路域植被等效水厚度估算模型研究

doi:10.12082/dqxxkx.2020.190254

摘要/Abstract

摘要：

植被等效水厚度对路域生态环境的监测评估具有重要意义。本研究以湖南醴潭高速一段为研究对象,以地面实测光谱和等效水厚度以及PRO4SAIL模拟光谱和模拟等效水厚度为数据源,利用PRO4SAIL冠层模型模拟光谱与地面实测光谱建立12种常用水分指数,引入随机森林算法对水分指数与等效水厚度进行重要性分析,得到12种水分指数的重要性排序;利用调整R ²确定建立等效水厚度估算模型中输入水分指数的最佳个数;在优选水分指数基础上,以PRO4SAIL模拟光谱计算得到水分指数和等效水厚度为训练集,分别构建随机森林耦合偏最小二乘（RF-PLS）、随机森林耦合支持向量机（RF-SVM）和随机森林耦合遗传算法优化支持向量机（RF-GA-SVM）等效水估算模型,并用地面实测等效水厚度对估算模型进行精度验证与分析。结果表明：RF-SVM估算模型中输入重要性前9的水分指数（NDWI、NMDI、SRWI、SR、NDII、WI、DWI、MSI、SAVI）时,调整R ²最高,验证集决定系数为0.8877;RF-PLS和RF-GA-SVM估算模型中输入重要性前4的水分指数（NDWI、NMDI、SRWI、SR）时,调整R ²最高,验证集决定系数分别为0.8053、0.8952,其中RF-GA-SVM模型估算等效水厚度效果最佳,其精度满足路域植被等效水厚度监测要求。本文研究成果为等效水厚度估算提供一种有效且精确的方法,同时为发展基于高光谱遥感的路域环境监测提供重要支撑。

关键词: 等效水厚度, 随机森林, PRO4SAIL模型, 水分指数, 机器学习, 路域植被, 湖南醴潭高速

Abstract:

Vegetation water content is an important evaluation index for vegetation health monitoring. The Equivalent Water Thickness (EWT) of vegetation is of great significance for the monitoring and evaluation of the ecological conditions in the road area, it could provide a guideline in road area environment management. Taking the Litan highway in Hunan Provinces as an example, this research used field data of canopy reflectance and equivalent water thickness of vegetation on the ground, and simulated reflectance and simulated equivalent water thickness established by PRO4SAIL. In total, 12 kinds of water indices were established by using the simulated reflectance of the PRO4SAIL canopy model and the ground measured reflectance. The random forest algorithm (RF) was introduced to analyze the importance of the 12 water indices and equivalent water thickness. We determined the ordination between water indices and equivalent water thickness as well as the optimal number of input water index in the equivalent water thickness estimation model by using the adjusted coefficient of determination. Based on the selected water index, the water index and equivalent water thickness were calculated by the PRO4SAIL simulation reflectance as the training set. Three equivalent water thickness estimation models were constructed: Random Forest Coupled Partial Least Squares (RF-PLS), Random Forest Coupled Support Vector Machine (RF-SVM) model, and Random Forest coupled Genetic Algorithm to optimize the Support Vector Machine (RF-GA-SVM) model. The applicability of 12 water indices in the estmation of equivalent water thickness in road-area of vegetation was also analyzed. The accuracy of the model was validated by measured equivalent water thickness on the ground. The experimental results show: (1) The adjusted determination coefficient of RF-SVM model was the highest, established by Normalized Difference Water Index (NDWI), Normalized Multi-band Drought Index (NMDI), Simple Ratio Water Index (SRWI), Simple Ratio (SR), Normalized Difference Infrared Index (NDII), Water Index (WI), Dattwater Index (DWI), Moisture Stress Index (MSI) and Soil Adjusted Vegetation Index (SAVI), with the determination coefficient of verification set reaching 0.8877. (2) The RF-PLS and RF-GA-SVM models with the four water indices of NDWI, NMDI, SRWI, and SR had the highest adjusted determination coefficient, with the validation set's determination coefficients reaching 0.8053 and 0.8952, respectively. (3) Among them, the RF-GA-SVM model was the best for estimating equivalent water thickness, which met the requirements of vegetation equivalent water thickness monitoring in road area. Our findings provide an effective and accurate method for the estimation of equivalent water thickness, and provide support for road area environment monitoring based on hyper-spectral remote sensing.

Key words: Equivalent water thickness, random forest, PRO4SAILmodel, water index, machine learning, vegetation of road area, Litan highway in Hunan Province

郭云开, 张晓炯, 许敏, 刘雨玲, 钱佳, 章琼. 路域植被等效水厚度估算模型研究[J]. 地球信息科学学报, 2020, 22(2): 308-315.DOI:10.12082/dqxxkx.2020.190254

GUO Yunkai, ZHANG Xiaojiong, XU Min, LIU Yuling, QIAN Jia, ZHANG Qiong. Estimation Model of Equivalent Water Thickness in the Road Area[J]. Journal of Geo-information Science, 2020, 22(2): 308-315.DOI:10.12082/dqxxkx.2020.190254

图/表 6

表1

表2

12种水分指数的计算方法列表

水分指数	计算公式	公式编号	参考文献
新近红外肩部区域光谱比值指数（Spectral Ratio Index in theNIR Shoulder Region, NSRI）	$NSRI = ρ 890 ρ 780$	（1）	[15]
水分指数（Water Index, WI）	$WI = ρ 900 ρ 970$	（2）	[15]
归一化差值植被指数（Normalized difference vegetation index, NDVI）	$NDVI = (ρ 860 - ρ 680) (ρ 860 + ρ 680)$	（3）	[15]
归一化差值红外指数（Normalized Difference Infrared Index, NDII）	$NDII = (ρ 850 - ρ 1650) (ρ 850 + ρ 1650)$	（4）	[15]
Datt水分指数（DattWater Index, DWI）	$DWI = (ρ 816 - ρ 2218) (ρ 816 + ρ 2218)$	（5）	[15]
简单水比指数（Simple Ratio Water Index, SRWI）	$SRWI = ρ 860 ρ 1240$	（6）	[15]、[20]
湿度胁迫指数（Moisture Stress Index, MSI）	$MSI = ρ 1600 ρ 819$	（7）	[15]、[19]
水比指数（Simple Ratio, SR）	$SR = ρ 895 ρ 675$	（8）	[15]
归一化差值水分指数（Normalized Difference Water Index, NDWI）	$NDWI = (ρ 860 - ρ 1240) (ρ 860 + ρ 1240)$	（9）	[15]
归一化多波段干旱指数（NormalizedMulti-band Drought Index, NMDI）	$NMDI = ρ 860 - (ρ 1640 - ρ 2130) ρ 860 + (ρ 1640 + ρ 2130)$	（10）	[15]
全球植被水分指数（Global Vegetation Moisture Index, GVWI）	$GVWI = (ρ 820 + 0.1) - (ρ 1600 - 0.02) (ρ 820 + 0.1) + (ρ 1600 + 0.02)$	（11）	[18]
土壤调整植被指数（Soil Adjusted Vegetation Index, SAVI）	$SAVI = ρ 860 - ρ 680 × (1 + L) (ρ 860 + ρ 860 + L), L = 0.5$	（12）	[18]

表2

表3

图1

表4

图2

参考文献 24

[1]	张峰, 周广胜 . 植被含水量高光谱遥感监测研究进展[J]. 植物生态学报, 2018,42(5):517-525.
	[ Zhang F, Zhou G S . Research progress on monitoring vegetation water content by using hyperspectral remote sensing[J]. Chinese Journal of Plant Ecology, 2018,42(5):517-525. ]
[2]	Dzikiti S, Verreynne J S, Stuckens J , et al. Determining the water status of Satsuma mandarin trees "citrus unshiumarcovitch" using spectral indices and by combining hyperspectral and physiological data[J]. Agricultural and Forest Meteorology, 2009,150(3):369-379.
[3]	De Jong S M, Addink E A, Hoogenboom P , et al. The spectral response of Buxussempervirens to different types of environmental stress: A laboratory experiment[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2012,74:56-65.
[4]	刘良云, 王纪华, 张永江 , 等. 叶片辐射等效水厚度计算与叶片水分定量反演研究[J]. 遥感学报, 2007,11(3):289-295.
	[ Liu L Y, Wang J H, Zhang Y J , et al. Detection of leaf EWT by calculating REWT from reflectance spectra[J]. Journal of Remote Sensing, 2007,11(3):289-295. ]
[5]	赵英时等. 遥感应用分析原理与方法[M]. 北京: 科学出版社, 2003.
	[ Zhao Y S , et al. Principles and methods of remote sensing application analysis[M]. Beijing: Science Press, 2003. ]
[6]	Seelig H D, Hoehn A, Stodieck L S , et al. Relations of remote sensing leaf water indices to leaf water thickness in cowpea, bean, and sugarbeet plants[J]. Remote Sensing of Environment, 2007,112(2):445-455.
[7]	Clevers J G P W, Kooistra L, Schaepman M E . Using spectral information from the NIR water absorption features for the retrieval of canopy water content[J]. International Journal of Applied Earth Observations and Geo-information, 2008,10(3):388-397.
[8]	李丹娜, 郭云开, 朱善宽 , 等. 利用PRO4SAIL与支持向量机回归的组合模型反演植被等效水厚度[J]. 测绘通报, 2017(8):50-55.
	[ Li D N, Guo Y K, Zhu S K , et al. Quantitative inversion vegetation equivalent water thickness by combined model of PRO4SAIL and the support vector regression[J]. Bulletin of Surveying and Mapping, 2017(8):50-55. ]
[9]	侯学会, 王猛, 刘思含 , 等. 基于Prosail模型和Landsat 8数据的小麦冠层含水量反演比较[J]. 麦类作物学报, 2018,38(4):493-497.
	[ Hou X H, Wang M, Liu S H , et al. Comparison between prosailmodel and Landsat 8 images in inversion of water content of wheat canopy[J]. Journal of Triticeae Crops, 2018,38(4):493-497.]
[10]	Liu L L, Zhang S, Zhang B . Evaluation of hyperspectral indices for retrieval of canopy equivalent water thickness and gravimetric water content[J]. International Journal of Remote Sensing, 2016,37(14):3384-3399.
[11]	Gao B C . NDWI: A normalized difference water index for remote sensing of vegetation liquid water from space[J]. Remote Sensing of Environment, 1996,58(3):257-266.
[12]	程晓娟, 杨贵军, 徐新刚 , 等. 新植被水分指数的冬小麦冠层水分遥感估算[J]. 光谱学与光谱分析, 2014,34(12):3391-3396.
	[ Cheng X J, Yang G J, Xu X G , et al. Estimating canopy water content in wheat based on new vegetation water index[J]. Spectroscopy and Spectral Analysis, 2014,34(12):3391-3396. ]
[13]	郑兴明, 丁艳玲, 赵凯 , 等. 基于Landsat 8 OLI数据的玉米冠层含水量反演研究[J]. 光谱学与光谱分析, 2014,34(12):3385-3390.
	[ Zhen X M, Ding Y L, Zhao K , et al. Estimation of vegetation water content from Landsat 8 OLI data[J]. Spectroscopy and Spectral Analysis, 2014,34(12):3385-3390. ]
[14]	Liang L, Di L P, Zhang L P , et al. Estimation of crop LAI using hyperspectral vegetation indices and a hybrid inversion method[J]. Remote Sensing of Environment, 2015,165:123-134.
[15]	张春兰, 杨贵军, 李贺丽 , 等. 基于随机森林算法的冬小麦叶面积指数遥感反演研究[J]. 中国农业科学, 2018,51(5):855-867.
	[ Zhang C L, Yang G J, Li H L , et al. Remote sensing inversion of leaf area index of winter wheat based on random forest algorithm[J]. Scientia Agricultural Sinica, 2018,51(5):855-867. ]
[16]	Jacquemoud S, Verhoef W . PROSPECT+SAIL models: A review of use for vegetation characterization[J]. Remote Sensing of Environment, 2009,113:56-66.
[17]	Jacquemoud S, Baret F . Prospect: A model of leaf optical properties spectra[J]. Remote Sensing of Environment, 1990,34:75-91.
[18]	Hedia C, Florent M, Abdelaziz H . Regional equivalent water thickness modeling from remote sensing across a tree Cover/LAI gradient in mediterranean forests of northern Tunisia[J]. Molecular Diversity Preservation International (MDPI), 2015,7(2):1937-1961.
[19]	Pietro C, Stéphane F, Stefano T S et al Detecting vegetation leaf water content using reflectance in the optical domain[J]. Remote Sensing of Environment, 2001,77(1):22-33.
[20]	Zarco-Tejada P J, Rueda C A, Ustin S A . Water content estimation in vegetation with MODIS reflectance data and model inversion methods[J]. Remote Sensing of Environment, 2003,85(1):109-124.
[21]	牛亚莉 . 遗传优化支持向量机的交通流量预测模型[J]. 微型电脑应用, 2017,33(8):72-74.
	[ Niu Y L . Traffic flow forecasting model by Support Vector Machine optimized by genetic algorithm[J]. Microcomputer Applications, 2017,33(8):72-74. ]
[22]	付俊娥, 苏庆珣, 潘世兵 , 等. 基于支持向量机理论的地下水动态遥感监测模型与应用[J]. 地球信息科学学报, 2010,12(4):466-472.
	[ Fu J E, Su Q X, Pan S B , et al. Support vector machine based ground water level monitoring model by using remote scence images[J]. Journal of Geo-information Science, 2010,12(4):466-472. ]
[23]	沈润平, 郭佳, 张婧娴 , 等. 基于随机森林的遥感干旱监测模型的构建[J]. 地球信息科学学报, 2017,19(1):125-133.
	[ Shen R J, Guo J, Zhang J X , et al. Construction of a drought monitoring model using the randform forest based remote sensing[J]. Journal of Geo-information Science, 2017,19(1):125-133. ]
[24]	李鹏霞 . 基于偏最小二乘和支持向量机的生物分子的定量结构—活性/性质关系研究[D]. 兰州:西北师范大学, 2011.
	[ Li X P . Quantitative structure-activity/property relationship studies in biomolecules based on partial least squares and support vector machine[D]. Lanzhou: Northwest Normal University, 2011. ]

输入参数	输入值
结构参数（N）	1.22
叶绿素含量（Cab）/（μg/cm²）	42.93
类胡萝卜素含量（Car）/（μg/cm²）	8
褐色素含量（Cbrown）/（μg/cm²）	0
干物质重量（Cm）/（μg/cm²）	0.02316
平均叶倾角（LAD）	30
叶面积指数（LAI）	4~6,步长：0.2
等效水厚度（EWT）	0.01~0.03,步长：0.001
热点参数（hspot）	0.15
太阳天顶角（θs）/°	23.9
观测天顶角（θv）/°	0
土壤反射率（rsoil）	0.2

反演模型	建模样本		预测样本
反演模型	自变量	应变量	自变量	应变量
PLS	PRO4SAIL模拟NDWI、NMDI、SRWI、SR	PRO4SAIL模拟EWT	实测NDWI、NMDI、SRWI、SR	实测EWT
SVM	PRO4SAIL模拟NDWI、NMDI、SRWI、SR、NDII、WI、DWI、MSI、SAVI	PRO4SAIL模拟EWT	实测NDWI、NMDI、SRWI、SR、NDII、WI、DWI、MSI、SAVI	实测EWT
GA-SVM	PRO4SAIL模拟NDWI、NMDI、SRWI、SR	PRO4SAIL模拟EWT	实测NDWI、NMDI、SRWI、SR	实测EWT