基于GF-1影像的渭-库绿洲外围土壤含盐量定量反演研究

苏雯 丁建丽 杨爱霞

PDF(709 KB)
中国农村水利水电 ›› 2017 ›› (2) : 9-13.
农田水利

基于GF-1影像的渭-库绿洲外围土壤含盐量定量反演研究

  • 苏雯1,2,丁建丽1,2,杨爱霞1,2
作者信息 +

Quantitative inversion of Soil Salinity in Weigan-Kuqa River Oasis Based on GF-1 Image

  • SU Wen1,2 ,DING Jian-li 1,2 ,YANG Ai-xia1,2 
Author information +
稿件信息 +

摘要

为探讨国产GF-1卫星影像在干旱区土壤盐渍化监测中的适用性, 本文以渭-库绿洲外围荒漠交错带为研究对象,利用BP神经网络和RBF神经网络2种建模算法,以GF-1影像的4个波段的反射率,及影像提取的归一化差异植被指数(NDVI)、差值植被指数(DVI)、土壤调节植被指数(SAVI)、及盐度指数(SI1、SI2、SI-T)共十个指标构建土壤含盐量反演模型。结果表明:(1)在2种算法中,BP神经网络模型预测精度最高,R2为0.818,RMSE为0.194。(2)发现利用植被指数更能提高模型的预测精度。(3)利用BP神经网络预测模型反演研究区的土壤含盐量,发现预测情况与研究区实际情况相符,说明利用GF-1数据结合BP神经网络构建的反演模型适用于监测研究区土壤盐渍化问题。

Abstract

The applicability of domestic GF-1 satellite images in the monitoring of soil salinization is discussed,using BP neural network and RBF neural network modeling method with ten indexes including four bands reflectance of GF-1 images、the normalized difference vegetation index ( NDVI) 、difference vegetation index ( DVI) 、soil regulating vegetation index ( SAVI) and salinity index ( SI1,SI2,SI-T) . The results show that: The BP neural network model with the highest prediction accuracy is the highest in the 2 kinds of algorithms,R2 is 0.818,RMSE is 0.194. The vegetation index can improve the prediction accuracy of the model effectively. Using BP neural network to predict the soil salinity in the study area,found that soil containing salt content is consistent with the reality of the study area. It shows that it is feasible to use GF -1 data combined with BP neural network model to monitor the salinization of study area.

关键词

GF-1 遥感影像 / BP 神经网络 / RBF 神经网络 / 土壤盐渍化监测

Key words

GF-1 / BP neural network / RBF neural network / soil salinization monitoring

基金

国 家 自 然 科 学 基 金 项 目 ( U1303381,4126090, 41161063) ; 

教育部长江学者计划创新团队计划 ( IRT1180) ; 

自治区科技支疆项目( 201504051064) ; 

自治区重点实验室专项基金( 2014KL005) ; 

高分辨率 对地观测重大专项( 民用部分) ( 95-Y40B02-9001-13 / 15-03-01) 。

引用本文

导出引用
苏雯 丁建丽 杨爱霞. 基于GF-1影像的渭-库绿洲外围土壤含盐量定量反演研究[J].中国农村水利水电, 2017(2): 9-13
SU Wen, DING Jian-li, YANG Ai-xia. Quantitative inversion of Soil Salinity in Weigan-Kuqa River Oasis Based on GF-1 Image[J].China Rural Water and Hydropower, 2017(2): 9-13

参考文献

[1] Wang Y G,Xiao D N,Li Y,et al.Response of salt accumulation in soilto groundwater changes at the oasis in the lower reaches of SangongRiver,Xinjiang[J].Acta Ecologica Sinica,2007,27( 10) .
[2] Acosta J A,Jansen B,Kalbitz K,et al. Salinity increases mobility ofheavy metals in soils[J].Chemosphere,2011,85( 8) : 1 318-1 324.
[3] Aldakheel Y Y.Assessing NDVI spatial pattern as related to irrigationand soil salinity management in Al-Hassa Oasis,Saudi Arabia[J].Journal of the Indian Society of Remote Sensing,2011,39( 2) : 171-180.
[4] Muyen Z,Moore G A,Wrigley RJ.Soil salinity and sodicity effects ofwaste water irrigation in South East Australia[J].Agricultural WaterManagement,2011,99( 1) : 33-41.
[5] Dehaan R L,TayLor G R.Field derived spctra of salinized soils andvegetation as indicators of irrigation induced soil salinization[J].Remote sensing of environment,2002,80( 3) : 406-417.
[6] Mettmicht GL,Zinck JA.Remote sensing of soil salinity potentials andconstraints[J].Remote Sensing of Environment,2003,85( 1) : 1-20.
[7] 陈 实,徐 斌,金云翔,等. 北疆农区土壤盐渍化遥感监测及其时空特征分析[J].地理科学,2015,35( 12) : 1 608-1 615.
[8] 翁永玲,宫 鹏.土壤盐渍化遥感应用研究进展[J].地理科学,2006,26( 3) : 371-372.
[9] Rao B R M,Ravisankar T,Dwivedi R S,et al. Spectral behaviour ofsa1taflfected soils[J].International Journal of Remote Sensing,1995,16( 12) : 2 125-2 136.
[10] Letey J,Hoffman G J,Hopmans J W,et al. Evaluation of soil salinityleaching requirement guidelines[J].Agricultural Water Management,2011,98( 4) : 502-506.
[11] Viscarra Rossel R A,Webster R.Discrimination of Australian soil horizons and classes from their visible-near infrared spectra[J].Eur JSoil Sci,2011,62: 637 – 647.
[12] Xiaoyan Guan,Shaoli Wang,Zhanyi Gao,et al.Dynamic prediction ofsoil salinization in an irrigation district based on the support vectormachine[J].Mathematical and Computer Modelling,2013,56: 719-724.
[13] 丁建丽,姚 远,王 飞.干旱区土壤盐渍化特征空间建模[J].生态学报,2014,34( 16) : 4 620-4 631.
[14] 张 芳,熊黑钢,栾福明,等.土壤碱化的实测光谱响应特征[J].红外与毫米波学报,2011,30( 1) : 55-60.
[15] 赵振亮,塔西甫拉提·特依,孙 倩,等.土壤光谱特征分析及盐渍化信息提取———以新疆渭干河/库车河绿洲为例[J].地理科学进展,2014,33( 2) : 280-288.
[16] 陈 实,高 超,徐 斌.新疆石河子农区土壤含盐量定量反演及其空间格局分析[J].地理研究,2014,33( 11) .
[17] 王 爽,丁建丽,王 璐,等.基于地表光谱建模的区域土壤盐渍化遥感监测研究[J].干旱区地理,2016,39( 1) : 190-198.
[19] 姚 远,丁建丽,张 芳,等.基于电磁感应技术的塔里木盆地北缘绿洲土壤盐分空间变异特性[J].中国沙漠,2014,34( 3) : 765-772.
[20] Ayetiguli Sidikea,Shuhe Zhao,Yuming Wen. Estimating soil salinity in Pingluo County of China using Quick Birddata and soil reflectance spectra[J]. International Journal of Applied Earth ObservationandGeoinformation,2014,26( 2) : 156-175.
[21] Yinghai Ke,Jungho Im,Junghee Lee,et al. Characteristics of Landsat 8 OLI-derived NDVI by comparison with multiple satellite sensors and in-situ observations[J]. Remote Sensing of Environment,2015,164: 298-313.
[22] Yao Zhengyang,Liu Jianjun,Zhao Xiaowen,et al. Spatial dynamicsof aboveground carbon stock in urban green space: A case study ofXi’an,China[J]. Journal of Arid Land,2015,7( 3) : 350-360.
[23] Huete A R. A soil -adjusted vegetation index ( SAVI) [J]. RemoteSensing of Environment,1988,25( 88) : 295-309.
[24] 王玉刚,肖笃宁,李 彦,等. 新疆三工河流域尾闾绿洲地下水变化与土壤积盐的响应[J]. 生态学报,2007,27 ( 10) .
[25] A E K Douaoui,H Nicolas,C Walter,et al. Detecting salinity hazards within a semiarid context by means of combining soil and remote-sensing data[J]. Geoderma,2006,134( 1) : 217-230.
[26] N K Tripathi,Brijesh. Spatial modelling of soil alkalinity in GIS environment using IRS data[C]∥ Paper presented at the 18th AsianConference in Remote Sensing,1997.
[27] 杨爱霞,丁建丽.新疆艾比湖湿地土壤有机碳含量的光谱测定方法对比[J].农业工程学报,2015,31( 18) : 162-168.
[28] 蒋宗礼.人工神经网络导论[M].北京: 高等教育出版社,2001: 7-13.
PDF(709 KB)

323

访问

0

引用

详细情况

段落导航
相关文章

/