基于多光谱遥感的典型绿洲棉田春季土壤盐分反演及验证

doi:10.12118/j.issn.1000-6060.2021.477

摘要/Abstract

摘要：

为探索快速提取典型绿洲棉田土壤盐分的有效方法,获取区域尺度的土壤盐渍化特征及空间分布,进而为土壤盐渍化防治提供参考。以新疆兵团农二师31团为研究区域,2019、2021年春季Landsat 8 OLI多光谱影像和野外实测土壤含盐量为数据源,将波段组、光谱指数组和全变量组作为模型输入变量组,采用多元逐步回归（Multiple stepwise regression, MSR）、偏最小二乘回归（Partial least squares regression, PLSR）、极限学习机（Extreme learning machine, ELM）、支持向量机（Support vector machine, SVM）和BP神经网络（Back propagation neural network, BPNN）构建基于3个输入变量组的土壤盐分遥感反演模型,探究输入变量和建模方法对模型精度的影响效果,通过对比确定春季土壤盐分最优反演模型,定量反演地表土壤含盐量。结果表明：（1）研究区主要为非盐化土和轻度盐化土,总样本变异系数为0.67,呈中等变异性;光谱反射率与土壤盐渍化程度的关系表现为土壤盐渍化越重,光谱反射率越高。（2）海岸波段（b1）、蓝波段（b2）、绿波段（b3）、红波段（b4）和盐分指数（SI1、SI2、SI3、SI4、S3、S4、S5）均通过显著性检验P<0.01,相关系数均达到0.4以上。（3）所有模型中,基于全变量组建立的BPNN反演模型精度最高,建模集R²为0.705;验证集R²为0.556。（4）由反演结果可知,2019、2021年春季耕作区土壤主要为非盐化土,分别占耕作区总面积的55.55%和64.62%,其次为轻度盐化土,分别占44.31%和35.17%;2021年土壤盐渍化程度较2019年有所减轻。

关键词: 多光谱遥感反演, 土壤盐分, 光谱反射率, 变量组, 机器学习

Abstract:

This study aimed to explore an effective method for extracting soil salinity from cotton fields in oasis and determine the characteristics and spatial distribution of soil salinization on a regional scale to provide reference for soil salinization control. The 31^st Regiment of the 2^nd Division of Xinjiang Production and Construction Corps was taken as the research area; Landsat 8 OLI multispectral images and field measurements of soil salinity in spring 2019 and 2021 were taken as data sources; and the band group, spectral index group, and total variable group were taken as the model input variable group. Multiple stepwise regression (MSR), partial least squares regression, extreme learning machine, support vector machine, and back propagation neural network (BPNN) were used to construct a remote sensing inversion model of soil salt based on the three input variable groups. Precision evaluation was conducted, and the effects of input variables and modeling methods on model accuracy were explored. The best inversion model of soil salt in spring was determined through comparison and quantitatively inverted surface soil salt content. Results showed that (1) the study area was mainly composed of non-salinized soil and slightly salinized soil, and the coefficient of variation of total samples was 0.67, implying moderate variability. The relationship between spectral reflectance and soil salinization is as follows: serious soil salinization leads to great spectral reflectance. (2) Significance tests were conducted on coastal band(b1), blue(b2), green(b3), and red(b4) and the salinity indices of SI1, SI2, SI3, SI4, S3, S4, and S5 (P<0.01). The obtained correlation coefficients were all above 0.4, which can represent soil salinity to a certain extent. (3) Among the six linear regression models, the MSR model established on the band group had the best inversion effect, and the BPNN inversion model established on the full variable group had the highest accuracy. (4) According to the inversion results, the soil in spring 2019 and 2021 was mainly nonsalinized soil accounting for 55.55% and 64.62% of the total tillage area, respectively, followed by mild salinized soil accounting for 44.31% and 35.17%, respectively. Soil salinization in 2021 was reduced compared with that in 2019.

Key words: multispectral remote sensing inversion, soil salinity, spectral reflectance, variable group, machine learning

刘旭辉,白云岗,柴仲平,张江辉,丁邦新,江柱. 基于多光谱遥感的典型绿洲棉田春季土壤盐分反演及验证[J]. 干旱区地理, 2022, 45(4): 1165-1175.

LIU Xuhui,BAI Yungang,CHAI Zhongping,ZHANG Jianghui,DING Bangxin,JIANG Zhu. Inversion and validation of soil salinity based on multispectral remote sensing in typical oasis cotton field in spring[J]. Arid Land Geography, 2022, 45(4): 1165-1175.

图/表 10

图1

表1

光谱参量及相关计算公式"

光谱参量		计算公式	参考文献
波段	b1	CA	[16]
	b2	B	[16]
	b3	G	[16]
	b4	R	[16]
	b5	NIR	[16]
	b6	SWIR1	[16]
	b7	SWIR2	[16]
盐分指数	NDSI	$R - N I R R + N I R$	[14]
	SI-T	$100 × R - N I R$	[14]
	SI1	$G × R$	[2]
	SI2	$G 2 + R 2 + N I R 2$	[2]
	SI3	$G 2 + R 2$	[2]
	SI4	$B - G 2 + G - R 2$	[10]
	S1	$B / R$	[2]
	S2	$B - R / B + R$	[2]
	S3	$G × R / B$	[2]
	S4	$B × R$	[2]
	S5	$B × R / G$	[2]
	S6	$R × N I R / G$	[2]

表1

表2

图2

表3

表4

表5

图3

图4

表6

参考文献 37

[1]	梁萌, 米晓军, 李晨华, 等. 新疆准噶尔盆地未开垦盐碱土盐分与盐生植被多样性分析[J]. 干旱区地理, 2022, 45(1): 185-196.
	[ Liang Meng, Mi Xiaojun, Li Chenhua, et al. Salinity characteristics and halophytic vegetation diversity of uncultivated saline-alkali soil in Junggar Basin, Xinjiang[J]. Arid Land Geography, 2022, 45(1): 185-196. ]
[2]	孙亚楠, 李仙岳, 史海滨, 等. 基于多源数据融合的盐分遥感反演与季节差异性研究[J]. 农业机械学报, 2020, 51(6): 169-180.
	[ Sun Ya’nan, Li Xianyue, Shi Haibin, et al. Remote sensing inversion of soil salinity and seasonal difference analysis based on multi-source data fusion[J]. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(6): 169-180. ]
[3]	李晓明, 杨劲松, 杨奇勇. 基于电磁感应的南疆典型田块土壤盐分空间异质性研究[J]. 水土保持学报, 2011, 25(1): 167-170.
	[ Li Xiaoming, Yang Jinsong, Yang Qiyong. Spatial heterogeneity of soil salinity in a typical field in south Xinjiang based on electromagnetic induction[J]. Journal of Soil and Water Conservation, 2011, 25(1): 167-170. ]
[4]	李彪, 王耀强. 土壤盐渍化雷达反演模拟研究[J]. 干旱区资源与环境, 2015, 29(8): 180-184.
	[ Li Biao, Wang Yaoqiang. Radar inversion and simulation of salty soil salinization[J]. Journal of Arid Land Resources and Environment, 2015, 29(8): 180-184. ]
[5]	Sreenivas K, Venkataratnam L, Rao P V N. Dielectric properties of salt-affected soils[J]. International Journal of Remote Sensing, 1995, 16(4): 641-649. doi: 10.1080/01431169508954431
[6]	Ding J L, Yu D L. Monitoring and evaluating spatial variability of soil salinity in dry and wet seasons in the Werigan-Kuqa Oasis, China, using remote sensing and electromagnetic induction instruments[J]. Geoderma, 2014, 235: 316-322.
[7]	Rao B R M, Ravi Sankar T, Dwivedi R S, et al. Spectral behaviour of salt-affected soils[J]. International Journal of Remote Sensing, 1995, 16(12): 2125-2136. doi: 10.1080/01431169508954546
[8]	奚雪, 赵庚星, 高鹏, 等. 基于Sentinel卫星及无人机多光谱的滨海冬小麦种植区土壤盐分反演研究--以黄三角垦利区为例[J]. 中国农业科学, 2020, 53(24): 5005-5016.
	[ Xi Xue, Zhao Gengxing, Gao Peng, et al. Inversion of soil salinity in coastal winter wheat growing area based on sentinel satellite and unmanned aerial vehicle multi-spectrum: A case study in Kenli District of the Yellow River delta[J]. Scientia Agricultura Sinica, 2020, 53(24): 5005-5016. ]
[9]	张素铭, 赵庚星, 王卓然, 等. 滨海盐渍区土壤盐分遥感反演及动态监测[J]. 农业资源与环境学报, 2018, 35(4): 349-358.
	[ Zhang Suming, Zhao Gengxing, Wang Zhuoran, et al. Remote sensing inversion and dynamic monitoring of soil salt in coastal saline area[J]. Journal of Agricultural Resources and Environment, 2018, 35(4): 349-358. ]
[10]	周晓红, 张飞, 张海威, 等. 艾比湖湿地自然保护区土壤盐分多光谱遥感反演模型[J]. 光谱学与光谱分析, 2019, 39(4): 1229-1235. doi: 10.3964/j.issn.1000-0593(2019)04-1229-07
	[ Zhou Xiaohong, Zhang Fei, Zhang Haiwei, et al. A study of soil salinity inversion based on multispectral remote sensing index in Ebinur Lake Wetland Nature Reserve[J]. Spectroscopy and Spectral Analysis, 2019, 39(4): 1229-1235. ] doi: 10.3964/j.issn.1000-0593(2019)04-1229-07
[11]	张晓光, 姜子璇, 孔繁昌. 滨海盐渍土可见近红外高光谱特征[J]. 遥感技术与应用, 2019, 34(4): 816-821.
	[ Zhang Xiaoguang, Jiang Zixuan, Kong Fanchang. Hyperspectral characteristics of coastal saline soil with visible/near infrared spectroscopy[J]. Remote Sensing Technology and Application, 2019, 34(4): 816-821. ]
[12]	Zhang T T, Qi J G, Gao Y, et al. Detecting soil salinity with MODIS time series VI data[J]. Ecological Indicators, 2015, 52: 480-489. doi: 10.1016/j.ecolind.2015.01.004
[13]	Alexakis D D, Daliakopoulos I N, Panagea I S, et al. Assessing soil salinity using WorldView-2 multispectral images in Timpaki, Crete, Greece[J]. Geocarto International, 2016, 33(4): 321-338. doi: 10.1080/10106049.2016.1250826
[14]	张智韬, 魏广飞, 姚志华, 等. 基于无人机多光谱遥感的土壤含盐量反演模型研究[J]. 农业机械学报, 2019, 50(12): 151-160.
	[ Zhang Zhitao, Wei Guangfei, Yao Zhihua, et al. Soil salt inversion model based on UAV multispectral remote sensing[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(12): 151-160. ]
[15]	冯雪力, 刘全明. 基于多源遥感协同反演的区域性土壤盐渍化监测[J]. 农业机械学报, 2018, 49(7): 127-133.
	[ Feng Xueli, Liu Quanming. Regional soil salinity monitoring based on multi-source collaborative remote sensing data[J]. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(7): 127-133. ]
[16]	刘恩, 王军涛, 常步辉, 等. 小开河引黄灌区土壤盐渍化定量遥感反演[J]. 中国农村水利水电, 2019(12): 20-24.
	[ Liu En, Wang Juntao, Chang Buhui, et al. Quantitative remote sensing inversion of soil salinization in Xiaokaihe Yellow River irrigation district[J]. China Rural Water and Hydropower, 2019(12): 20-24. ]
[17]	张智韬, 杜瑜燕, 劳聪聪, 等. 基于雷达遥感的不同深度土壤含盐量反演模型[J]. 农业机械学报, 2020, 51(10): 243-251.
	[ Zhang Zhitao, Du Yuyan, Lao Congcong, et al. Inversion model of soil salt content in different depths based on radar remote sensing[J]. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(10): 243-251. ]
[18]	何宝忠, 丁建丽, 刘博华, 等. 渭库绿洲土壤盐渍化时空变化特征[J]. 林业科学, 2019, 55(9): 185-196.
	[ He Baozhong, Ding Jianli, Liu Bohua, et al. Spatiotemporal variation of soil salinization in Weigan-Kuqa River Delta Oasis[J]. Scientia Silvae Sinicae, 2019, 55(9): 185-196. ]
[19]	阿尔达克·克里木, 塔西甫拉提·特依拜, 张东, 等. 基于高光谱的ASTER影像土壤盐分模型校正及验证[J]. 农业工程学报, 2016, 32(12): 144-150.
	[ Kelimu Ardak, Tiyip Tashpolat, Zhang Dong, et al. Calibration and validation of soil salinity estimation model based on measured hyperspectral and ASTER image[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(12): 144-150. ]
[20]	丁邦新, 白云岗, 柴仲平, 等. 塔里木河下游绿洲灌区土壤盐渍化特征及季节性变化规律[J]. 水土保持通报, 2020, 40(2): 77-84.
	[ Ding Bangxin, Bai Yungang, Chai Zhongping, et al. Soil salinization characteristics and its seasonal variation in oasis irrigation district of lower reaches of Tarim River[J]. Bulletin of Soil and Water Conservation, 2020, 40(2): 77-84. ]
[21]	叶勤, 姜雪芹, 李西灿, 等. 基于高光谱数据的土壤有机质含量反演模型比较[J]. 农业机械学报, 2017, 48(3): 164-172.
	[ Ye Qin, Jiang Xueqin, Li Xican, et al. Comparison on inversion model of soil organic matter content based on hyperspectral data[J]. Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(3): 164-172. ]
[22]	Marshall F E, Smith D T, Nickerson D M. Empirical tools for simulating salinity in the estuaries in Everglades National Park, Florida[J]. Estuarine Coastal & Shelf Science, 2011, 95(4): 377-387.
[23]	冯娟, 丁建丽, 杨爱霞, 等. 干旱区土壤盐渍化信息遥感建模[J]. 干旱地区农业研究, 2018, 36(1): 266-273.
	[ Feng Juan, Ding Jianli, Yang Aixia, et al. Remote sensing modeling of soil salinization information in arid areas[J]. Agricultural Research in the Arid Areas, 2018, 36(1): 266-273. ]
[24]	吴忠强, 毛志华, 王正, 等. 基于极限学习机的浅海水深遥感反演研究[J]. 海洋测绘, 2019, 39(3): 11-15.
	[ Wu Zhongqiang, Mao Zhihua, Wang Zheng, et al. Research on remote sensing inversion of shallow water depth based on extreme learning machine[J]. Hydrographic Surveying and Charting, 2019, 39(3): 11-15. ]
[25]	陆军胜, 陈绍民, 黄文敏, 等. 采用 SE_PLS_ELM 模型估算夏玉米地上部生物量和叶面积指数[J]. 农业工程学报, 2021, 37(18): 128-135.
	[ Lu Junsheng, Chen Shaomin, Huang Wenmin, et al. Estimation of aboveground biomass and leaf area index of summer maize using SE_PLS_ELM model[J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(18): 128-135. ]
[26]	姜雯, 吴陈. 基于改进粒子群算法的支持向量机遥感影像分类[J]. 江苏科技大学学报(自然科学版), 2020, 34(5): 66-72.
	[ Jiang Wen, Wu Chen. Remote sensing image classification by support vector machine based on improved particle swarm optimization[J]. Journal of Jiangsu University of Science and Technology (Natural Science Edition), 2020, 34(5): 66-72. ]
[27]	赵洪莹, 舒清态, 罗文秀, 等. 基于Landsat 8-OLI的高山松叶面积指数采样尺度优化分析[J]. 西南林业大学学报(自然科学版), 2021, 41(5): 114-120.
	[ Zhao Hongying, Shu Qingtai, Luo Wenxiu, et al. Optimization analysis of sampling scale of Pinus densata leaf area index based on Landsat 8-OLI[J]. Journal of Southwest Forestry University (Natural Sciences Edition), 2021, 41(5): 114-120. ]
[28]	李柏年, 吴礼斌. MATLAB数据分析方法[M]. 北京: 机械工业出版社, 2012: 170-171.
	[ Li Bainian, Wu Libin. MATLAB data analysis methods[M]. Beijing: China Machine Press, 2012: 170-171. ]
[29]	张录, 张芳, 熊黑钢, 等. 不同季节强碱土土壤呼吸影响因子分析与模型预测[J]. 干旱地区农业研究, 2017, 35(1): 71-78.
	[ Zhang Lu, Zhang Fang, Xiong Heigang, et al. Impact factor analysis and model prediction of strong alkaline soil respiration in different seasons[J]. Agricultural Research in the Arid Areas, 2017, 35(1): 71-78. ]
[30]	新疆维吾尔自治区农业厅, 新疆土壤普查办公室. 新疆土壤[M]. 北京: 科学出版社, 1996: 151-521.
	[Xinjiang Department of Agriculture, Xinjiang Soil Survey Office. The soil of Xinjiang[M]. Beijing: Science Press, 1996: 151-521. ]
[31]	Peck A J. Note on the role of a shallow aquifer in dryland salinity[J]. Soil Research, 1978, 16(2): 237-240. doi: 10.1071/SR9780237
[32]	柳菲, 陈沛源, 于海超, 等. 民勤绿洲不同土地利用类型下土壤水盐的空间分布特征分析[J]. 干旱区地理, 2020, 43(2): 406-414.
	[ Liu Fei, Chen Peiyuan, Yu Haichao, et al. Spatial distribution characteristics of soil water and salt under different land use types in Minqin Oasis[J]. Arid Land Geography, 2020, 43(2): 406-414. ]
[33]	姜凌, 李佩成, 胡安焱, 等. 干旱区绿洲土壤盐渍化分析评价[J]. 干旱区地理, 2009, 32(2): 234-239.
	[ Jiang Ling, Li Peicheng, Hu Anyan, et al. Analysis and evaluation of soil salinization in oasis of arid region[J]. Arid Land Geography, 2009, 32(2): 234-239. ]
[34]	王雪梅, 周晓红. 渭干河-库车河三角洲绿洲棉田土壤盐分估算及遥感反演[J]. 干旱地区农业研究, 2018, 36(6): 250-254, 262.
	[ Wang Xuemei, Zhou Xiaohong. Estimation and inversion modeling of salinity of cotton field soil using remote sensing in the delta oasis of Weigan and Kuqa Rivers[J]. Agricultural Research in the Arid Areas, 2018, 36(6): 250-254, 262. ]
[35]	边玲玲, 王卷乐, 郭兵, 等. 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.
	[ Bian Lingling, Wang Juanle, Guo Bing, et al. Remote sensing extraction of soil salinity in Yellow River delta Kenli County based on feature space[J]. Remote Sensing Technology and Application, 2020, 35(1): 211-218. ]
[36]	贾萍萍, 张俊华, 孙媛, 等. 基于实测高光谱和Landsat 8 OLI影像的土壤盐化和碱化程度反演研究[J]. 土壤通报, 2020, 51(3): 511-520. doi: 10.1046/j.1365-2389.2000.00318.x
	[ Jia Pingping, Zhang Junhua, Sun Yuan, et al. Inversion of soil salinity and pH degree based on measured hyper-spectral and Landsat 8 OLI image[J]. Chinese Journal of Soil Science, 2020, 51(3): 511-520. ] doi: 10.1046/j.1365-2389.2000.00318.x
[37]	杨小虎, 罗艳琴, 杨海昌, 等. 玛纳斯河流域绿洲农田土壤盐分反演及空间分布特征[J]. 干旱区资源与环境, 2021, 35(2): 156-161.
	[ Yang Xiaohu, Luo Yanqin, Yang Haichang, et al. Soil salinity retrieval and spatial distribution of oasis farmland in Manasi River Basin[J]. Journal of Arid Land Resources and Environment, 2021, 35(2): 156-161. ]

数据集	样本数/个				土壤含盐量/g·kg^-1			标准差	变异系数
数据集	总样本	非盐化土	轻度盐化土	中度盐化土	最小值	最大值	平均值	标准差	变异系数
总样本	77	50	21	6	0.37	11.82	3.02	2.03	0.67
建模集	51	33	14	4	0.42	11.82	3.05	2.07	0.68
验证集	26	17	7	2	0.37	9.64	2.97	1.95	0.66

变量组	建模方法	表达式	建模集（n=51）		验证集（n=26）
变量组	建模方法	表达式	R²	RMSE	R²	RMSE
波段组	MSR	Y=624.08b1-440.71b2-29.42	0.411	1.588	0.473	1.414
波段组	PLSR	Y=40.75b1+35.51b2+25.32b3+19.94b4-21.19	0.314	1.713	0.496	1.382
光谱指数组	MSR	Y=-94.05SI2+170.77SI4+246.21S5-14.30	0.341	1.679	0.377	1.544
光谱指数组	PLSR	Y=13.32SI1+6.02SI2+9.3SI3+26.55SI4+7.86S3+16S4+16.58S5-16.74	0.269	1.768	0.449	1.447
全变量组	MSR	Y=592.81b1-318.31b3-75.19SI2+462.30SI4-23.33	0.439	1.549	0.359	1.562
全变量组	PLSR	Y=15.29b1+13.33b2+9.5b3+7.48b4+8.53SI1+3.86SI2+5.96SI3+17.01SI4+5.03S3+10.25S4+10.62S5-18.73	0.290	1.742	0.470	1.418

变量组	建模方法	建模集（n=51）		验证集（n=26）
变量组	建模方法	R²	RMSE	R²	RMSE
波段组	ELM	0.617	1.279	0.533	1.998
	SVM	0.782	0.985	0.476	1.996
	BPNN	0.736	1.067	0.538	1.460
光谱指数组	ELM	0.621	1.274	0.508	1.986
	SVM	0.769	0.997	0.409	2.001
	BPNN	0.730	1.080	0.459	1.567
全变量组	ELM	0.540	1.403	0.490	1.565
	SVM	0.533	1.515	0.488	1.658
	BPNN	0.705	1.133	0.556	1.409

土壤类型	土壤含盐量/g·kg^-1	2019年		2021年
土壤类型	土壤含盐量/g·kg^-1	像元个数	面积占比/%	像元个数	面积占比/%
非盐化土（0~3 g·kg^-1）	0~1	4508	0.35	14	0.00
	1~2	92995	7.17	366240	28.25
	2~3	622671	48.03	471548	36.37
轻度盐化土（3~6 g·kg^-1）	3~4	428918	33.08	398125	30.71
	4~5	128548	9.92	47257	3.65
	5~6	16989	1.31	10556	0.81
中度盐化土（6~10 g·kg^-1）	6~10	1848	0.14	2737	0.21