威海市褚岛北部海域CDOM和COD分布特征遥感分析

根据2017年7月和9月在山东威海褚岛北部海域现场测量的COD_(Mn)(Chemical oxygen demand)值和水体表观光学量,结合COMS(Communication,OceanMeteorological Satellite)上搭载的传感器GOCI(Geostationary Ocean Color Imager)所提供的有色可溶性有机物(Chromophoric dissolved organic matter,CDOM)产品,利用星地同步观测数据对现有的基于遥感反射比反演CDOM的模式进行验证,确定适合该海域的CDOM浓度遥感反演模式;通过对测试海域化学需氧量与遥感反演的水体CDOM浓度相关性分析,建立利用CDOM反演COD_(Mn)的遥感模式,并将该模式应用于测试海域LANDSAT 8/OLI(Operational Land Imager)遥感图像上,获取该海域COD_(Mn)浓度专题图,基于这些专题图分析了测试海域COD_(Mn)时空分布特征。结果表明:(1)基于GOCI产品的CDOM浓度值随时间和站点动态变化大,离岸越近数值越高,同一地点水体前后相差近1 h的数值变化也较大;(2)基于LANDSAT 8/OLI遥感数据反演的COD_(Mn)浓度时间动态变化大,总体来看褚岛附近水体的COD_(Mn)含量相对较低,褚岛以北海域水体COD_(Mn)含量有所增加,褚岛西侧水体的COD_(Mn)含量较东侧水体COD_(Mn)含量来说整体偏高。     

基于高光谱数据的珠江口表层水体悬浮泥沙遥感反演模式

利用2006年12月4日珠江口海域实测的高光谱遥感反射率数据及悬浮泥沙质量浓度数据,进行了该海域表层水体悬浮泥沙遥感反演模式的研究。研究结果表明,悬浮泥沙质量浓度与Rrs(λ1)/Rr s(λ2)-Rrs(λ1)/Rrs(λ3)的相关性较好,其中Rrs(λ)代表遥感反射率,λ代表波长,λ1=762.6 nm,λ2=559.09 nm,λ3=772.78 nm,建立了悬浮泥沙质量浓度定量遥感反演模式,该模式的均方根误差为4.67 mg/L,可以用于珠江口海域的悬浮泥沙质量浓度的遥感监测。     

水体石油类信息遥感提取模式研究

石油类是水体有机污染物中的一种,其对水体吸收系数的影响主要体现在有色可溶性有机物(chromophoric dissolved organic matter,CDOM)的光吸收特性上。在水色遥感领域,CDOM光吸收特性主要用其在参考波段的吸收系数和光谱斜率来表征。利用2008年5月、2009年8月和2010年6月在辽宁省盘锦市辽河油田境内获取的CDOM吸收系数、水体表观光学特性以及石油类污染浓度等试验数据,确定表征石油类污染水体CDOM吸收光谱特性的光谱斜率;根据光谱斜率以及表征CDOM浓度的440 nm参考波段吸收系数,建立遥感反演水体石油类污染浓度的模式,并利用31个野外现场实测值对模式精度进行了验证,结果表明,该模式的相对误差为7%;将该模式应用于国产卫星环境一号遥感数据,获取双台子河及辽东湾近岸水体石油类污染空间分布图。     

珠江口海域主导光学因子的遥感分类及其变化特征

基于2014年5月和8月珠江口海域的现场和实验测量数据,分析了光学因子和生态参数的动态变化特征,并找出该海域的主导光学因子;基于模拟的MODIS数据,构建了水体组分特征波长处吸收系数的遥感反演模型,并结合MODIS卫星遥感影像,依据IOCCG规则对主导光学因子进行遥感分类,揭示了珠江口海域2014年逐月水体主导光学因子的动态变化特征。结果表明,珠江口海域水体以a_d、a_g以及两种共同主导类型为主,并且a_g、a_d两种光学因子主导类型所占比例最大,其次是a_g主导类型,a_d主导类型所占比例最小。同时发现a_d（442）、a_dg（442）均与无机悬浮物浓度有较好的正相关关系,说明CDOM和悬浮泥沙具有相似的来源和动力变化过程。     

基于现场光谱数据的珠江口MERIS悬浮泥沙分段算法

利用珠江口海域4个航次共59个站位的实测遥感反射比和悬浮泥沙数据(悬浮泥沙浓度范围为4-140g·m-3),建立了利用MERIS遥感数据反演珠江口悬浮泥沙浓度的分段算法.算法以Rrs(620)/Rrs(560)=0.9为阈值,当Rrs(620)/Rrs(560)<0.9时,红绿波段比值可以较好地反演悬浮泥沙浓度.随着悬浮泥沙浓度的增加,Rrs(620)/Rrs(560)>0.9,红绿比值趋于饱和,对悬浮泥沙的变化响应不敏感,此时采用包含红波段和近红外波段的波段比值提取水体的悬浮泥沙浓度.利用该分段算法从MERIS遥感图像中提取珠江口水体的悬浮泥沙浓度分布,得到较好的结果.     

珠江口海域基于吸收测量的MODIS半分析算法修正

利用珠江口2002年11月颗粒物质(包括浮游植物与非色素颗粒)及有色溶解有机物质(CDOM)吸收系数的测量,尝试修正MODIS半分析算法经验方程.原算法中建立的浮游植物吸收系数aφ(λ)与aφ(675)之间的双曲线正切函数关系式可能并不符合珠江口这一高污染河口区的光学性质,故给出了aφ(λ)与aφ(443)的简单二次函数关系式.非色素颗粒吸收系数(ad)总体上远大于CDOM吸收系数(ag),所以,没有分离ad和ag而直接导出ag并不切实可行;建议参数S取值0.012 nm-1.当然实用的珠江口MODIS半分析算法还需要进一步的调试与优化.     

夏季珠江口海域海水CO2分压的卫星遥感反演

基于控制因子分析的方法,本研究建立了夏季珠江口海域海水CO₂分压(pCO₂)的遥感反演模型。基于珠江水与黑潮水的两端元混合,建立了水平混合和热力学作用的量化模型,并生成了查找表。同时,建立了基于黄色物质(含碎屑)吸收系数的盐度遥感算法,实现珠江口海域表层盐度的遥感反演。利用走航pCO₂和匹配的遥感叶绿素质量浓度产品,建立了生物作用的量化模型。通过集成水平混合和生物作用,最终实现夏季珠江口海域pCO₂的遥感反演。与走航pCO₂比较表明,仅考虑水平混合和热力学作用的遥感结果会显著高估,考虑生物作用后,遥感结果无论在量值和空间变化趋势上均与实测结果相符。此外,遥感反演结果表明,夏季珠江口近岸水域为CO₂的汇区,而离岸的陆架水域则为CO₂的弱源。     

HY-1C卫星海岸带成像仪叶绿素a浓度反演研究

叶绿素a作为最重要的水质参数之一,是评价水体富营养化和初级生产力状况的主要因素。我国海洋一号C（HY-1C）卫星海岸带成像仪（CZI）具有高时空分辨率的观测优势。本文基于东海和南海现场实测数据建立了HY-1C卫星CZI叶绿素a浓度反演模型并在实测水域进行反演,与MODIS叶绿素a浓度反演产品进行了对比验证,应用CZI叶绿素a浓度模型在珠江口、长江口、渤海湾水域进行了叶绿素a浓度反演示例试验。结果表明,叶绿素a浓度模型估算浓度与实测浓度相关系数为0.774 3,平均相对误差为24.58%,利用实测叶绿素a浓度对模型进行精度验证,相关系数达到0.993 9,平均相对误差为18.49%。模型在实测水域反演得到的叶绿素a浓度分布与MODIS叶绿素a浓度产品分布大体一致。在珠江口水域反演得到叶绿素a浓度空间分布为由西北向东南逐级递减,峰值出现在珠江口西沿岸。在长江口、渤海湾反演叶绿素a浓度空间分布均符合地理实情。研究表明HY-1C卫星CZI数据可应用于中国近海水色定量化研究。     

基于半分析算法的赤潮水体固有光学性质反演

在珠江口海域海洋光学浮标实验期间获取了赤潮生消过程的水体光学数据和相应的生化数据.利用该实验数据开展了准分析算法(quasi-analytical algorithm,QAA)反演赤潮水体固有光学参数的精度检验和模型修正工作.(1)利用遥感反射率和QAA反演主要水色波段(412,443,490,510,560,620和...     

基于GOCI影像的长江口及其邻近海域CDOM遥感反演及其日内变化研究

采用静止轨道海洋水色卫星（GOCI）数据对长江口及其邻近海域有色溶解有机物（CDOM）反演。以QAA-CDOM算法为基础,根据实测数据,利用BP神经网络模型来拟合QAA-CDOM算法中需要针对长江口水体进行优化的悬浮颗粒后向散射系数b_bp与吸收系数a_p的关系,从而准确估算CDOM的浓度。结果表明,反演结果准确度较高,平均相对误差为0.35。基于GOCI日内连续成像的优势,选取2014年3月15日8景GOCI影像,利用优化后的QAA-CDOM-BP算法,对长江口及其邻近海域CDOM的日内变化进行反演和分析,得到的变化规律如下:长江口及其邻近海域的CDOM日变化主要受潮流、长江径流等共同影响。长江口内CDOM浓度在涨潮期高于退潮期,由于受长江冲淡水的作用,CDOM从口外往外海区呈现逐渐递减趋势。     

大洋河河口海域有色溶解性有机物的光学特性及遥感反演模型

有色溶解性有机物(CDOM)在河口混合过程中近似呈保守行为,可作为水团运动的示踪剂.基于2009年5月6日大洋河河口海域水体调查的实测数据,对该区域CDOM光学特性及遥感反演模型进行了研究,结果表明:研究区域CDOM主要来自河流输入,成分相对较稳定,属于典型的近岸二类水体区域;波长275～295 nm的光谱斜率和波长4...     

珠江口溶解有机物的遥感动态监测

河口有色溶解有机物（colored dissolved organic matter,CDOM）的分布是各种物理-生物地球化学过程共同作用的结果。为实现河口高动态变化CDOM的监测,遥感是一种重要的手段。由珠江口四个不同季节的航次获得的实测数据,本文构建了一个遥感算法以反演CDOM在400 nm的吸收系数（a_CDOM （400））。该算法使用以波段反射率比值R_rs （667）/R_rs （443）和R_rs （748）/R_rs （412）为自变量。将构建的算法应用于2002-2014年的MODIS/Aqua数据,本文计算了珠江口不同季节的a_CDOM （400）气候态分布。CDOM的分布主要受珠江径流量和区域水下地形特征的影响。沿着垂直于水深梯度的断面,气候态a_CDOM （400）呈指数减少（y=ae^bx,b<0）,但不同季节差异很大。珠江口CDOM主要是河流淡水输运而来。其中,富里酸比例随盐度的增加而降低。基于构建的算法、CDOM保守混合方程和径流量,本文由MODIS/Aqua数据进一步估算了2002-2014年夏季和冬季珠江DOC的有效入海浓度和有效入海通量。珠江的有效入海浓度和有效入海通量都与流量存在正相关关系,且在夏季的相关性更明显,R2分别为0.698和0.9657。     

An optical model for the remote sensing of coloured dissolved organic matter in coastal/ocean waters

An optical model is developed for the remote sensing of coloured dissolved organic matter (CDOM) in a wide range of waters within coastal and open ocean environments. The absorption of CDOM (denoted as a_g) is generally considered as an exponential form model, which has two important parameters – the slope S and absorption of CDOM at a reference wavelength a_g(λ₀). The empirical relationships for deriving these two parameters are established using in-situ bio-optical datasets. These relationships use the spectral remote sensing reflectance (R_rs) ratio at two wavelengths R_rs(670)/R_rs(490), which avoids the known atmospheric correction problems and is sensitive to CDOM absorption and chlorophyll in coastal/ocean waters. This ratio has tight relationships with a_g(412) and a_g(443) yielding correlation coefficients between 0.77 and 0.78. The new model, with the above parameterization applied to independent datasets (NOMAD SeaWiFS match-ups and Carder datasets), shows good retrievals of the a_g(λ) with regression slopes close to unity, little bias and low mean relative and root mean square errors. These statistical estimates improve significantly over other inversion models (e.g., Linear Matrix-LM and Garver-Siegel-Maritorena-GSM semi-analytical models) when applied to the same datasets. These results demonstrate a good performance of the proposed model in both coastal and open ocean waters, which has the potential to improve our knowledge of the biogeochemical cycles and processes in these domains.     

应用CDOM光学特性估算水体COD——以辽宁省盘锦市双台子河和辽东湾为例

有色可溶性有机物(chromophoric dissolved organic matter,CDOM)是水体中一类重要的光吸收物质,在水色遥感中,其光学特性主要以440 nm处的吸收系数和光谱斜率来表征,利用这些光学特性进行水环境要素遥感反演具有较广泛的应用前景.化学需氧量(chemical oxygen deman...     

The conservative and non-conservative behavior of chromophoric dissolved organic matter in Chinese estuarine waters

The spectral absorption properties of chromophoric dissolved organic matter (CDOM) and their distributions in two Chinese estuaries, the Yangtze River Estuary and the Jiulong River Estuary, were studied during August 2003 (wet season) and during different seasons between 2003−2005, respectively. The CDOM concentrations (a₃₅₅) of fresh end members in the Jiulong River Estuary varied seasonally, while its quality remained relatively stable. However, the a₃₅₅ of the marine end members exhibited less variability. Application of a conservative mixing model indicated that CDOM behaved conservatively in the Yangtze River Estuary. No photobleaching removal was observed at high salinity region of this estuary. Although CDOM showed conservative behavior for many cruises in the Jiulong River Estuary, there was evidence for removal in the low salinity regions during some cruises. Laboratory mixing experiments and a salt addition experiment suggested that particle sorption of CDOM maybe the possible reason for the removal. These results showed that absorption properties of CDOM can be used as a tool to observe the quantitative and qualitative dynamics of DOM during estuarine mixing.     

Seasonal dynamics of light absorption by chromophoric dissolved organic matter (CDOM) in the NW Mediterranean Sea (BOUSSOLE site)

We analyze a two-year time-series of chromophoric dissolved organic matter (CDOM) light absorption measurements in the upper 400 m of the water column at the BOUSSOLE site in the NW Mediterranean Sea. The seasonal dynamics of the CDOM light absorption coefficients at 440 nm (a_cdom(440)) is essentially characterized by (i) subsurface maxima forming in spring and progressively reinforcing throughout summer, (ii) impoverishment in the surface layer throughout summer and (iii) vertical homogeneity in winter. Seasonal variations of the spectral dependence of CDOM absorption, as described by the exponential slope value (S_cdom), are characterized by highest values in summer and autumn at the surface and low values at the depths of a_cdom(440) subsurface maxima or just below them. Variations of a_cdom(440) are likely controlled by microbial digestion of phytoplankton cells, which leads to CDOM production, and by photochemical destruction (photobleaching), which leads to CDOM degradation. Photobleaching is also the main driver of S_cdom variations. Consistently with previous observations, a_cdom(440) for a given chlorophyll a concentration is higher than expected from Case I waters bio-optical models. The total non-water light absorption budget shows that surface waters at the BOUSSOLE site are largely dominated by CDOM during all seasons but the algal bloom in March and April. These results improve the knowledge of CDOM absorption dynamics in the Mediterranean Sea, which is scarcely documented. In addition, they open the way to improved algorithms for the retrieval of CDOM absorption from field or satellite radiometric measurements.     

Characterization of dissolved organic matter fluorescence in the South Atlantic Bight with use of PARAFAC model: Relationships between fluorescence and its components,absorption coefficients and organic carbon concentrations

In this study, the CDOM absorption coefficient at 350 nm [a_CDOM(350)] and CDOM excitation emission matrix (EEM) fluorescence were used to estimate annual fluxes of dissolved organic carbon (DOC) from the Cape Fear River to Long Bay in the South Atlantic Bight. Water samples were collected during a 3.5 year period, from October 2001 through March 2005, in the vicinity of the Cape Fear River (CFR) outlet and adjacent Onslow Bay (OB). Parallel factor analysis (PARAFAC) of CDOM EEM spectra identified six components: three terrestrial humic-like, one marine humic-like and two protein-like. Empirical relationships were derived from the PARAFAC model between DOC concentration and a_CDOM(350), total fluorescence intensity and the intensities of respective EEM components. DOC concentration and CDOM optical parameters were very well correlated and R² values ranged from 0.77 to 0.90. Regression analyses revealed that the non-absorbing DOC fraction, in DOC concentration estimated from CDOM optical parameters, varied with the qualitative composition of the CDOM. DOC concentration and intensity of the humic-like CDOM components characterized by excitation maxima at longer wavelengths have significantly higher estimated non-absorbing DOC compared to the analogous relationships between DOC and intensity of the humic-like CDOM components characterized by excitation maxima at shorter wavelengths. The relationships between DOC concentration and intensity of one of the protein-like components resulted in significantly reduced non-absorbing DOC fraction in DOC concentration estimation. Results of regression analyses between fluorescence intensities of specific EEM components and CDOM-specific absorption coefficients suggest that the relative proportion of humic-like CDOM components (characterized by excitation maximum at longer wavelengths) and the main protein-like component have the most impact on the values of a?_CDOM(350). Based on the relationships between a_CDOM(350), Cape Fear River flow, and DOC concentrations, DOC fluxes were estimated for 2002, 2003 and 2004. DOC fluxes varied from 1.5 to 6.2 × 10¹⁰ g C yr^? 1, depending on river flow.     

海水石油类物质吸收系数遥感化提取算法研究

Establishing the remote sensing algorithm of retrieving the absorption coefficient of seawater petroleum substances is an efficient way to improve the accuracy of retrieving a seawater petroleum concentration using a remote sensing technology. A remote sensing reflectance is a basic physical parameter in water color remote sensing. Apply it to directly retrieve the absorption coefficient of seawater petroleum substances is of potential advantage. The absorption coefficient of waters containing petroleum [ACWCP, a_o(λ)], consists of the absorption coefficient of pure water [ACPW, a_w(λ)], plankton [ACP, a_(ph)(λ)], colored scraps [ACCS, a_(d,g)(λ)], and petroleum substance [ACPS, a_(oil)(λ)]. Among those, ACCS consists of the absorption coefficient of nonalgal particle [ACNP, a_d(λ)] and colored dissolved organic matter [ACCDOM, a_g(λ)]. For waters containing petroleum, the retrieved ACCS using the existing method is a combination absorption coefficient of ACNP,ACCDOM and ACPA [CAC, a_(d,g,oil)(λ)]. Therefore, the principle question is how to extract ACPS from CAC.Through the analysis of the three proportion tests conducted between the year of 2013 and 2015 and the corresponding remote sensing data, an algorithm of retrieving the absorption coefficient of petroleum substances is proposed based on remote sensing reflectance. First of all, ACPS and CAC are retrieved from the reflectance using the quasi-analytical algorithm(QAA), with some parameter modified. Secondly, given the fact that the backscatter coefficient [BC, b_(bp)(555)] of total particles at 555 nm can be obtained completely from the reflectance, the relation between BC and ACNP in petroleum contaminated water can be established. As a result, ACNP can be calculated. Then, combining the remote sensing retrieving algorithm of a_g(440), the method of achieving the spectral slope of the absorption coefficient can be established, from which ACCDOM,can be calculated. Finally, ACPS can be computed as the residual. The accuracy of ACPS based on this algorithm is 86% compared with the in situ measurements.     

Chromophoric dissolved organic matter and dissolved organic carbon in Chesapeake Bay

Chromophoric dissolved organic matter (CDOM) is the light absorbing fraction of dissolved organic carbon (DOC). The optical properties of CDOM potentially permit remote sensing of DOC and CDOM, and correction for CDOM absorption is essential for remote sensing of chlorophyll a (chl a) in coastal and estuarine waters. To provide data for this purpose, we report the distributions of CDOM, DOC, and chl a from seven cruises in Chesapeake Bay in 1994–1997. We observed non-conservative distributions of chl a and DOC in half of the cruises, indicating net accumulations within the estuary; however, there were no net accumulations or losses of CDOM, measured as absorption at 355 nm or as fluorescence. Freshwater end member CDOM absorption varied from 2.2 to 4.1 m⁻¹. Coastal end member CDOM absorption was considerably lower, ranging over 0.4–1.1 m⁻¹. The fluorescence/absorption ratio was similar to those reported elsewhere for estuarine and coastal waters; however, in the lower salinity/high CDOM region of the Bay, the relationship was not constant, suggestive of the mixing of two or more CDOM sources. Chl a was not correlated with the absorption for most of the cruises nor for the data set as a whole; however, CDOM and DOC were significantly correlated, with two groups evident in the data. The first group had high CDOM concentrations per unit DOC and corresponded to the conservative DOC values observed in the transects. The second group had lower CDOM concentrations per unit DOC and corresponded to the non-conservative DOC values associated with net DOC accumulation near the chl a maximum on the salinity gradient. This indicates the production of non-chromophoric DOC in the region of the chl a maximum of Chesapeake Bay. In terms of remote sensing, these data show that (1) the retrieval of the absorption coefficient of CDOM from fluorescence measurements in the Bay must consider the variability of the fluorescence/absorption relationship, and (2) estimates of DOC acquired from CDOM absorption will underestimate DOC in regions with recent, net accumulations of DOC.