.
8 Ma Fangkai, Gao Zhaobo, Ye Bangling. Remote sensing inversion of Tangxun Lake water quality based on high score remote sensing satellite image[J]. Water resources development and management, 2021(5): 69-75.
8 马方凯, 高兆波, 叶帮玲. 基于高分遥感卫星影像的汤逊湖水质遥感反演[J]. 水资源开发与管理, 2021(5): 69-75.
9 Li Yijing, Sun Xiaomin, Guo Yuyin, et al. Remote sensing retrieval of water quality parameters in Poyang Lake based on the gradient boosting decision tree algorithm[J]. Space Return and Remote Sensing, 2020, 41(6): 90-102.
9 李怡静, 孙晓敏, 郭玉银, 等. 基于梯度提升决策树算法的鄱阳湖水环境参数遥感反演[J]. 航天返回与遥感, 2020, 41(6): 90-102.
10 Park Y, Cho K H, Park J, et al. Development of early-warning protocol for predicting Chlorophyll-a concentration using machine learning models in freshwater and estuarine reservoirs, Korea[J]. Science of The Total Environment, 2015, 502: 31-41. DOI: 10.1016/j.scitotenv.2014.09.005.
doi: 10.1016/j.scitotenv.2014.09.005
11 Cheng Chunmei, Wei Yuchun, Wang Guoxiang, et al. Using spectral smoothing method to improve the validation precision of the Chlorophyll-a estimation models in turbidity waters[J]. Remote Sensing Technology and Application, 2013, 28 (6): 941-948.
11 程春梅, 韦玉春, 王国祥, 等.使用光谱平滑提高浑浊水体叶绿素a浓度估算模型的应用精度[J]. 遥感技术与应用, 2013, 28(6): 941-948.
12 Huang Changchun, Li Yunmei, Xu Liangjiang, et al. Study on influencing factors and universality of Chlorophyll-a retrieval model in inland water body[J]. Environmental Science, 2013, 34 (2): 525-531.
12 黄昌春, 李云梅, 徐良将, 等.内陆水体叶绿素反演模型普适性及其影响因素研究[J]. 环境科学, 2013, 34(2): 525-531.
13 Cui, T W, Zhang J, Wang K, et al. Remote sensing of chlorophyll a concentration in turbid coastal waters based on a global optical water classification system[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 163: 187-201. DOI: 10.1016/j.isprsjprs.2020.02.017.
doi: 10.1016/j.isprsjprs.2020.02.017
14 Tavora J E, Boss S, Doxaran D, et al. An algorithm to estimate suspended particulate matter concentrations and associated uncertainties from remote sensing reflectance in coastal environments[J]. Remote Sensing, 2020, 12(13): 2172. DOI: 10.3390/rs12132172.
doi: 10.3390/rs12132172
15 Chen Jun, Wen Zhenhe, Fu Jun. The application of the subsection mapping retrieval model to water qualities quantitative analysis[J]. Spectroscopy and Spectral Analysis, 2010, 30 (10): 2784-2788.
15 陈军, 温珍河, 付军. 基于分段映射模型的水质参数遥感反演研究[J]. 光谱学与光谱分析, 2010, 30(10): 2784-2788.
16 Xue Pengsong, Feng Minquan, Xing Xiaopeng. Water quality prediction model based on Markov chain improved grey neural network[J]. Engineering Journal of Wuhan University, 2012, 45 (3): 19-24.
16 薛鹏松, 冯民权, 邢肖鹏. 基于马尔科夫链改进灰色神经网络的水质预测模型[J]. 武汉大学学报工学版, 2012, 45(3): 19-24.
17 Liu S Y, Tai H J, Ding Q S, et al. A hybrid approach of support vector regression with genetica gorithm optimization for aquaculture water quality prediction[J]. Mathematical and Computer Modelling,2013,58(1):458-465. DOI: 10.1016/j.mcm.2011.11.021.
doi: 10.1016/j.mcm.2011.11.021
18 Yang Liu, Han Yu, Wang Zumao, et al. Study on retrieval model of water quality parameter in Wenyu River based on BP neural network[J]. Journal of Water Resources and Water Engineering, 2013, 24 (6): 25-28.
18 杨柳, 韩瑜, 汪祖茂, 等. 基于 BP 神经网络的温榆河水质参数反演模型研究[J]. 水资源与水工程学报, 2013, 24(6): 25-28.
19 Zhang Sisi, Gao Xuguang, Hua Wenqiang. Remote sensing image information extraction method based on clustering and artificial neural network[J]. Electronic Design Engineering, 2020, 28(15): 106-109.
19 张思思, 高旭光, 滑文强. 基于聚类与人工神经网络的遥感图像信息提取方法[J].电子设计工程,2020,28(15): 106-109.
20 Yang Guofan, Yan Mengdong, Yin Fei. Inversion modeling of Chlorophyll-a in Qinghe Reservoir[J]. Remote Sensing Information, 2016, 31 (5): 74-78.
20 杨国范, 阎孟冬, 殷飞. 清河水库叶绿素a浓度反演模型研究[J]. 遥感信息, 2016, 31 (5): 74-78.
21 Pan Jun, Liang Haitao, Yue Dandan, et al. Application of BP artificial neural network model in the evaluation of groundwater quality[J]. Water Supply Technology, 2015, 9 (6): 6-11.
21 潘俊, 梁海涛, 岳丹丹, 等. BP人工神经网络模型在地下水水质评价中的应用[J]. 供水技术, 2015, 9 (6): 6-11.
22 Guo Qingchun, He Zhenfang, Li Li. Water quality assessment model of Yangtze River based on artificial neural networks[J]. Information Technology, 2013 (8): 54-56,60.
22 郭庆春, 何振芳, 李力. 基于人工神经网络的长江水质评价模型[J]. 信息技术, 2013(8): 54-56,60.
23 Li M S,Wu W,Chen B S,et al.Water quality evaluation using back propagation artificial neural network based on self-adaptive particle swarm optimization algorithm and chaos theory[J]. Computational Water Energy & Environmental Engineering,2017,6(3):229-242. DOI:10.4236/cweee.2017.63016.
doi: 10.4236/cweee.2017.63016
24 Zheng Yanhui, Zhang Yuanbo, He Yanhu. Study on the multi-parameter inversion of reservoir water quality based on GF-1 WFV images and neural network model[J]. People's Pearl River, 2020, 41(7): 57-62,84.
24 郑炎辉, 张园波, 何艳虎. 基于GF-1 WFV影像与神经网络模型的水库水质多参数反演研究[J].人民珠江, 2020, 41(7): 57-62,84.
25 Xiao Qian, Yang Kun, Hong Liang. Comparative study of extraction methods of lake water in Yunnan-Guizhou plateau based on multi-source images[J]. Journal of Yunnan Normal University (Natural Science Edition), 2018, 38(2): 73-78.
25 肖茜, 杨昆, 洪亮. 基于多源数据的云贵高原湖泊水体提取方法对比研究[J]. 云南师范大学学报(自然科学版), 2018, 38(2): 73-78.
[1] 孙景霞,张冬有,侯宇初. 基于多源遥感数据协同反演森林地表土壤水分研究[J]. 遥感技术与应用, 2021, 36(3): 564-570.
[2] 孙奇,关琳琳,焦全军,刘新杰,戴华阳. 基于植被指数融合的冬小麦生物量反演研究[J]. 遥感技术与应用, 2021, 36(2): 391-399.
[3] 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.
[4] 肖尧,马明国,闻建光,于文凭. 复杂地表地表温度反演研究进展[J]. 遥感技术与应用, 2021, 36(1): 33-43.
[5] 郭利彪,刘桂香,运向军,张勇,孙世贤. 基于数据机理的植被叶面积指数遥感反演研究[J]. 遥感技术与应用, 2020, 35(5): 1047-1056.
[6] 薛华柱,王昶景,周红敏,王锦地,万华伟. 基于模拟退火算法的BP神经网络模型估算高分辨率叶面积指数[J]. 遥感技术与应用, 2020, 35(5): 1057-1069.
[7] 程娟,肖青,闻建光,唐勇,游冬琴,卞尊健,吴胜标,郝大磊,钟守熠. 地物波谱数据库应用方法及遥感应用现状[J]. 遥感技术与应用, 2020, 35(2): 267-286.
[8] 王艺晴,韩震,周玮辰,吴义生. SMAP卫星海表面亮温仿真及海表面盐度遥感反演[J]. 遥感技术与应用, 2020, 35(2): 365-371.
[9] 王作亮,文军,刘蓉,李振朝,郑东海,王欣. 基于地基微波辐射计数据评估不同土壤介电模型反演土壤湿度的适用性[J]. 遥感技术与应用, 2020, 35(1): 97-110.
[10] 郑贵洲,熊良超,廖艳雯,王红平. 利用MODIS数据反演南海南部海表温度及时空变化分析[J]. 遥感技术与应用, 2020, 35(1): 132-140.
[11] 刘俊,孟庆岩,葛小三,刘顺喜,陈旭,孙云晓. 基于BP神经网络的夏玉米多生育期叶面积指数反演研究[J]. 遥感技术与应用, 2020, 35(1): 174-184.
[12] 边玲玲,王卷乐,郭兵,程凯,魏海硕. 基于特征空间的黄河三角洲垦利县土壤盐分遥感提取[J]. 遥感技术与应用, 2020, 35(1): 211-218.
[13] 胡路,赵天杰,施建成,李尚楠,樊东,王平凯,耿德源,肖青,崔倩,陈德清. 基于地基微波辐射观测的土壤水分反演算法评估[J]. 遥感技术与应用, 2020, 35(1): 74-84.
[14] 劳从坤,杨娜,徐少博,汤燕杰,张恒杰. 反演策略对SMOS土壤水分反演算法的影响研究[J]. 遥感技术与应用, 2020, 35(1): 65-73.
[15] 徐卫星,薛华柱,靳华安,李爱农. 融合遥感先验信息的叶面积指数反演[J]. 遥感技术与应用, 2019, 34(6): 1235-1244.