A simple semi-analytical model to estimate total suspended sediment matter (3S) was established for estimating TSM concentrations in Changjiang River Estuary. The results indicate that 3S model with near-infrared wavelengths provide good estimates of TSM concentrations in the study region. Furthermore, the applicability of 3S model was evaluated using an independent data set taken from Oujiang river estuary during September 2012. The results indicate that providing an available atmospheric correction scheme for satellite imagery, the 3S model could be used for quantitative monitoring of TSM concentration in coastal waters, even though local bio-optical information is still needed to reinitialize the model.
©2013 Optical Society of America
Total suspended sediment matter (TSM) in coastal waters plays an important role in biogeochemical cycles in estuarine and coastal environments, because the fine-grained particles are an important carrier of various chemical compounds . TSM concentrations in estuaries are strongly influenced by a combination of hydrodynamic, physico-chemical, and biological processes. Understanding the spatial and temporal dynamics of TSM in estuarine systems can thus allow for the estimation of the transport of terrestrial and anthropogenic materials to pelagic oceans. Additionally, the interaction between TSM and seawater constituents may strongly modify the nutrient concentration in estuarine systems . However, estuarial ecosystems typically exhibit TSM concentrations with high temporal and spatial variability, which are often too difficult to characterize . Fortunately, ´ satellite imagery can be used to rapidly assessed the optical biophysical variables in coastal environments dominated by phytoplankton at temporal and spatial scales difficult to attain with direct field measurements .
Many methods of TSM estimation have been developed using empirical, semi-analytical, or physical approaches. These models can be used to process satellite data efficiently , but some problems are encountered when they are applied to the Chanjiang River Estuary. The empirical methods estimate TSM based on the relationship between optical properties and TSM concentration [6–8]. This kind of model is simple and easy to implement; however, it lacks a physical foundation and the relationships are more geographically specific and cannot be applied to other areas . The physical model uses the radiative transfer theory to simulate the spectral at the top-of-atmosphere (TOA) with different TSM concentrations and atmosphere conditions [10–13]. Using the radiance at the TOA measured by satellite detector, TSM concentrations can be accurately determined by approaches such as neural network  optimization, and principal component analysis . However, this model depends on accurate profile information of the inherent optical properties of water and atmosphere, but this kind of information is not always available for the general application of remote sensing. The semi-analytical approach is based on the relationship between the inherent optical properties of sea water and TSM concentration, combined with some empirical relationships [5, 15–19]. The model combines physicals with statistical methods and is a promising technique for TSM concentration retrieval.
Optical properties in Changjiang River Estuary are quite complicated due to sediment resuspension in the seabed  and anthropogenic impacts such as agricultural and industrial sewage. According to the classification method described by Morel and Prieur , the water quality in Changjiang River Estuary falls into the “Case II” category. Considering the sediment-dominated coastal waters in Changjiang River Estuary, refined models are needed to accurately estimate the TSM concentration from satellite remote sensing data . In this study, a new simple semi-analytical TSM model (3S model) was developed for estimating TSM concentrations in Changjiang River Estuary, China. The main objectives of this study were to: (1) develop and construct the 3S model, (2) find out the optimal positions and width of each band of 3S model, and (3) evaluate the performance of 3S model with MODIS, SeaWiFS and MERIS spectral bands.
2. Materials and Methods
2.1 Study area
The Changjiang River Estuary, located between longitudes 121°18′E and 122°45′E, and latitudes 30°47′N and 31°52′N, is a partially mixed estuary. The Changjiang River is the fourth largest in the world. Annual mean water discharge and suspended sediment load is ~4.86 × 108 tons of TSM . 40% of the suspended sediment is deposited within the estuary and the grain size of these fluvial sediment is fine with the medium grain size of 2.7 μm . Such large quantities of river water, terrestrial sediments, and associated chemical species like nutrients and pollutants delivered into the Changjiang River Estuary have substantial impacts on the biogeochemistry cycle of this region. Recently, rapid developments in industry and urbanization in China particularly around east coastal areas covering the Changjiang Valley have caused relatively pristine environments to become more polluted . Much more treated or untreated waste sewage and associated pollutants are discharged into Changjiang River Estuary. Shen et al.  studied circulation of the Changjiang and its effect on the transport of suspended sediment in the Changjiang River estuary and found that the tidal hydrodynamics determine the transport of fine sediments in the Changjiang River Estuary. Both estuarine gravitational circulation and tidal asymmetry are of importance with the complex flow patterns largely determined by the interaction of Changjiang River runoff, tidal currents, and varying topography, along with density differences, winds and the Coriolis effect . The complicated hydrodynamics lead to very high TSM concentrations and high turbidity in the Changjiang River Estuary .
The Oujiang River Estuary (Fig. 1) is located between longitudes 120°48′E and 121°32′E, and latitudes 27°36′N and 28°12′N, near the center of Wenzhou City, Zhejiang Province, a city which is well known in China for its rapid economic development over the past two decades. The optical properties of coastal waters such as the Oujiang River Estuary are vital to local human activities and needs, and play a critical role in the regional ecosystem. Due to the rapid economic development and population growth of this area, enormous quantities of nutrients and other pollutants are transported from land to the Oujiang River Estuary. These pollutants may be among the main causes of the increasing number and scale of harmful algal bloom events in the coastal waters . Thus, there is an urgent need to effectively monitor and manage the aquatic environment in the Oujiang River Estuary, and to better understand the optical, biological and ecological processes and phenomena which occur .
2.2 Statistical criteria
To evaluate model performance, the RMSE (Root-Mean-Square) and MRE (Mean-Relative-Error) statistics were used in this study. These statistics were described by 
Where xmod,i was the modeled value of the ith element, xm was the average values of in situ measurements, xobs,i was the observed (or in situ measured) value of the ith element, and m was the number of elements.
2.3 Data used
To evaluate the accuracy of the model for estimating TSM concentrations, two independent data sets composed of remote sensing reflectance (Fig. 2) and TSM concentration of the water column were used, firstly for model calibration and the secondly for model validation. The calibration data set containing 20 samples were collected in Changjiang River Estuary, China, on 14 October, 2009. The validation data sets containing 16 samples were collected in Changjiang River Estuary, China, on 15 October, 2009. Furthermore, to validate the applicability of the 3S model to other coastal waters, another independent data set including simultaneous measurements of above-water remote sensing reflectance and TSM were taken in Oujiang River Estuary, China during September 2012, containing 66 stations.
2.4 Field Measurements
Field measurements of the two data sets were carried out from 10:00 to 14:00 local time. At each station, remote sensing reflectance measurements were taken from a boat. The reflectance was measured with a spectroradiometer with 25°fiber-optic, covering the spectral range of 350nm-2500nm (Spectral Devices, Boulder, CO, ASD). Although data were collected in the range of 350-2500nm, only data similar to ocean color sensors in the 400-900nm range at 1.4 nm resolution were used [29–33]. Following the ocean optics protocols for satellite ocean color sensor validation , at each station three measurements were repeated in a short time in order to estimate the uncertainty (MRE values of three repeated measurements) associated with each measurement. The average measurements (mean values of three repeated measurements) with <5% RMS at each station were selected for model calibration and validation.
During the measurements, the tip of the optical fiber was kept ~1m above the water surface by means of a 3m long, hand-held black pole. Radiance of both the water surface (Lsw(λ)) and a standard gray board (Lp(λ)) was measured. Ten curves were acquired for each target. In order to effectively avoid interference by the ship with the water surface and the influence of direct solar radiation, the instrument was positioned at an angle β of 90-135° with the plane of the incident radiation away from the sun . The view of the water surface, α, was controlled between 30 and 45° with the aplomb direction. In this way most of the direct sunlight was eliminated while the impact of the ship’s shadow was minimized. Immediately after measuring the water radiance, the spectroradiometer was rotated upwards by 90-120° to measure skylight. The view azimuth angle in this measurement was kept the same as that when measuring the water radiance .
Remote sensing reflectance, Rrs(λ), was calculated as follows:
Where Lw(λ) is the water-leaving radiance, Ed,0(λ) is the total incident radiant flux of the water surface. Lw(λ) and Ed,0(λ) in Eq. (3) are further calculated as follows:
Where Lsw(λ) denotes the total radiance received from the water surface; Lsky(λ) represents the diffuse sky radiation; r refers to the reflectance of the skylight at the air-water interface, whose value depends upon the solar azimuth, measurement geometry, wind speed, and surface roughness; Lp(λ) is the radiance of the gray plate; and ρp(λ) is the reflectance of the gray plate. In this study, r is calculated with assumption of the black water body at wavelength from 1000 to 1020nm  and wavelength-independent .
2.5 Laboratory measurements
Water samples were collected immediately after the radiance measurements. At each station a standard set of water quality parameters were measured. These included TSM concentration, water depth, turbidity, salinity and temperature. The surface water samples were collected at a depth of 0.5m below the water-air interface. After sampling, water samples in bottles were stored at a low temperature and sent for laboratory analysis. The laboratory analyses were carried out within 24 h following sample collection. The TSM concentration was measured by a weighing method. The water sample was filtered with 0.45 μm filter (Whatman GF/F filters) and vacuum filtration system. The filter pad was flushed with 0.00005 m3 distilled water 3 times in order to flush away the salt. Filters were dried overnight at 70 °C for determination of TSM. The dry-weight of the filter-pad was weighed by an electronic analytic scale. The blank filter and sampled filter-pad were weighed until the difference two successive TSM concentrations calculated from the scale reading was within 0.01 mg/l.
2.6 Construction of 3S model
Remote sensing reflectance in the visible and NIR spectrum provides qualitative and quantitative information about optically significant materials present in natural water .
Where λ is the wavelength, γ is unchanging with respect to wavelength and viewing geometric conditions [20, 36, 37], bb(λ) is the total backscattering coefficient of water (bbw(λ)), TSM (bbs(λ)) and phytoplankton pigments, and a(λ) is the total absorption coefficient of colored dissolved organic matter ag(λ), phytoplankton pigments ap (λ), inorganic particles ad(λ), and pure water aw (λ).
To retrieve TSM concentrations from spectral reflectance, one needs to isolate the TSM backscattering coefficient, bbs(λ), which is a part of total backscattering coefficient. According to the study by Bukata et al. , at wavelengths greater than 690nm the volume reflectance is independent of phytoplankton pigments concentrations. At these wavelengths, the backscattering coefficient of phytoplankton pigment concentrations becomes insignificant . Thus, the total backscattering coefficients can be approximated as follows [4, 28]:
According to Smith and Baker , it is readily observed that for λ>~580 nm, scattering by water molecules becomes insignificant. Thus, the light scattering at wavelengths greater than ~580nm is due to TSM scattering. Consequently, Eq. (6) can be approximated as:
In order to accurately estimate TSM concentration using satellite remote sensing, Rrs(λ) should be maximally sensitive to backscattering by TSM; this means that λ should be restricted within NIR regions, because the backscattering properties of phytoplankton pigments and water molecules are insignificant or slightly significant within these regions. In addition to backscattering by TSM, R−1rs(λ) is also affected by absorption by CDOM, phytoplankton pigments, water, and TSM. The effect of ag(λ) + ad(λ) + ap(λ) can be minimized using a second spectral band that meets the following requirement of [ag(λ1) + ad(λ1) + ap(λ1)]/bbs(λ1)≈[ag(λ2) + ad(λ2) + ap(λ2)]/bbs(λ2). Thus, this band has to be quite close to the first band and has to be restricted within the NIR wavelengths [40–42]. Then, the 3S model can be expressed as following formula:
Where [TSM] refers to the TSM concentrations. Therefore, the conceptual 3S model developed by this study can be denoted as follows:
3.1 Characterizing TSM concentration
The data sets used to calibrate and validate the performance of the TSM models contain 35 water samples (Table 1). The calibration data set, including 20 water samples, were collected from Changjiang River Estuary on October 14, 2009. The TSM concentrations in calibration data set range from 0.07 to 0.58 km/m3, and the corresponding average value was 0.22 kg/m3. The TSM concentrations in the validation data set ranged from 0.14 to 0.29 kg/m3, and the corresponding average value was 0.19 kg/m3 (Table 1). The TSM concentration for Oujiang River Estuary ranges from 0.01 to 0.32 kg/m3, the average being 0.04 kg/m3 (Table 1). By comparison, the TSM concentration in Oujiang River Estuary is much lower than that in Changjiang River Estuary, so the waters in Changjiang River Estuary are much more turbid than Oujiang River Estuary.
|c. Descriptive statistics of the TSM concentration for second data set taken in Oujiang River Estuary|
3.2 Spectral characteristics
In general, the remote sensing reflectance is highly variable over the visible and near-infrared spectral regions in turbid “Case II” waters. In Changjiang River Estuary, spectra is very low in the blue range (400-500 nm), which is lower than 0.045 sr−1. Remote sensing reflectance in the green range (500-600 nm) is much higher than in the blue range. In the red region (600-700 nm) reflectance had several spectral features. Finally, reflectance in the near-infrared range (700-840 nm) varied widely at the low reflectance with a distinct peak located between 690nm and 710 nm. This peak is the result of both high backscattering and a minimum in absorption by all optically active constituents including pure water . As shown in Fig. 2, the dramatic impact of TSM concentrations on volume reflectance is clear. Even a small concentration of TSM concentration can substantially increase the volume reflectance in a manner that becomes more pronounced as the wavelength becomes longer. At very large concentrations of TSM (>0.15 kg/m3), the reflectance spectrum displays an asymptotic approach to a broad peak ~700nm. Remote sensing reflectance recorded in Oujiang River Estuary is quite similar in shape to that in Changjiang River Estuary as well as other turbid waters [3, 27, 40, 41], but the magnitude is much lower than that in Changjiang River Estuary. The reason for this result may be the much lower TSM concentration in Oujiang River Estuary than Changjiang River Estuary (Table 1).
3.3 Optimal spectral position and bandwidth of 3S model
In order to determine the optimal positions of λ1 and λ2 and the width of each band in the conceptual 3S model as shown in Eq. (12), we adjusted the model according to the optimal optical properties of the water bodies studied. The calibration data set contained 20 samples taken over water bodies with various properties (Table 1a). The optimal spectral positions of 3S model were determined by band tuning from 400nm to 900nm, although the 3S model requires that the wavelengths should be greater than 690nm, where the backscattering by water molecules becomes insignificant . The non-linear iterative method proposed by Chen and Quan  was used to find out the most ideal functions of the 3S model. We selected the final values of the bandwidths to be consistent with the bandwidth of visible bandpass filters . The results (Fig. 3) indicate that if 35% MRE is considered as an acceptable error upper-bound of TSM estimation, band 1 must be between 690 and 900 nm, and band 2 must be defined as 720 through 780 nm or 840 through 900 nm. Thus, the 3S model has the following form:
3.3 Optimal 3S model with MODIS, MERIS, and SeaWiFS spectral bands
3.3.1 Model calibration
To match the bandwidth of MODIS (Moderate Resolution Imaging Spectroradiometer), SeaWiFS (Sea-viewing Wide Field-of-view Sensor) and MERIS (Medium Resolution Imaging Spectrometer) sensors, the spectra recorded by ASD were aggregated using the spectral response functions of these satellite sensors [7, 49, 50]. The optimal 3S models with MODIS, SeaWiFS and MERIS spectral bands were determined using band turning method. Results show that the optimal position for λ1 and λ2 are 869 and 748 nm for MODIS, 865 and 765 nm for SeaWiFS, and 865 and 761 nm for SeaWiFS, respectively. The results also reveal that all models give a good estimate of TSM concentrations for the turbid waters of Changjiang River Estuary, China; however, the 3S model for the MERIS band (R2 = 0.8917) perform a little better than the MODIS (R2 = 0.8846) and SeaWiFS bands (R2 = 0.8760) (Fig. 4).
3.3.2 Model Evaluation
The stability and accuracy of 3S model with MODIS, SeaWiFS and MERIS spectral bands were evaluated using the first validation data set taken in Changjiang River Estuary, China, on October 15, 2009 (Fig. 2(b)). Figure 5shows the performance of these three models in determining TSM from Changjiang River Estuary. We found that the RMSE values of 3S models with MODIS, SeaWiFS and MERIS spectral bands are 0.0550, 0.0566 and 0.0521 kg/m3, respectively, and the corresponding MRE values are 28.99%, 29.83% and 27.47%, respectively. By comparison, the 3S model with MERIS spectral bands is a little better than both 3S models with MODIS and SeaWiFS spectral bands. Use of 3S model with MERIS spectral bands in estimating TSM concentration in Changjiang River Estuary decreases the uncertainty of estimation by 1.52% and 2.36%, respectively from 3S model with MODIS and SeaWiFS spectral bands. Moreover, Fig. 5 also indicates that the bias between 3S model-derived and field-measured TSM concentration increases with increasing TSM concentrations, but there is no statistically significant relationship between them. Thus, the 3S model can be used to estimate TSM concentration in turbid coastal waters. These findings indicate that, provided that an atmospheric correction scheme for the NIR bands is available, the extensive database of MERIS, MODIS, and SeaWiFS imagery can be used for quantitative monitoring of TSM concentration in Changjiang River Estuary using 3S model.
3.3 Model comparison
The performances of several other models (Table 2) were also evaluated for model comparison, including Doxaran model , Zhang model , Fettweis model , and Miller model . The MRE values (Fig. 6) between estimation from each of the four models and the field-measured TSM concentration were 58.45%, 165.4%, 83.73%, and 55.38%, respectively. The performance of all of the four models is poor and cannot meet the requirements for the TSM concentration retrieval in Bohai Sea. By comparison, 3S model produces a superior performance to all of these four models. Using of the 3S model in estimating TSM concentration in Changjiang River Estuary decreases the MRE values of estimation by >18%.
4.1 Future model evaluation and comparison
The 3S models with MODIS, SeaWiFS, and MERIS spectral bands were calibrated using a data set collected in Oujiang River Estuary, China, during September 2012 and the specific form of this model as expressed by 3S model in Eq. (13) was applied for predicting TSM concentrations for Oujiang River Estuary. We found that these models do not require further optimization of spectral band positions to accurately estimate TSM in water bodies with widely varying bio-optical characteristics, even though the statistical parameters of 3S model require further optimization. They account for more than 90% of the variation in TSM concentration, and hence can be used to accurately estimate TSM concentration in turbid estuarial waters. These findings demonstrate that the optimal position of the 3S model always lies within a wavelength range as shown in Eq. (13), but their specific statistical parameters vary with water constituents and their optical properties, especially the size, shape, and refractive index .
By comparison, the 3S model works better in retrieval of TSM concentration in the Oujiang River Estuary than in Changjiang River Estuary. The TSM concentration in Changjiang River Estuary (Table 1) ranges from 0.07 to 0.71 kg/m3, the average being 0.22 kg/m3, and the TSM concentration in Oujiang River Estuary varies from 0.01 to 0.32 kg/m3, and the corresponding averaged value is 0.04 kg/m3, so the waters in Changjinag River Estuary are more turbid than that in Oujiang River Estuary. This is to say, the 3S model is capable of the estimation in highly turbid waters, but works better in the relatively clear Oujinag River Estuary waters than highly turbid Changjiang River Estuary waters. Therefore, the 3S model can be used to derive TSM concentration from turbid estuarial waters, but the performance may be dependent on optical properties of aquatic environments.
In order to qualitatively understand the impact of optical properties of TSM on 3S model in estimating TSM concentration for turbid estuarial waters, an innovative definition, specific remote sensing reflectance (R*rs(λ)), was developed (Appendix.I). Figure 8 shows the average specific remote sensing reflectance taken in Oujiang River estuary and Changjiang River Estuary, respectively. We found that R*rs(λ) in Changjiang River Estuary is ~2 times larger than that in Oujiang River Estuary. It seems that TSM in Changjiang River Estuary are more effective at scattering photons penetrating into water than that in Oujiang River Estuary. Bowers and Binding  reported that the smaller sized sediments generally lead to a higher spectral reflectance. This is to say, the particle size in Changjiang River Estuary may be smaller than that in Oujinag River Estuary. These findings indicate that the 3S model may produce a superior performance in waters with larger particle size than that with smaller particle size. It is noteworthy that the slope of 3S model in Changjiang River Estuary is much lower than Oujiang River Estuary (Fig. 5 and Fig. 8). This may also be attributed to differences in particle size between Changjiang River Estuary and Oujiang River Estuary.
4.2 Some limitations
There are a few caveats that need to be considered when attempting to apply the conceptual model to satellite data. Firstly, since the 3S model relies strongly on reflectance in NIR region, there are some specific hurdles that are to be expected. The strong absorption by water in the NIR greatly reduces the magnitude of the recorded signal in this region, thus reducing the signal-to-noise (SNR) ratio and enhancing the effect of inherent noise in the recorded reflectance . However, as yet, no operational atmospheric correction procedures have been developed that have proved universally robust in the NIR region across waters of varying geophysical features . In theory, the model proposed by Wang et al. , which involves the use of shortwave infrared (SWIR) bands for aerosol model selection, seems a feasible option, but some practical experiments with MODIS data has shown that even though ocean color products in turbid coastal waters can be improved using SWIR bands-based model, the extent of improvement is very limited due to the considerably lower sensor SNR values for the MODIS SWIR bands [55, 56]. The success of the application of 3S model to satellite data depends heavily on the accuracy of the atmospheric correction procedure and the retrieval of accurate reflectance. Secondly, the optical properties of TSM change with their size, shape, scattering phase function, refractive index, and others , so that the spectral slope for backscattering of TSM, εbb, in Eq. (10) is a site-specific parameter. This is to say, an optimal 3S model would heavily depend on the optical properties of TSM. Moreover, the local bio-optical information or improved model is still needed to reinitialize the site-specific parameters 3S model while the optical properties of water bodies are different from these used for model development. Finally, the calibration and validation data sets only contained a narrow range of optical properties of natural coastal waters. It is impossible to completely validate the accuracy of the model in other waters with different bio-optical properties. Thus, the improved model should be used for estimating TSM concentration in coastal waters, even though it may be essential to optimize site-specific parameters of 3S model for the given aquatic bio-optical conditions accordingly. We also suggest calibration and validation of the models based on more in situ measurements of waters with different optical properties.
Distribution of the TSM concentration is a key issue for analyzing the deposition and erosion variety of the coast and estuary, evaluating the material fluxes from river to sea, and studying oceanic sediment velocity and environment. Satellite remote sensing is a useful tool to investigate the spatial variation of TSM concentration in estuarial zones. In this study, we developed a 3S algorithm by specifying the optimal wavelengths to estimate TSM concentrations. The 3S was tested against in situ data obtained in the Changjiang River Estuary, China and results indicated that the 3S algorithm produces good performance in estimating TSM concentrations in Changjiang River Estuary, China.
While the backscattering properties of phytoplankton pigments and water molecules are insignificant or slightly significant at NIR regions, the reflectance at these wavelengths may be maximally sensitive to TSM concentration. The in situ measurements taken in Changjiang River Estuary were used to find the optimal positions of λ1 and λ2 and the width of each band in the conceptual 3S model. The results indicate that band 1 should be between 690 and 900 nm, and band 2 has to be restricted within the range from 720 to 780 nm or 840 to 900 nm.
The performance of 3S model was further evaluated using field measurements taken in Oujiang River Estuary. We found that 3S model provides an accurate prediction of TSM concentration after reinitializing the site-specific parameters of 3S model. This is to say, the 3S model is a regional model and local optical information is still needed to improve the performance of 3S model. Additionally, since the 3S model relies strongly on reflectance in NIR region, the success of the application of 3S model to satellite data depends heavily on the accuracy of the atmospheric correction procedure and the retrieval of accurate reflectance.
Appendix I. Definition on specific reflectance
For convenience in discussing the optical properties of TSM on remote sensing reflectance measured above the water-surface, an innovative definition, specific reflectance, was defined in this study. At the most fundamental, microscopic level, all scattering arises from interactions between photons and molecules or atoms. Nevertheless, scattering in natural waters is viewed as being caused by small-scale (<<λ) density fluctuations attributable to random molecular motions, by the ubiquitous large (>λ) organic and inorganic particles . Due to differences in optical properties of suspended particles, especially the size, shape, scattering phase function, and refractive index, the remote sensing reflectance measured above the water-surface may be not linearly respond to suspended particle concentration in water column. Therefore, it may be meaningful to establish a definition to identify the reflection efficiency of suspended particles or specific reflectance. The specific reflectance is defined as the ratio of reflectance to all backscattering contributors’ concentration expect for water molecular, i.e., backscattering contributions of per unit TSM concentration to reflectance (kg·sr−1·m−3). Obviously, the specific reflectance is an optical property, because its value depends on both the medium and on the directional structure of the ambient light field.
Due to the strong absorption by water in the NIR regions, the increased total reflectance in these wavelengths can be written simply as :
Where, λnir represents the band at NIR wavelengths. Therefore, the specific remote sensing reflectance at NIR wavelengths can be calculated using following formula:
Where R*rs refers to the specific remote sensing reflectance, and b*bs represents the specific backscattering coefficient. b*bs is an inherent optical property and mainly depended on particles’ size, shape, scattering phase function, and refractive index. Therefore, even though R*rs is an apparent optical property, it also can be used to identify the particle type.
This study is supported by the China State Major Basic Research Project (2013CB429701), Projects of International Cooperation and Exchanges of National Natural Science Foundation of China (41210005), the Science Foundation for 100 Excellent Youth Geological Scholars of China Geological Survey, open fund of Key Laboratory of Marine Hydrocarbon Resources and Environmental Geology (MRE201109), Serial Maps of Geology and Geophysics on China Seas and Land on the Scale of 1:1000000 (200311000001), the National Natural Science Foundation of China (No. 41106154), High-tech Research and Development Program of China (2007AA092102), and Dragon 3 Project (10470). We would also like to express our gratitude to two anonymous reviewers for their useful comments and suggestions.
References and Links
1. A. Turner and G. E. Millward, “Suspended particles: their role in estuarine biogeochemical cycles,” Estuar. Coast. Shelf Sci. 55(6), 857–883 (2002). [CrossRef]
2. L. Gao, D. J. Li, and P. X. Ding, “Variation of nutrients in response to the highly dynamic suspended particle matter in the Chang Jiang (Yangtze River) Plume,” Cont. Shelf Res. 28(17), 2393–2403 (2008). [CrossRef]
3. J. Chen, W. T. Quan, Z. H. Wen, and T. W. Cui, “An improved three-band semi-analytical algorithm in estimating chlorophyll-a concentration in high turbid Yellow River Estuary,” Environ. Earth Sci., doi.. (2013). [CrossRef]
4. N. M. Komick, M. P. F. Costa, and J. Gower, “Bio-optical algorithm evaluation for MODIS for western Canada coastal waters: An exploratory approach using in situ reflectance,” Remote Sens. Environ. 113(4), 794–804 (2009). [CrossRef]
5. Z. Mao, J. Chen, D. Pan, B. Tao, and Q. Zhu, “A regional remote sensing algorithm for total suspended matter in the East China Sea,” Remote Sens. Environ. 124, 819–831 (2012). [CrossRef]
6. R. L. Miller and B. A. Mckee, “Using MODIS Terra 250 m imagery to map concentration of total suspended matter in coastal waters,” Remote Sens. Environ. 93(1-2), 259–266 (2004). [CrossRef]
7. S. Ouillon, P. Douillet, A. Petrenko, J. Neveux, C. Dupouy, J.-M. Froidefond, S. Andréfouët, and A. Muñoz-Caravaca, “Optical algorithms at satellite wavelengths for total suspended matter in tropical coastal waters,” Sensors (Basel Switzerland) 8(7), 4165–4185 (2008). [CrossRef]
8. S. Tassan, “An improved in-water algorithm for the determination of chlorophyll and suspended sediment concentration from Thematic Mapper data in coastal waters,” Int. J. Remote Sens. 14(6), 1221–1229 (1993). [CrossRef]
9. J. Chen, M. W. Zhang, T. W. Cui, and Z. H. Wen, “A review of some important technical problems in respect of satellite remote sensing of chlorophyll-a concentration in coastal waters,” IEEE J. Sel. Topics Appl. Earth Observ ., 99, 1–15 (2013), . [CrossRef]
10. C. D. Mobley, Light and Water: Radiative Transfer in Natural Waters (Academic Press, 1994).
11. T. M. Pedersen, C. L. Gallegos, and S. L. Nielsen, “Influence of near-bottom re-suspended sediment on benthic light availability,” Estuar. Coast. Shelf Sci. 106, 93–101 (2012). [CrossRef]
12. V. Volpe, S. Silvestri, and M. Marani, “Remote sensing retrieval suspended sediment concentration in shallow waters,” Remote Sens. Environ. 115(1), 44–54 (2011). [CrossRef]
13. R. Doerffer and J. Fischer, “Concentrations of chlorophyll, suspended matter, and Gelbstoff in case II waters derived from satellite coastal coastal zone color scanner data with inverse modeling methods,” J. Geophys. Res. 99(C4), 7457–7466 (1994). [CrossRef]
14. Y. Zhang, J. Pulliainen, S. Koponen, and M. Hallikainen, “Application of an empirical neural network to surface water quality estimation in the Gulf of Finland using combined optical data and microwave data,” Remote Sens. Environ. 81(2-3), 327–336 (2002). [CrossRef]
15. A. G. Dekker, R. J. Vos, and S. W. M. Peters, “Analytical algorithms for lake water TSM estimation for retrospective analysis of TM and SPOT sensor data,” Int. J. Remote Sens. 23(1), 15–35 (2002). [CrossRef]
16. D. Doxaran, J. M. Froidefond, S. Lavender, and P. Castaing, “Spectral signature of highly turbid water application with SPOT data to quantify suspended particulate matter concentration,” Remote Sens. Environ. 81(1), 149–161 (2002). [CrossRef]
17. S. Sathyendranath, L. Prieur, and A. Morel, “A three component model of ocean color and its application to remote sensing of phytoplankton pigments in coastal waters,” Int. J. Remote Sens. 10(8), 1373–1394 (1989). [CrossRef]
18. M. W. Zhang, J. W. Tang, Q. Dong, Q. T. Song, and J. Ding, “Retrieval of total suspended matter concentration in the Yellow and East China Seas from MODIS imagery,” Remote Sens. Environ. 114(2), 392–403 (2010). [CrossRef]
19. B. Nechad, K. G. Ruddick, and Y. Park, “Calibration and validation of a generic multisensor algorithm for mapping of total suspended matter in turbid waters,” Remote Sens. Environ. 114(4), 854–866 (2010). [CrossRef]
20. A. Morel and L. Prieur, “Analysis of variances in ocean color,” Limnol. Oceanogr. 22(4), 709–722 (1977). [CrossRef]
21. J. Z. Shi, “Tidal resuspension and transport processes of fine sediment within the river plume in the partially-mixed Changjiang River Estuary, China: A personal perspective,” Geomorphology 121(3-4), 133–151 (2010). [CrossRef]
22. S.-L. Chen, G.-A. Zhang, S.-L. Yang, and J. Z. Shi, “Temporal variations of fine suspended sediment concentration in the Changjiang River Estuary and adjacent coastal waters, China,” J. Hydrol. (Amst.) 331(1-2), 137–145 (2006). [CrossRef]
23. S.-C. Hsu and F.-J. Lin, “Elemental characteristics of surface suspended particulates off the Changjiang estuary during the 1998 flood,” J. Mar. Syst. 81(4), 323–334 (2010). [CrossRef]
24. H. T. Shen, J. Li, H. F. Zhu, M. B. Han, and F. G. Zhou, “Transport of the suspended sediment in the Changjiang Estuary,” Int. J. Sediment Res. 7, 45–63 (1993).
25. F. Shen, M. H. D. Suhyb Salama, Y.-X. Zhou, J.-F. Li, Z. Su, and D.-B. Kuang, “Remote-sensing reflectance characteristics of highly turbid estuarine waters – a comparative experiment of the Yangtze River and the Yellow River,” Int. J. Remote Sens. 31(10), 2639–2654 (2010). [CrossRef]
26. L. Gao, D. Fan, D. Li, and J. Cai, “Fluorescence characteristics of chromophoric dissolved organic matter in shallow water along the Zhejiang coasts, southeast China,” Mar. Environ. Res. 69(3), 187–197 (2010). [CrossRef] [PubMed]
27. J. Chen, W. Quan, Z. Wen, and T. Cui, “A simple “clear water” atmospheric correction algorithm for Landsat-5 sensors. I: A spectral slope based method,” Int. J. Remote Sens. 34(11), 3787–3802 (2013). [CrossRef]
28. K. L. Carder, F. R. Chen, Z. P. Lee, S. K. Hawes, and J. P. Cannizzaro, “MODIS Ocean Science Team Algorithm Theoretical Basis Document: Case 2 chlorophyll a ,” ATBD 19, Version 7. (2003)
29. M. Deng and Y. Li, “Use of SeaWiFS imagery to detect three-dimensional distribution of suspended sediment,” Int. J. Remote Sens. 24(3), 519–534 (2003). [CrossRef]
30. H. J. Gons, M. T. Auer, and S. W. Effler, “MERIS Satellite Chlorophyll Mapping of Oligotrophic and Eutrophic Waters in the Laurentian Great Lakes,” Remote Sens. Environ. 112(11), 4098–4106 (2008). [CrossRef]
31. H. Ouaidrai and E. F. Vermote, “Operational atmospheric correction of Landsat TM data,” Remote Sens. Environ. 70(1), 4–15 (1999). [CrossRef]
32. K. Tachiiri, “Calculating NDVI for NOAA/AVHRR data after atmospheric correction for extensive images using 6S code: a case study in the Marsabit District, Kenya,” ISPRS J. Photogramm. 59(3), 103–114 (2005). [CrossRef]
33. M. H. Wang, S. H. Son, and W. Shi, “Evaluation of MODIS SWIR and NIR-SWIR atmospheric correction algorithms using SeaBASS data,” Remote Sens. Environ. 113(3), 635–644 (2009). [CrossRef]
34. J. L. Mueller, C. O. Davis, R. A. Arnone, R. Frouin, K. L. Carder, Z. P. Lee, R. G. Steward, S. Hooker, C. D. Mobley, and C. R. McClain, “Above-water radiance and remote sensing measurement and analysis protocols,” Ocean Optics Protocols for Satellite Ocean-Color Sensor Validation, vol. Revision 4, III: Radiometric Measurements and Data Analysis Protocols, pp. NASA Tech. Memo (2003).
36. H. R. Gordon, O. B. Brown, R. H. Evans, J. W. Brown, R. C. Smith, K. S. Baker, and D. K. Clark, “A Semianalytic Radiance Model of Ocean Color,” J. Geophys. Res. 93(D9), 10909–10924 (1988). [CrossRef]
38. R. P. Bukata, J. H. Jerome, K. Y. Kondratyev, and D. V. Pozdnyakov, Optical Properties and Remote Sensing of Inland and Coastal Waters, 1st Ed.(CRC Press, 1995).
39. R. C. Smith and K. S. Baker, “Optical classification of natural waters,” Limnol. Oceanogr. 23(2), 260–267 (1978). [CrossRef]
40. G. Dall’Olmo and A. A. Gitelson, “Effect of bio-optical parameter variability on the remote estimation of chlorophyll-a concentration in turbid productive waters: experimental results,” Appl. Opt. 44(3), 412–422 (2005). [CrossRef] [PubMed]
41. A. A. Gitelson, G. Dall'Olmo, W. Moses, D. C. Rundquist, T. Barrow, T. R. Fisher, D. Gurlin, and J. Holz, “A simple semi-analytical model for remote estimation of chlorophyll-A in turbid waters: Validation,” Remote Sens. Environ. 112(9), 3582–3593 (2008). [CrossRef]
42. K. L. Carder, F. R. Chen, Z. P. Lee, S. K. Hawes, and D. Kamykowski, “Semi-analytical moderate resolution imaging spectrometer algorithms for chlorophyll a and absorption with bio-optical domains based on nitrate-depletion temperatures,” J. Geophys. Res. 104(C3), 5403–5421 (1999). [CrossRef]
43. T. J. Smyth, G. F. Moore, T. Hirata, and J. Aiken, “Semi-analytical model for the derivation of ocean color inherent optical properties: description, implementation, and performance assessment,” Appl. Opt. 45(31), 8116–8131 (2006). [CrossRef] [PubMed]
44. D. Doxaran, J. M. Froidefond, and P. Castaing, “Remote-sensing reflectance of turbid sediment-dominated waters. Reduction of sediment type variations and changing illumination conditions effects by use of reflectance ratios,” Appl. Opt. 42(15), 2623–2634 (2003). [CrossRef] [PubMed]
45. D. H. Schoellhamer, T. E. Mumley, and J. E. Leatherbarrow, “Suspended sediment and sediment-associated contaminants in San Francisco Bay,” Environ. Res. 105(1), 119–131 (2007). [CrossRef] [PubMed]
46. A. A. Gitelson and M. N. Merzlyak, “Spectral Reflectance Changes Associated with Autumn Senescence of Aesculus Hippocastanum L. and Acer Platanoides L. Leaves. Spectral Features and Relation to Chlorophyll Estimation,” J. Plant Physiol. 143(3), 286–292 (1994). [CrossRef]
47. P. B. Robert, S. Alexander, and Y. K. Kirill, Optical Properties and Remote Sensing of Inland and Coastal Waters, 1st Ed. (CRC Press, 1995).
48. J. Chen and W. T. Quan, “An improved algorithm for retrieving chlorophyll-a from the Yellow River Estuary using MODIS imagery,” Environ. Monit. Assess. 185(3), 2243–2255 (2013). [CrossRef] [PubMed]
49. P. Huck, B. Light, H. Eicken, and M. Haller, “Mapping sediment-laden sea ice in the Arctic using AVHRR remote sensing data: Atmospheric correction and determination of reflectances as a function of ice type and sediment load,” Remote Sens. Environ. 107(3), 484–495 (2007). [CrossRef]
50. F. D’Ortenzio, S. Marullo, M. Ragni, M. R. d’Alcala, and R. Santoleri, “Validation of empirical SeaWiFS algorithms for chlorophyll-a retrieval in the Mediterranean Sea: A case study for oligotrophic seas,” Remote Sens. Environ. 82(1), 79–94 (2002). [CrossRef]
51. D. Doxaran, J. M. Froidefond, P. Castaing, and M. Babin, “Dynamics of the turbidity maximum zone in a macrotidal estuary (the Gironde, France): Observations from field and MODIS satellite data,” Estuar. Coast. Shelf Sci. 81(3), 321–332 (2009). [CrossRef]
52. M. Fettweis, B. Nechad, and D. V. Eynde, “An esitmate of the suspended particulate matter transport in the southern North Sea using SeaWiFS images, in situ measurements and numverical model results,” Cont. Shelf Res. 27, 1568–1583 (2007).
53. D. G. Bowers, K. M. Braithwaite, W. A. M. Nimmo-Smith, and G. W. Graham, “Light scattering by particles suspended in the sea: the role of particle size and density,” Cont. Shelf Res. 29(14), 1748–1755 (2009). [CrossRef]
54. D. G. Bowers and C. E. Binding, “The optical properties of mineral suspended particles: A review and synthesis,” Estuar. Coast. Shelf Sci. 67(1-2), 219–230 (2006). [CrossRef]
55. J. Chen, W. T. Quan, M. W. Zhang, and T. W. Cui, “A simple atmospheric correction for MODIS in shallow turbid waters: a case study in Taihu Lake,” IEEE J. Sel. Topics Appl. Earth Observ (2012), doi. [CrossRef] .
56. P. J. Werdell, B. A. Franz, and S. W. Bailey, “Evaluation of shortwave infrared atmospheric correction for ocean color remote sensing of Chesapeake Bay,” Remote Sens. Environ. 114(10), 2238–2247 (2010). [CrossRef]