基于辐射传输模型定量遥感反演水深和水底反射率
|
摘要
近年来我国水环境变化剧烈,面积持续萎缩,水质持续恶化,水体的功能和效益不断下降,成为制约社会经济可持续发展的瓶颈。
研究中采用辐射传输模型,对石头口门水库的水底反射率进行了反演。在水深较大(＞2m)的假设下,选取水库较清澈污染较小的区域,假设水体全部发生镜面反射,由水底反射率衰减模型求出水深。利用已求水深与离水辐射率的进行拟合,得出水深与离水辐射率的关系式。并将离水辐射率带入求出水库的所有区域水深。将所求水深代入水底反射率衰减模型中反演水底反射率。
以2000年ETM+卫星遥感图像为例,基于辐射传输模型对水深和水底反射率进行反演,对水深结果进行评价分析,反演出水底反射率,而且随之时间推移,水底反射率变小,可能污染物浓度逐渐增加。最后为了了解模型中各参数对反演结果的影响,进行敏感性分析,主要是研究太阳天顶角、水深的变化等对结果的影响。。
Water quality pollution is a serious problem,highly needed be solved,in China.Remote sensing monitoring of water-bottom,using the relationship between reflectance spectrum characteristics of water bodies and water bottom parameter,inverses water bottom parameter. There are three ways,which are customarily used to resolve the problem,including empirical,semi-empirical and mechanism modeling methods.At present,most researches are focused on the measurement of various water body spectrum characteristics,and regression analysis using remote sensing data and other methods to establish a visible light or the ratio of band parameters and the bottom of the empirical formula,in order to extract quantitative information of water bottom.The method can be carried out very easily,but it is depended upon the measured data of water surface and synchronous data of remote sensing. However,the accuracy of this method will be significantly influenced by the changing of the water sediment concentration.The principle of mechanism model,based on two relation ship: water component and amount of inherent optical,amount of inherent optical and apparent optical volume,simulates the distribution of light field in water and then inverse the bottom-water parameter.The theoretical basis of mechanism model is the principle of light transmission in water,therefore,mechanism model is an important method of remote sensing monitoring of water quality.The advantages of the mechanism model have a clear physical meaning,the bottom of the reliability of inversion parameters,and subjected to time and geographical constraints,applicability.Therefore,this study based on radioactive transfer model.
According to the radiation from the water-leaving radiance,water depth and bottom reflectance modeling is to achieve the bottom reflectance effective way of quantitative remote sensing At depths greater(＞2m) under the assumption that,in the clear waters of the assumption that spectrum reflection occurs,the bottom reflectance is 1.According to the water depth and rate of water-leaving radiance to establish the relationship between the depth of the statistical model,the request will be incorporated into the water depth of reflection attenuation model,then inverse water bottom reflectance.To Shitoukoumen Reservoir as an example,inverse water bottom reflectance use of water attenuation model.Three dates of Landsat data are acquired for these models,that is,Landsat on 1993-09-07 and Landsat data on 2000-10-20 and 2001-09-21.
Carries on the radiation to the ETM+/TM data to calibrate,DN(Digital Number) values the header files from the data obtained by are converted into dimensionless values of the radiance L(Radiance).Through the establishment of spatial distortion correction space and the corresponding relationship between the distortions of space will transform all the pixels to the calibration of space,using two sets of coordinates corresponding to the relationship between the remote sensing images of the geometric correction.Remote sensing image geometric correction is conducted with ground control points(GCP),eight points and features obvious points to control error in 0.5 pixels.Through the establishment the relationship of spatial distortion correction and the distortion of space,all the pixels transforms to the calibration of space,using the relationship of two sets of coordinates corresponding to geometric correction for remote sensing images.The purpose of atmospheric correction is to eliminate factors such as air and light reflection on the ground, gain ground reflectance and radiation ratio,temperature and real physical surface model parameters used to eliminate the atmosphere,water,oxygen and carbon dioxide and methane and reflection on the ground of ozone,eliminate air molecules and aerosol scattering effects. In remote sensing image processing,often requires the removal of atmospheric scattering effects of the image brightness,that is,the need for atmospheric correction.Using the atmospheric correction module FLAASH ENVI of ETM+/TM data,eliminates the atmosphere correction of remote sensing image by air,so as to obtain the true features of the radiation information.Corrected through a 1,2,3-band calculated from the water-leaving radiance.Assumptions in the clear waters of the spectrum reflection occurred underwater; input the inversed depth to water reflectance attenuation equation.The request will be incorporated into the water depth of reflectivity attenuation model,in order to inverse the underwater anti-reflectivity.By the relationship of water depth and the water-leaving radiance establish a statistical model so as to come up the depth of the reservoir in other areas.
The depth of Shitoukoumen reservoir from the reservoir edge to middle deepens gradually deepened,which conforms to the reservoir actual water depth distributed condition. The average depth is 9.63014m,which is consistent with the calculated value.
The water bottom reflectance of Shitoukoumen reservoir high area concentrates in north and south some areas but the reservoir's center of the lake area under water index of reflection is low.There is a low reflection strip obvious from Shitoukoumen reservoir west to middle,which is Shitoukoumen reservoir major activity regions.The artificial active influence causes the index of reflection to be lower than other areas obviously.But Shitoukoumen reservoir's actual situation is:the southern part of the reservoir sediment concentration was significantly higher than other region,which distribution is similar to the underwater terrain.And the waters near central reservoir suspended sediment concentration are high,north to gradually decrease.High in the southern area most distribution,which is mainly due to the shallows in natural conditions,the water in the sediment concentration distribution and underwater topography there exists the relationship of dynamic contact.
The average water depth of 1993-09-07 remote sensing image of the error is 14.99%. The average water depth of 2000-10-20 remote sensing image of the error is 17.44%.The average water depth of 2001-09-21 remote sensing image of the error is 3.53%.Results show that water reflectivity attenuation model is accurate in inversion underwater parameter precision.From the table,Shitoukoumen reservoir bottom reflectivity in 2001 obviously lower than 2000 underwater reflectivity,and reservoir bottom reflectivity in 2000 obviously lower than 1993 underwater reflectivity.Shitoukoumen reservoir pollution becomes more and more serious.
In order to verify the sensitivity of radioactive transfer model,the solar azimuth angle and the depth are evaluated for the inverse.The Vertex Angle with the depth of the value trend is same,increased significantly,and the trend is clear,viewing Angle from 30°to 90°,the depth increase about 2.7(m) in the other conditions remain unchanged.With the water bottom reflectance increasing,the zenith angle reduces,but the scope of minor changes in the 0.01 range.With the increase in water depth,bottom reflectance gradually becomes smaller.
研究中采用辐射传输模型,对石头口门水库的水底反射率进行了反演。在水深较大(＞2m)的假设下,选取水库较清澈污染较小的区域,假设水体全部发生镜面反射,由水底反射率衰减模型求出水深。利用已求水深与离水辐射率的进行拟合,得出水深与离水辐射率的关系式。并将离水辐射率带入求出水库的所有区域水深。将所求水深代入水底反射率衰减模型中反演水底反射率。
以2000年ETM+卫星遥感图像为例,基于辐射传输模型对水深和水底反射率进行反演,对水深结果进行评价分析,反演出水底反射率,而且随之时间推移,水底反射率变小,可能污染物浓度逐渐增加。最后为了了解模型中各参数对反演结果的影响,进行敏感性分析,主要是研究太阳天顶角、水深的变化等对结果的影响。。
Water quality pollution is a serious problem,highly needed be solved,in China.Remote sensing monitoring of water-bottom,using the relationship between reflectance spectrum characteristics of water bodies and water bottom parameter,inverses water bottom parameter. There are three ways,which are customarily used to resolve the problem,including empirical,semi-empirical and mechanism modeling methods.At present,most researches are focused on the measurement of various water body spectrum characteristics,and regression analysis using remote sensing data and other methods to establish a visible light or the ratio of band parameters and the bottom of the empirical formula,in order to extract quantitative information of water bottom.The method can be carried out very easily,but it is depended upon the measured data of water surface and synchronous data of remote sensing. However,the accuracy of this method will be significantly influenced by the changing of the water sediment concentration.The principle of mechanism model,based on two relation ship: water component and amount of inherent optical,amount of inherent optical and apparent optical volume,simulates the distribution of light field in water and then inverse the bottom-water parameter.The theoretical basis of mechanism model is the principle of light transmission in water,therefore,mechanism model is an important method of remote sensing monitoring of water quality.The advantages of the mechanism model have a clear physical meaning,the bottom of the reliability of inversion parameters,and subjected to time and geographical constraints,applicability.Therefore,this study based on radioactive transfer model.
According to the radiation from the water-leaving radiance,water depth and bottom reflectance modeling is to achieve the bottom reflectance effective way of quantitative remote sensing At depths greater(＞2m) under the assumption that,in the clear waters of the assumption that spectrum reflection occurs,the bottom reflectance is 1.According to the water depth and rate of water-leaving radiance to establish the relationship between the depth of the statistical model,the request will be incorporated into the water depth of reflection attenuation model,then inverse water bottom reflectance.To Shitoukoumen Reservoir as an example,inverse water bottom reflectance use of water attenuation model.Three dates of Landsat data are acquired for these models,that is,Landsat on 1993-09-07 and Landsat data on 2000-10-20 and 2001-09-21.
Carries on the radiation to the ETM+/TM data to calibrate,DN(Digital Number) values the header files from the data obtained by are converted into dimensionless values of the radiance L(Radiance).Through the establishment of spatial distortion correction space and the corresponding relationship between the distortions of space will transform all the pixels to the calibration of space,using two sets of coordinates corresponding to the relationship between the remote sensing images of the geometric correction.Remote sensing image geometric correction is conducted with ground control points(GCP),eight points and features obvious points to control error in 0.5 pixels.Through the establishment the relationship of spatial distortion correction and the distortion of space,all the pixels transforms to the calibration of space,using the relationship of two sets of coordinates corresponding to geometric correction for remote sensing images.The purpose of atmospheric correction is to eliminate factors such as air and light reflection on the ground, gain ground reflectance and radiation ratio,temperature and real physical surface model parameters used to eliminate the atmosphere,water,oxygen and carbon dioxide and methane and reflection on the ground of ozone,eliminate air molecules and aerosol scattering effects. In remote sensing image processing,often requires the removal of atmospheric scattering effects of the image brightness,that is,the need for atmospheric correction.Using the atmospheric correction module FLAASH ENVI of ETM+/TM data,eliminates the atmosphere correction of remote sensing image by air,so as to obtain the true features of the radiation information.Corrected through a 1,2,3-band calculated from the water-leaving radiance.Assumptions in the clear waters of the spectrum reflection occurred underwater; input the inversed depth to water reflectance attenuation equation.The request will be incorporated into the water depth of reflectivity attenuation model,in order to inverse the underwater anti-reflectivity.By the relationship of water depth and the water-leaving radiance establish a statistical model so as to come up the depth of the reservoir in other areas.
The depth of Shitoukoumen reservoir from the reservoir edge to middle deepens gradually deepened,which conforms to the reservoir actual water depth distributed condition. The average depth is 9.63014m,which is consistent with the calculated value.
The water bottom reflectance of Shitoukoumen reservoir high area concentrates in north and south some areas but the reservoir's center of the lake area under water index of reflection is low.There is a low reflection strip obvious from Shitoukoumen reservoir west to middle,which is Shitoukoumen reservoir major activity regions.The artificial active influence causes the index of reflection to be lower than other areas obviously.But Shitoukoumen reservoir's actual situation is:the southern part of the reservoir sediment concentration was significantly higher than other region,which distribution is similar to the underwater terrain.And the waters near central reservoir suspended sediment concentration are high,north to gradually decrease.High in the southern area most distribution,which is mainly due to the shallows in natural conditions,the water in the sediment concentration distribution and underwater topography there exists the relationship of dynamic contact.
The average water depth of 1993-09-07 remote sensing image of the error is 14.99%. The average water depth of 2000-10-20 remote sensing image of the error is 17.44%.The average water depth of 2001-09-21 remote sensing image of the error is 3.53%.Results show that water reflectivity attenuation model is accurate in inversion underwater parameter precision.From the table,Shitoukoumen reservoir bottom reflectivity in 2001 obviously lower than 2000 underwater reflectivity,and reservoir bottom reflectivity in 2000 obviously lower than 1993 underwater reflectivity.Shitoukoumen reservoir pollution becomes more and more serious.
In order to verify the sensitivity of radioactive transfer model,the solar azimuth angle and the depth are evaluated for the inverse.The Vertex Angle with the depth of the value trend is same,increased significantly,and the trend is clear,viewing Angle from 30°to 90°,the depth increase about 2.7(m) in the other conditions remain unchanged.With the water bottom reflectance increasing,the zenith angle reduces,but the scope of minor changes in the 0.01 range.With the increase in water depth,bottom reflectance gradually becomes smaller.
引文
[1]谭衢林,邵云.遥感技术在环境监测中的应用[J].遥感术与应用,2000,15(4):247-251.
[2]李旭文,季耿善,杨静.苏州运河水质的TM分析[J].环境遥感,1993,8(1):36-44.
[3]匡定波.卫星遥感解译太湖水质研究[J].863-308通讯,1998,2:4-5.
[4]张渊智,聂跃平,蔺启忠等.表面水质遥感监测研究[J].遥感技术与应用,2000,15(4):215-219.
[5]ZHAN G Y,PULL IAIN EN,JOUN I,etal.Application of an empirical neural network to surface water quality estimatin in the Gulf of Finland using combined optical data and microwave data[J].Remote sens Environ,2002,8(2-3):327-336.
[6]DEKKER A G.Detection of optical water quality parameters for eutrophic waters by high resolution remote sensing[D].Amsterdam:Free University,1993.
[7]周艺,周伟奇,王世新等.遥感技术在内陆水体水质监测中的应用[J].水科学进展,2004,5:312-317.
[8]徐希孺等.遥感物理[M].北京大学出版社,2005:272-274.
[9]Sathyendranath,Morel A.Remote Sensing Applications in Manine Science and Technology[M].London Kluwer Academic Publishers,1983:53-54.
[10]Carpenter D J,Carpenter S M,Modeling inland water quality using landsat data[J].Remote sensing of Environment,1983,13:345-352.
[11]王学军,马延.应用遥感技术监测和评价太湖水质状况[J].环境科学,2000,21(11):65-68.
[12]李炎,李京.基于海面-遥感器光谱反射率斜率传递现象的悬浮泥沙遥感算法[J].科学通报,1999,44(17):1892-1898.
[13]Mgengel V,J Spitzer R.Application of remote sensing data to maping of shallow sea-floor near by Netherlands[J].International Journal of Remote Sensing,1991,7:473-479.
[14]V.Lafon,Friodefond J.M.Bathymetric mapping by SPOT images to quantify sand movements in the tidal inlet of Ateachon(France)[J].Earth Surface Sensing,1997:456-466
[15]王旭楠.基于辐射传输模型的水质参数定量遥感反演[D].吉林:吉林大学地探学院,2008.
[16]JenniferL.lrish,L.lrish.W.Jeff.Liffycrop.Scanning laser mapping of the coastal zone:the SHOALS System[J].Remote Sensing,1999,54:123-129.
[17]李铁芳,易建春,厉银喜.浅海水下地形地貌遥感信息提取与应用[J].环境遥感,1991,6(1):22-28.
[18]ALLEE R J,JOHNSON J E.Use of satellite imagery to estimate surface chlorophyll-a and secchidisc depth of Bull reservoir,Arkanass,USA[J].International Journal of Remote Sensing,1999,20(6):1057-1072.
[19]石爱业,徐立中,杨先一等.基于神经网络-证据理论的遥感图像数据融合与湖泊水质状况识别[J].中国图像图形学报,2005,3:372-377.
[20]疏小舟,尹球,匡定波.内陆水体藻类叶绿素浓度与反射光谱特征的关系[J].遥感学报,2000,4(1):41-45.
[21]张东,王文,张鹰.长江口水域水深遥感研究[J].河海大学学报,1998,26(6):86-90.
[22]张东,王文,张鹰.统计相关水深遥感模式的建立[J].河海大学学报,1998,26(6):95-99.
[23]周建军.水深遥感模型及其在闽江口冲淤变化分析中的应用[D].北京:中国地质大学,2004.
[24]Hoogenboom H J,Dekker A G,Althuis IJ A.Simulation of AVIRIS Sensitivity for Detecting Chlorophyll
[25]over Coastal and Inland Waters[J].Remote Sensing of Environment,1998,65:333-340.
[26]邓孺孺,何执兼,陈晓翔等.珠江口水域水污染遥感定量分析[J].中山大学学报,2002,3:99-103.
[27]邓孺孺,何执兼,陈晓翔等.基于二次散射的水污染遥感模型及其在珠江口水域的应用[J].海洋学报,2003,11:69-78.
[28]Gordon H R.Remove of atmospheric effectsfrom satellite imageryof the oceans.Applied Optics[J].1978,17(10):1631-1636.
[29]Stumpf R P,Pennock J R.Calibration of a general optical equation for remote sensing of suspended sediments in a moderately turgid estuary[M].Geophysical Res.1989:14363-14371.
[30]M.Zhang,K Carder,F.E.Muller-Karger etc.Noise reduction and atmospheric correction for coast applications of landsat thematic mapper imagery[J].Remote Sensing.1999,70:167-180.
[31]David R.Lyzenga,D.R.Remote sensing of bottom reflectance and water attenuation parameters in shallow water using aircraft and Landsat data[J].International Journal of Remote Sensing.1981,2(1):71-82.
[32]Bierwirth,P.N.Mineral mapping and vegetation removal via data-calibrated pixel unmixing using multispectral images[J].International Journal of Remote Sensing].1990,11(11):1999-2017.
[33]任敬萍,赵进平.二类水体水色遥感的主要进展与发展前景[J].地球科学进展,2002,17(3):363-371.
[34]徐京萍,张柏,蔺枉等.结合高光谱数据反演吉林石头口门水库悬浮物含量和透明度[J].湖泊科学,2007,19(3):269-274.
[35]肖桂义,陆继龙,蔡波等.长春市石头口门水库水质演变及对策[J].地质与勘探,2003,39(6):61-63.
[36]Koponen S,Pulliainen J.Lake water quality classification with airborne hyperspectral spectrometer and simulated MERIS data[J].Remote Sensing of Environment,2002,79:51-59.
[37]李素菊,王学军.内陆水体水质参数光谱特征与定量遥感[J].地理学与国土研究,2002,5(2):26-30.
[38]John R.Jensen.遥感数字影像处理导论[M].机械工业出版社,2007.
[39]王历,龚家龙.遥感--遥感技术的发展及其应用[M].北京:科学出版社,1983.
[40]叶明,李仁东,许国鹏.多光谱水深遥感方法及研究进展[J].世界科技研究与发展,2007,29(2):76-79.
[41]upp,D.L.B.,Mayo,K.K.,Kuchler,D.A.,Heggan,S.J.,Kendall,B.,Radke,M.,and Ayling,T.Landsat based interpretation of the Cairns section of the Great Barrier Reef Marine Park.CSIRO div.of Water and Land Resources,Natural Resource Series[J].1985,4:39-51.
[42]Kirk J TO.Light and photosynthesis in aquatic ecosystems[M].Cambridge,Cambridge University Press,1994.
[43]李旭文,季耿善,杨静.苏州运河水质的TM分析[J].环境遥感,1993,8(1):36-44.
[44]袁曾任.人工神经网络及其应用[M].北京:清华大学出版社,2000.
[45]王文,李子扬,唐伶俐.中国遥感卫星地面站Landsat-5 TM的辐射定标.中国遥感卫星地面站用户简讯,2004,(4):3.
[46]Gyanesh Chander,Brian Markham.Revised Landsat-5 TM radiometric calibration procedures and postcalibration dynamic ranges[J].IEEE Transactions on Geoscience and remote sensing,2003,41(11):2674-2677.
[47]况昶.水色遥感理论模型的识别与求解途径探讨[D].北京:清华大学,1999.
[48]陶秋香,陶华学,张连蓬.线性混合光谱模型在植被高光谱遥感分类中的应用研究[M].勘察科学技术,2004(1):21-24.
[49]李素菊.利用分析方法建立湖泊光学水质参数反演算法研究[D].北京:北京大学,2003.
[50]高永年,张万昌.遥感影像地形校正物理模型的简化与改进[J].测绘学,2008,37(1):89-94.
[51]张亭禄,贺明霞.基于人工神经网络的一类水域叶绿素a浓度反演方法[J].遥感学报,2002,6(1):40-44.
[52]阿布都瓦斯提.吾拉木.绿洲-荒漠交错带地下水位分布的遥感模型研究[D].新疆:新疆大学资源环境信息信息科学院,2000.
[53]KIRK J TO.Light and Photosynthesis in Aquatic Ecosystems[M].Cambridge:Cambridge University Press,1994.184-185.
[54]DEKKER A G,HOOGENBOOM H J,RIJKEBOER M.Estimation of water quality parameters of inland waters with the use of a toolkit for processing of remote sensing data[C].Proceedings of the 4th International Conference on Remote Sensing for Marine and Coastal Environments.Orlando:1997,269-275.
[55]HARMA P,VEPSALAINEN J,HANNONEN T,et al.Detection of water quality using simulated satellite data and semi-empirical algorithms in Finland[J].The Science of the Total Environment,2001,268:107-121.
[2]李旭文,季耿善,杨静.苏州运河水质的TM分析[J].环境遥感,1993,8(1):36-44.
[3]匡定波.卫星遥感解译太湖水质研究[J].863-308通讯,1998,2:4-5.
[4]张渊智,聂跃平,蔺启忠等.表面水质遥感监测研究[J].遥感技术与应用,2000,15(4):215-219.
[5]ZHAN G Y,PULL IAIN EN,JOUN I,etal.Application of an empirical neural network to surface water quality estimatin in the Gulf of Finland using combined optical data and microwave data[J].Remote sens Environ,2002,8(2-3):327-336.
[6]DEKKER A G.Detection of optical water quality parameters for eutrophic waters by high resolution remote sensing[D].Amsterdam:Free University,1993.
[7]周艺,周伟奇,王世新等.遥感技术在内陆水体水质监测中的应用[J].水科学进展,2004,5:312-317.
[8]徐希孺等.遥感物理[M].北京大学出版社,2005:272-274.
[9]Sathyendranath,Morel A.Remote Sensing Applications in Manine Science and Technology[M].London Kluwer Academic Publishers,1983:53-54.
[10]Carpenter D J,Carpenter S M,Modeling inland water quality using landsat data[J].Remote sensing of Environment,1983,13:345-352.
[11]王学军,马延.应用遥感技术监测和评价太湖水质状况[J].环境科学,2000,21(11):65-68.
[12]李炎,李京.基于海面-遥感器光谱反射率斜率传递现象的悬浮泥沙遥感算法[J].科学通报,1999,44(17):1892-1898.
[13]Mgengel V,J Spitzer R.Application of remote sensing data to maping of shallow sea-floor near by Netherlands[J].International Journal of Remote Sensing,1991,7:473-479.
[14]V.Lafon,Friodefond J.M.Bathymetric mapping by SPOT images to quantify sand movements in the tidal inlet of Ateachon(France)[J].Earth Surface Sensing,1997:456-466
[15]王旭楠.基于辐射传输模型的水质参数定量遥感反演[D].吉林:吉林大学地探学院,2008.
[16]JenniferL.lrish,L.lrish.W.Jeff.Liffycrop.Scanning laser mapping of the coastal zone:the SHOALS System[J].Remote Sensing,1999,54:123-129.
[17]李铁芳,易建春,厉银喜.浅海水下地形地貌遥感信息提取与应用[J].环境遥感,1991,6(1):22-28.
[18]ALLEE R J,JOHNSON J E.Use of satellite imagery to estimate surface chlorophyll-a and secchidisc depth of Bull reservoir,Arkanass,USA[J].International Journal of Remote Sensing,1999,20(6):1057-1072.
[19]石爱业,徐立中,杨先一等.基于神经网络-证据理论的遥感图像数据融合与湖泊水质状况识别[J].中国图像图形学报,2005,3:372-377.
[20]疏小舟,尹球,匡定波.内陆水体藻类叶绿素浓度与反射光谱特征的关系[J].遥感学报,2000,4(1):41-45.
[21]张东,王文,张鹰.长江口水域水深遥感研究[J].河海大学学报,1998,26(6):86-90.
[22]张东,王文,张鹰.统计相关水深遥感模式的建立[J].河海大学学报,1998,26(6):95-99.
[23]周建军.水深遥感模型及其在闽江口冲淤变化分析中的应用[D].北京:中国地质大学,2004.
[24]Hoogenboom H J,Dekker A G,Althuis IJ A.Simulation of AVIRIS Sensitivity for Detecting Chlorophyll
[25]over Coastal and Inland Waters[J].Remote Sensing of Environment,1998,65:333-340.
[26]邓孺孺,何执兼,陈晓翔等.珠江口水域水污染遥感定量分析[J].中山大学学报,2002,3:99-103.
[27]邓孺孺,何执兼,陈晓翔等.基于二次散射的水污染遥感模型及其在珠江口水域的应用[J].海洋学报,2003,11:69-78.
[28]Gordon H R.Remove of atmospheric effectsfrom satellite imageryof the oceans.Applied Optics[J].1978,17(10):1631-1636.
[29]Stumpf R P,Pennock J R.Calibration of a general optical equation for remote sensing of suspended sediments in a moderately turgid estuary[M].Geophysical Res.1989:14363-14371.
[30]M.Zhang,K Carder,F.E.Muller-Karger etc.Noise reduction and atmospheric correction for coast applications of landsat thematic mapper imagery[J].Remote Sensing.1999,70:167-180.
[31]David R.Lyzenga,D.R.Remote sensing of bottom reflectance and water attenuation parameters in shallow water using aircraft and Landsat data[J].International Journal of Remote Sensing.1981,2(1):71-82.
[32]Bierwirth,P.N.Mineral mapping and vegetation removal via data-calibrated pixel unmixing using multispectral images[J].International Journal of Remote Sensing].1990,11(11):1999-2017.
[33]任敬萍,赵进平.二类水体水色遥感的主要进展与发展前景[J].地球科学进展,2002,17(3):363-371.
[34]徐京萍,张柏,蔺枉等.结合高光谱数据反演吉林石头口门水库悬浮物含量和透明度[J].湖泊科学,2007,19(3):269-274.
[35]肖桂义,陆继龙,蔡波等.长春市石头口门水库水质演变及对策[J].地质与勘探,2003,39(6):61-63.
[36]Koponen S,Pulliainen J.Lake water quality classification with airborne hyperspectral spectrometer and simulated MERIS data[J].Remote Sensing of Environment,2002,79:51-59.
[37]李素菊,王学军.内陆水体水质参数光谱特征与定量遥感[J].地理学与国土研究,2002,5(2):26-30.
[38]John R.Jensen.遥感数字影像处理导论[M].机械工业出版社,2007.
[39]王历,龚家龙.遥感--遥感技术的发展及其应用[M].北京:科学出版社,1983.
[40]叶明,李仁东,许国鹏.多光谱水深遥感方法及研究进展[J].世界科技研究与发展,2007,29(2):76-79.
[41]upp,D.L.B.,Mayo,K.K.,Kuchler,D.A.,Heggan,S.J.,Kendall,B.,Radke,M.,and Ayling,T.Landsat based interpretation of the Cairns section of the Great Barrier Reef Marine Park.CSIRO div.of Water and Land Resources,Natural Resource Series[J].1985,4:39-51.
[42]Kirk J TO.Light and photosynthesis in aquatic ecosystems[M].Cambridge,Cambridge University Press,1994.
[43]李旭文,季耿善,杨静.苏州运河水质的TM分析[J].环境遥感,1993,8(1):36-44.
[44]袁曾任.人工神经网络及其应用[M].北京:清华大学出版社,2000.
[45]王文,李子扬,唐伶俐.中国遥感卫星地面站Landsat-5 TM的辐射定标.中国遥感卫星地面站用户简讯,2004,(4):3.
[46]Gyanesh Chander,Brian Markham.Revised Landsat-5 TM radiometric calibration procedures and postcalibration dynamic ranges[J].IEEE Transactions on Geoscience and remote sensing,2003,41(11):2674-2677.
[47]况昶.水色遥感理论模型的识别与求解途径探讨[D].北京:清华大学,1999.
[48]陶秋香,陶华学,张连蓬.线性混合光谱模型在植被高光谱遥感分类中的应用研究[M].勘察科学技术,2004(1):21-24.
[49]李素菊.利用分析方法建立湖泊光学水质参数反演算法研究[D].北京:北京大学,2003.
[50]高永年,张万昌.遥感影像地形校正物理模型的简化与改进[J].测绘学,2008,37(1):89-94.
[51]张亭禄,贺明霞.基于人工神经网络的一类水域叶绿素a浓度反演方法[J].遥感学报,2002,6(1):40-44.
[52]阿布都瓦斯提.吾拉木.绿洲-荒漠交错带地下水位分布的遥感模型研究[D].新疆:新疆大学资源环境信息信息科学院,2000.
[53]KIRK J TO.Light and Photosynthesis in Aquatic Ecosystems[M].Cambridge:Cambridge University Press,1994.184-185.
[54]DEKKER A G,HOOGENBOOM H J,RIJKEBOER M.Estimation of water quality parameters of inland waters with the use of a toolkit for processing of remote sensing data[C].Proceedings of the 4th International Conference on Remote Sensing for Marine and Coastal Environments.Orlando:1997,269-275.
[55]HARMA P,VEPSALAINEN J,HANNONEN T,et al.Detection of water quality using simulated satellite data and semi-empirical algorithms in Finland[J].The Science of the Total Environment,2001,268:107-121.