Home Journals Articles Updates About
CN
image
收藏切换
The dynamic observation of dissolved organic matter in the Zhujiang (Pearl River) Estuary in China from space
收藏切换
PDF
Dong LIU1, 2, Yan BAI1, *, Xianqiang HE1, Delu PAN1, Difeng WANG1, Ji'an WEI1, Lin ZHANG1
Acta Oceanologica Sinica | 2018, 37(7) : 105 - 117
Less
收藏切换
Acta Oceanologica Sinica | 2018, 37(7): 105-117
• Marine Technology •
The dynamic observation of dissolved organic matter in the Zhujiang (Pearl River) Estuary in China from space
Full
Dong LIU1, 2, Yan BAI1, *, Xianqiang HE1, Delu PAN1, Difeng WANG1, Ji'an WEI1, Lin ZHANG1
Affiliations
  • 1 State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, State Oceanic Administration, Hangzhou 310012, China
  • 2 Key Laboratory of Watershed Geographic Sciences, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, China
Published: 2018-07-25 doi: 10.1007/s13131-017-1248-7
Outline
收藏切换

The distributions of estuarine colored dissolved organic matter (CDOM) are the combined results of physical-biogeochemical processes. Remote sensing is needed to monitor highly dynamically estuarine CDOM. Using in situ data from four seasonal cruises, an algorithm is developed to estimate CDOM absorption coefficient at 400 nm (aCDOM(400)) in the Zhujiang (Pearl River) Estuary (ZJE). The algorithm uses band ratios of Rrs(667)/Rrs(443) and Rrs(748)/Rrs(412). By applying it to moderate resolution imaging spectroradiometer onboard Aqua satellite (MODIS/Aqua) data from 2002 to 2014, seasonal climatology aCDOM(400) in the ZJE is calculated. CDOM distributions are majorly influenced by water discharge from the Zhujiang River and underwater topography. Along the section vertical to a water depth gradient, the seasonal aCDOM(400) exponentially decreased (y=aebx, b<0), but with great differences among seasons. Riverine fresh water is the primary source of CDOM in the ZJE. Fulvic acid fraction decreases with increasing salinity. Using developed algorithms, conservative CDOM mixing equation, and river discharge, effective riverine end-member concentration and flux of dissolved organic carbon (DOC) in summer and winter from 2002 to 2014 are first estimated from the MODIS/Aqua data. Both effective riverine end-member DOC concentration and flux are positively related to the river discharge, significantly in summer with R2 of 0.698 for concentration and 0.965 7 for flux.

colored dissolved organic matter  /  dissolved organic carbon  /  Zhujiang (Pearl River) Estuary  /  effective riverine end-member flux  /  remote sensing
Dong LIU, Yan BAI, Xianqiang HE, Delu PAN, Difeng WANG, Ji'an WEI, Lin ZHANG. The dynamic observation of dissolved organic matter in the Zhujiang (Pearl River) Estuary in China from space[J]. Acta Oceanologica Sinica, 2018 , 37 (7) : 105 -117 . DOI: 10.1007/s13131-017-1248-7
Full Text
Less
1 Introduction
Less
The global lateral flux of dissolved organic matter (DOM), determined as dissolved organic carbon (DOC) and hereinafter, into estuaries by rivers was estimated at 210 Mt/a (calculated by carbon) (Dai et al., 2012). Phytoplankton production and sediment pore water can also add DOM to estuarine waters. Riverine or autogenous DOM can influence toxic metal transport, oxygen consumption, etc. (Dai et al., 2012; Chen et al., 2013; Guo et al., 2013) so as to change estuarine water quality. Moreover, the optically active DOM fraction, colored dissolved organic matter (CDOM), can absorb light especially for that with short wavelength (Chen et al., 2004, 2013; Huang and Chen, 2009). As results, CDOM may increase primary production by absorbing harmful ultraviolet light on the one side, or reduce photosynthetic production in deep waters due to limitation of photosynthetic active radiation on the other (Huang and Chen, 2009). Therefore, monitoring CDOM distribution and its composition have great significance to estuarine water resource management. Except for CDOM, moreover, non-colored DOM can also influence estuarine carbon cycle, oxygen consumption, etc. (Dai et al., 2012; Chen et al., 2013; Guo et al., 2013).
Estuarine CDOM concentration, determined as light absorption coefficient at a specific wavelength, can be monitored from space in high tempo-spatial resolution using band ratio algorithm (Carder et al., 2003; Park et al., 2014; Siegel et al., 2002; Siswanto et al., 2011). Tiwari and Shanmugam (2011) recommended a linear equation using the ratio of remote sensing reflectance at 670 and 490 nm (Rrs(670)/Rrs(490)) to derive CDOM in both coastal and open ocean waters. For waters influenced by the Mississippi River during low flow conditions, D’sa and Miller (2003) reported that power law equation using the ratio of remote sensing reflectance at 510 and 555 nm (Rrs(510)/Rrs(555)) was the best CDOM inversion algorithm. However, these algorithms were not very appropriate for all kinds of waters with various inherent optical properties because of different constituents (Sathyendranath, 2000). Local algorithm should be developed to inverse estuarine CDOM accurately. For remote sensing monitoring of CDOM in the Zhujiang (Pearl River) Estuary (ZJE), Chen et al. (2003) first used Rrs(670)/Rrs(412) to inverse CDOM from simulated water reflectance from suspended particle, chlorophyll a (Chl a) and DOC concentrations. Based on in situ CDOM and Hyperion satellite data, Fang et al. (2009) also reported a weak relationship between CDOM and Rrs(703)/ Rrs(488). By using in situ data and extending the quasi-analytical algorithm (QAA) (Lee et al., 2002), Dong et al. (2013) recently derived the absorption coefficient of CDOM in the South China Sea and the Taiwan Strait close to the ZJE. From our in situ data in the ZJE, however, we got that all of these algorithms did not perform well for the local ZJE.
A great amount of DOM was discharged into the ZJE by the Pearl River, the 14th largest world river with discharge of about 343 km3/a (Cai et al., 2008). Riverine DOC flux was calculated by multiplying river discharge by average DOC concentration over a given time period (Bauer et al., 2013). However, the DOC concentration spatially varies along the river stream (Wu et al., 2007). Gao et al. (2002) estimated DOC flux at the Makou Hydrometric Station of the Xijiang River, accounting for 78% of drainage area of the Zhujiang River, to be 0.66 Mt C/a (calculated by carbon). Ni et al. (2008) reported the total DOC flux at eight outlets of the Zhujiang River to be 0.38 Mt/a using in situ data sampled monthly from March 2005 to February 2006. This estimation could denote annual riverine DOC flux into the ZJE. Once entering the estuary, however, there were rapid transformations from riverine DOC to particulate organic carbon (POC) through flocculation (Bauer et al., 2013) and from riverine POC to DOC through microbial activities (Dai et al., 2000) in the low salinity waters. Therefore, an effective riverine end-member DOC concentration was defined to denote the resultant riverine DOC after various transformations in low salinity waters (Section 4.3) (Cai et al., 2004).
In this study, remote sensing algorithms for inversing CDOM absorption coefficient at 400 nm (aCDOM(400)) and its spectral slope (SCDOM) were developed using synchronous in situ water reflectance spectra and CDOM data from four seasonal survey cruises in the ZJE firstly. Then, the developed algorithms were applied on moderate resolution imaging spectroradiometer onboard Aqua satellite (MODIS/Aqua) data to calculate a seasonal climatology aCDOM(400) and SCDOM in the ZJE. Thirdly, effective riverine end-member DOC fluxes in summer and winter from 2002 to 2014 were also estimated from the MODIS/Aqua data. Furthermore, we also discussed CDOM dynamics including its sources, impact factors, and distributions in the ZJE. In short, using the developed algorithms, CDOM dynamics in the ZJE can be monitored in high tempo-spatial resolution from space.
2 Study area
Less
ZJE receives fresh water from the second largest China’s river, the Zhujiang River, ranked after the Changjiang River and before the Huanghe River according to annual discharge (Cai et al., 2008). Fresh water is discharged into the ZJE through eight outlets, Humen, Jiaomen, Hongqili, Hengmen, Modaomen, Jitimen, Hutiaomen and Yamen (Fig. 1b). The Zhujiang River Basin is located in the subtropical zone and affected by the East Asian Monsoon. Annual fresh water from the Zhujiang River is unevenly seasonally distributed. About 80% of fresh water from the Zhujiang River occurs during the wet season from April to September, with only 20% happens in the dry season from October to March (Chen et al., 2004; Bai et al., 2015). The total suspended matter (TSM) concentration is high in the ZJE, especially for the near shore waters. On the one hand, the Zhujiang River is the 14th largest world river (Cai et al., 2008), with annual discharge of 343 km3/a. Under the influence of the East Asian Monsoon, it discharges more highly turbid water than Europe and America rivers into the ZJE (Ran et al., 2013). On the other hand, suspended matter can scatter sun light and increase water reflectance. Most of in situ Rrs(490) used by D’sa and Miller. (2003) from the Gulf of Mexico, Tiwari and Shanmugam (2011) from global coastal oceans were smaller than 0.01 sr–1. However, 55 of the 67 in situ Rrs(490) in the ZJE from our cruises were larger than 0.01 sr–1.
The DOM concentration intensely varied in the ZJE, with aCDOM(400) as high as 1.134 m–1 in the low salinity waters and only 0.04 m–1 in the high salinity waters. The ZJE connects the Zhujiang River to the SCS. The part north of Wanshan is also called as the Lingding Bay. The topography of the Lingding Bay presents as high in the west and low in the east (Fig. 1b). Most fresh water is discharged into the ZJE from the shallowly west side, where all outlets of the Zhujiang River are located. However, along the two eastern water channels saline waters can intrude into the ZJE from the deeply east part when tiding, even into the northernmost outlet at high tide (Fig. 1b) (Liu et al., 2015). As a result of interactions between fresh and saline waters, a counter-clockwise flow can be observed in the Lingding Bay (Chen et al., 2004). The area south of Wanshan is significantly influenced by coastal currents along Guangdong Province, China. DOM dynamics are the combined results of various physical-biogeochemical processes in the ZJE (Callahan et al., 2004; Chen et al., 2004; He et al., 2010).
3 Materials and ZJE processing
Less
3.1 In situ data
3.1.1 CDOM and DOC
Four seasonal survey cruises were conducted in autumn (November 2013), winter (February 2014), spring (May 2014), and summer (August 2014) in the ZJE. Sampling sites in different cruises are shown in Fig. 1b. CDOM samples were collected following the Ocean Optics Protocols proposed by the NASA (Mitchell et al., 2000). Surface water was collected using a Niskin water sampler (General Oceanics Inc, USA), stored in brown glass bottles until filtration so as to keep away from sun light (Liu et al., 2015), filtrated on board through polycarbonate filters (millipore, 0.22 μm, and Φ 47 mm) at a low vacuum around 125 mmHg (Bai et al., 2013). The CDOM samples were strained straightly into CNWTM brown glass bottles with teflon gaskets. All CNWTM bottles had been soaked in 10% HCl, rinsed with purified Milli-Q water, and combusted at 450°C for 4 h (Bai et al., 2013). After collection, the CDOM samples were immediately placed into a refrigerator with temperature around –20°C until laboratory measurement (Bai et al., 2013; Liu et al., 2014). Ranging from 250 to 800 nm with an interval of 1 nm, the light absorbance of CDOM sample (ODCDOM) and purified Milli-Q water (ODMilli-Q) were both measured using a Lambda 35 ultraviolet-visible light spectrophotometer (Perkin Elmer, Inc). Then, the CDOM absorption coefficient at a specific wavelength λ (aCDOM(λ)) was calculated using Eq. (1) (Bai et al., 2013).
$\begin{split} {{a}_{\rm{CDOM}}}(\text{λ} )= & (\text{2}.303/l)\times ((OD_{\rm{CDOM}}(\text{λ} )- \\ & O{{D}_{\rm{Milli-Q}}}(\text{λ} ))-O{{D}_{\rm{null}}}), \\ \end{split}$
where ODCDOM(λ) and ODMilli-Q(λ) are respectively measured the light absorbance of CDOM sample and purified Milli-Q water at wavelength λ; l denotes the quartz cell path length (0.1 m); ODnull is the residual absorbance offset at the long wavelength, average value of (ODCDOM-ODMilli-Q) from 695 to 705 nm was used. Some examples of CDOM absorption coefficient spectra for different salinities are shown in Fig. 2a. With known aCDOM(400), the absorption coefficient at any wavelength λ (aCDOM(λ)) can be calculated using the embedded equation, with SCDOM of the spectral slope (Fig. 2a).
DOC samples were also collected following the protocols established by the joint global ocean flux study (JGOFS) (Knap et al., 1994). The DOC samples were measured via a high temperature combustion method (680°C) using a total organic analyzer (Shimadzu, Inc) by referring to standard seawater samples provided by the Hansell Laboratory, Miami University. For a specific DOC sample, an analytic precision was within 2% on triplicate injections (Liu et al., 2014). More details about collection, storage, and measurement of DOC samples can be found in Liu et al. (2014).
3.1.2 Water reflectance
When collecting water, the radiances of water surface (Lws), sky light (Lsky), and a standard reflecting plate (Lp) were also measured by a field Spec spectroradiometer (Analytical Spectral Devices Inc,) using the above-water method following the ocean optics protocols proposed by the NASA (Mueller et al., 2003; Bai et al., 2013; Liu et al., 2015). Before the cruises, absolute calibration was performed to the instrument using an NIST traceable lamp in the laboratory (He et al., 2013). When measuring Lws and Lsky, the probe was positioned at an angle of 90°–135° to the plane of an incident radiation so as to minimize the effects of sun glint and ship shadow (Liu et al., 2015). A view angle to the aplomb direction was 135°–150° for Lws, but 30°–45° for Lsky. After measuring Lws and Lsky, Lp was collected similar to Lws but with probe vertical to the standard reflecting plate. The Field Spec Spectroradiometer recorded Lws, Lsky and Lp ranging from 350 nm to 2 390 nm with an interval of 1 nm. Then, water reflectance (Rrs) was calculated as follows:
${R_{{\rm{rs}}}}(\text{λ}) = \rho (\text{λ} )({L_{{\rm{ws}}}}(\text{λ} ) - r{L_{{\rm{sky}}}}(\text{λ} ))\text{/} (\text{π} {L_{\rm{p}}}(\text{λ} )),$
where for a specific wavelength λ, Rrs(λ) is the water reflectance; ρ(λ) is the reflectance of the standard reflecting plate; Lws(λ), Lsky(λ) and Lp(λ) are the radiance of water surface, sky light, and the standard reflecting plate respectively; r is the reflectance of air-water interface, with r=0.28 for the waters in the ZJE (Liu et al., 2015).
For a specific band of MODIS/Aqua, it majorly received radiance within a ZJE viously designed band range. To develop CDOM algorithms applicable to MODIS/Aqua data, equivalent water reflectance (requi, sr–1) for its ocean color bands of the in situ water reflectance was computed firstly. For a specific band, requi was calculated using Eq. (2) (Liu et al., 2015).
$r_{{\rm{equi}}}(\text{λ} ) = \frac{{\displaystyle\int\limits_{350}^{800}{f(\text{λ} ){R_{{\text{rs}}}}(\text{λ} )I(\text{λ} ){\rm{d}}\text{λ}} }}{{\displaystyle\int\limits_{350}^{800}{f(\text{λ})I(\text{λ}){\rm{d}}\text{λ} } }},$
where λ (412, 443, 488, 531, 551, 667, 678 or 748 nm) is the central wavelength representing the specific band. For band λ, f (λ) is the spectral response function from the OceanColor Website (http://oceancolor.gsfc.nasa.gov/); L(λ) is the solar irradiance at the mean earth-sun distance. For band λ of MODIS/Aqua, f (λ) is close to 0 beyond 350–800 nm, so we set the integration range of 350–800 nm.
3.1.3 Others
A SBE 19 plus CTD (Sea-Bird Electronics, Inc., USA) was used to record profile parameters at a frequency of 6 Hz. For a specific station, surface salinity was calculated by averaging CTD records within 0.5 m depth. In the summer cruise, the seapoint ultraviolet fluorometer was employed to record underway CDOM fluorescence, with an excitation wavelength of 370 nm and emission wavelength of 440 nm.
3.2 Satellite data
Daily level 1 B calibrated radiance products of MODIS/Aqua (MYD021KM, Version 2.16) were downloaded from the Level 1 and atmosphere archive and distribution system (http://ladsweb. nascom.nasa.gov/). Level 1 B data of MODIS/Aqua in January, April, July and November were downloaded from 2002 to 2014. All MODIS/Aqua data scanned the ZJE at about 1:30 pm local time with spatial resolution at the nadir point of 1 000 m×1 000 m for ocean color bands. On the basis of the SeaWiFS data analysis system (SeaDAS, Version 6.4), a near-infrared and shortwave infrared bands atmospheric correction algorithm was applied to MYD021KM to remove atmospheric effects and retrieve the water reflectance. When processing the MODIS/Aqua Level 1 B data, the L2 processing flags set by the atmospheric correction algorithm was used to identify and exclude questionable pixels, such as pixels of land, cloud, sun glint, etc. (Bailey and Werdell, 2006). Pixel with the reflectance of band centered at 2 030 nm larger than 0.018 was labeled as cloud. The atmospheric correction algorithm can derive the water reflectance of clearly open sea as well as turbid estuary from MODIS/Aqua Level 1 B data (Wang and Shi, 2007). Finally, the output reflectance data with precisely geographic coordinate was clipped for studying the ZJE (Fig. 1).
4 Remote sensing algorithms
Less
4.1 CDOM
As one of ocean color products, CDOM was often derived from optical satellite data (Siegel et al., 2002; Belanger et al., 2008; Mannino et al., 2008). Most previous CDOM algorithms were based on band ratios which could enhance CDOM signal and weaken effect of atmospheric correction error (Chen et al., 2003; Fang et al., 2009; Siswanto et al., 2011). However, algorithms for open oceans or other estuarine waters may not be suitable for intensively varied waters in the ZJE (Sections 1 and 2), of which two broad classes were identified: one was with high CDOM concentration and water reflectance, but the other one was opposite (Fig. 2). For the ZJE, temperature and discharge from the Zhujiang River are high in summer, but on the contrary in winter (Section 2). Spring and autumn are the two transitional seasons. Therefore, the CDOM data (Section 3.1) and equivalent water reflectance (Eq. (3)) from winter and summer cruises, 45 in all, were applied to develop and parameterize the CDOM algorithm. In this case, the developed algorithm can be used to all seasons.
For the ZJE, aCDOM(400) was positively linearly related to TSM concentration (R2=0.28, P<0.05). Suspended particle can significantly scatter light and increase the water reflectance at the long wavelength. Both water reflectance at 667 nm (Rrs(667)) and 748 nm (Rrs(748)) were positively linearly correlated with the TSM concentration, with R2=0.52 for Rrs(667) (P<0.05) and R2=0.17 for Rrs(748) (P<0.05). Moreover, strong CDOM light absorption at 412 nm and 443 nm could decease the water reflectance Rrs(667) and (Rrs(748)) (Fig. 2a). Therefore, both ratios of Rrs(667) to Rrs(443) and Rrs(748) to Rrs(412) can denote the CDOM concentration in the ZJE. To inverse CDOM concentration in the ZJE, in fact, Chen et al. (2003) and Fang et al. (2009) previously used ratios of Rrs(670) to Rrs(412) and Rrs(703) to Rrs(488), respectively. After testing various band combinations, the CDOM algorithm that was applicable to intensely dynamic waters in the ZJE was shaped and parameterized as follows (Fig. 3).
$\begin{gathered}{a_{{\rm{CDOM}}}}\left( {400} \right) = 0.158\,\,1{\left( {\frac{{{r_{{\rm{equi}}}}\left( {667} \right)}}{{{r_{{\rm{equi}}}}\left( {443} \right)}}} \right)^{1.626\,\,7}} {\left( {\frac{{{r_{{\text{equi}}}}\left( {784} \right)}}{{{r_{{\rm{equi}}}}\left( {412} \right)}}} \right)^{ - 0.981\,\,7}},\\ N = 45,\,\,\,{R^2} = 0.923\,\,0,\qquad\qquad\qquad\qquad\end{gathered}$
where aCDOM(400) denotes CDOM absorption coefficient at 400 nm; requi(667), requi(443), requi(748) and requi(412) were defined as Eq. (3). With R2=0.923 0, the used wavelengths can reflect different sources of CDOM in the ZJE (Eq. (4)). Arenz et al. (1996) reported a similar algorithm to inverse the DOC concentration in eight reservoirs of the Colorado Front Range. The detailed process was shown in Fig. 3.
Moreover, an absolute error (AE), a relative error (RE), and a mean square error (RMSE) were employed to evaluate the modeling results. They are defined by referring to Świrgoń and Stramska (2015) as follows:
$\left. \begin{array}{l}{{AE }} = {\rm{ }}{V_{{\rm{modeled}}}} - V{_{{{in}}\;\,\,{{situ}}}}\\{{RE }} = {\rm{ (}}{V_{{\rm{modeled}}}} - {V_{{{in}}\;\,\,{{situ}}}}{\rm{ )}} \text{/} {V_{{{in}}\;\,\,{{situ}}}} \times {\rm{100\% }}\\{{RMSE = }}\sqrt {\sum\limits_{i = 1}^N {{{({V_{{in}\,\,\;{situ}}} - {V_{{\rm{modeled}}}})}^2}{\text{/}} (N - 1)} } \end{array} \right\},$
where Vmodeled denotes the modeled value; Vin situ represents the in situ value; and N is the sample size.
4.2 CDOM spectral slope
Waters with same aCDOM(400) may vary in composition due to different sources. Optical properties can be used to explore the CDOM composition (Chen et al., 2004). The spectral slope is the fitting slope of the CDOM absorption coefficient (Fig. 2a). SCDOM has been successfully applied to indicate sources and transformation processes of CDOM (Jaffé et al., 2004). During the cruises, the in situ SCDOM varied between 0.010 7 nm–1 and 0.017 6 nm–1, which were consistent with the observations of Hong et al. (2005). For the Gulf of Mexico, Carder et al. (1989) calculated the spectral slope using a spectral range of 370–440 nm. In order to compare with their results, we also calculated SCDOM over the same spectral range. SCDOM was deduced from the CDOM absorption equation (Fig. 2a) and shaped as as follows (Fig. 3).
${S_{{\rm{CDOM}}}} = \frac{{\ln ({a_{{\rm{CDOM}}}}(370)/{a_{{\rm{CDOM}}}}(440))}}{{70}},$
where aCDOM(370) and aCDOM(440) are the CDOM absorption coefficient at 370 nm and 440 nm, respectively. For all in situ data, aCDOM(370) and aCDOM(440) were greatly linearly related to aCDOM(400), with R2 of 0.995 8 and 0.992 7, respectively. From the CDOM absorption equation embedded in the Fig. 2a, moreover, both aCDOM(370) and aCDOM(440) can be inversed using a similar satellite algorithm as aCDOM(400). Using the in situ data from four conducted cruises in the ZJE (57 in all), Eq. (6) was parameterized, with RMSE of 0.837 9 nm–1 and mean absolute RE of 4.364 7%. The parameterized SCDOM algorithm is shown as follows:
$\begin{array}{*{20}{l}} {{S_{{\text{CDOM}}}} \times 1\,000 = 14.235 + 3.0558\,\,{\text{ln}}\left( {\frac{{{r_{{\text{equi}}}}({\text{667}})}}{{{r_{{\text{equi}}}}({\text{443}})}}} \right)} - \\ { 1.1843\,\,{\text{ln}}\left( {\frac{{{r_{{\text{equi}}}}({\text{748}})}}{{{r_{{\text{equi}}}}({\text{412}})}}} \right),\,\,N = 57,\,\,{R^2} = 0.779 \,\, 1,} \end{array}$
where requi(667), requi(443), requi(748) and requi(412) were defined as Eq. (3).
4.3 Effective riverine end-member DOC
CDOM was usually conservatively mixed in estuaries (Pan et al., 2012). Hong et al. (2005) reported that CDOM was conservatively mixed in the ZJE. Our in situ data also evidenced the conservative mixing by a negatively linear relation between aCDOM(400) and salinity (Fig. 4a). For DOC, Dai et al. (2000) reported that DOC concentration followed a mixing line beyond salinity 5 until 25 in winter (March 1997) in the ZJE. The conservative mixing was disturbed by DOC from transformations of particulate organic carbon in low salinity waters (Dai et al., 2000) or DOC production by phytoplankton in high salinity waters. Figure 4 shows DOM mixing diagram in the ZJE. Supposing DOC is conservatively mixed in waters with salinity from Slow to Shigh (Fig. 4b), effective riverine end-member DOC concentration (Ce) can be calculated using the DOC conservative mixing line (Fig. 4b). However, DOC mixing in estuary was often not conservative and seasonally varied (Pan et al., 2012). For the four conducted cruises, conservative DOC mixing was only significant for waters in summer with salinities larger than 1.5 (R2=0.798 and N=28) and winter with salinities larger than 10 (R2=0.789 8 and N=27). Therefore, only seasonal Ce in summer and winter was estimated in this study.
For a specific time, using the CDOM mixing equation (Fig. 4a) and the developed CDOM algorithm (Eq. (4)), the salinities in Regions A (SA) and B (SB) were calculated (Fig. 1b). Furthermore, Chen et al. (2003) reported a remote sensing algorithm to inverse the DOC concentration in the ZJE. Using all in situ data from four conducted cruises in the ZJE (Section 3.1), their DOC algorithm was parameterized, with RMSE 0.156 6 mg/L of and mean absolute RE of 10.452 7%. The parameterized equation is shown as Eq. (8):
$\begin{split} \ln [{\text{DOC}}]{\text{ = }}0.265\;9 \times \ln \frac{{{r_{{\text{equi}}}}(667)}}{{{r_{{\text{equi}}}}(412)}} + 0.248\;8, \\ N{\text{ = 57}},{R^2} = 0.490\;8, \quad\qquad\qquad\qquad\end{split}$
where requi(667) and requi(412) are defined as in Eq. (3). Using Eq. (8), the DOC concentrations in Regions A (DOCA) and B (DOCB) were calculated. On the basis of the above estimated salinities and DOC concentrations in Regions A and B, the conservative mixing line of DOC in Fig. 4b was parameterized and applied to compute Ce shown as follows:
${C_{\rm{e}}} = ((DO{C_{\rm{A}}} - DO{C_{\rm{B}}}) \times {S_{\rm{A}}})/({S_{\rm{B}}} - {S_{\rm{A}}}) + DO{C_{\rm{A}}}{\rm{.}}$
Finally, effective riverine end-member DOC flux (Fe) was also calculated by multiplying Ce by discharge of the Zhujiang River. The detailed processes are shown in Fig. 3.
5 Results
Less
5.1 CDOM algorithm validation
5.1.1 Comparing with in situ aCDOM(400)
The in situ data from autumn cruise (12 in total) was used to validate the reliability of the developed CDOM algorithm (Eq. (4)). Figure 5 shows the validation results of modeled aCDOM(400) based on the in situ aCDOM(400). For all the 57 in situ data, RMSE, mean absolute AE (|AE|), and mean absolute RE (|RE|) of modeled aCDOM(400) were 0.073 6 m–1, 0.052 1 m–1, and 15.874 4%, respectively (Fig. 5). For the 12 testing data alone, RMSE, mean |AE|, and mean |RE| were 0.062 7 m–1, 0.048 2 m–1, and 12.428 7%, respectively. Application results of the new developed algorithm to in situ the water reflectance were satisfactory. Using the developed algorithm, the CDOM concentration in the ZJE can be inversed from the water reflectance.
5.1.2 Comparing with other algorithms
Furthermore, the developed CDOM algorithm was compared with other three algorithms which had been developed to inverse coastal CDOM concentration (Table 1). All the four algorithms got preferable modeling results for waters with aCDOM(400) around 0.35 m–1. For waters with aCDOM(400) higher than 0.6 m–1, however, algorithm proposed by Dong et al. (2013) significantly overestimated aCDOM(400), but for other three algorithms (Fig. 6). For waters with aCDOM(400) lower than 0.3 m–1, on the contrary, algorithm proposed by Dong et al. (2013) significantly underestimated aCDOM(400), but for other three algorithms (Fig. 6).
For algorithms proposed by D’sa and Miller (2003) and Tiwari and Shanmugam (2011), modeling results were similar for waters with high aCDOM(400), but the latter one got greater RE for waters with aCDOM(400) lower than 0.3 m–1, with unexpectedly great RE of 264.94% (Fig. 6). Water environment in the ZJE was locally characterized, with high TSM concentration (Ni et al., 2008) and aCDOM(400). When water was turbid, Zhu et al. (2011) previously reported that the accuracy of QAA algorithm generally degraded rapidly with increasing CDOM and non-algal particle concentrations. With poor regional and seasonal knowledge of inherent optical properties in the ZJE, the QAA algorithm also did not perform well in the ZJE (Fig. 6). With least |RE| (15.87%), in general, the new developed algorithm (Eq. (4)) was better than other three algorithms (Fig. 6a). The developed algorithm was more applicable to the highly turbid waters in the ZJE.
5.1.3 Application to MODIS/Aqua data
The developed algorithm was applied to the satellite water reflectance of MODIS/Aqua so as to further validate its availability (Section 3.2). When matching in situ measurement and satellite observation, we set the space window as 3×3 pixels so as to account for possible navigation error for the tempo-spatially dynamic waters in the ZJE (Bailey and Werdell, 2006). In order to allow for a greatest match possibility, the time window was set as ±3 h (Bailey and Werdell, 2006; He et al., 2013). To do the matching up for the coastal waters in the ZJE, mean value was calculated only for sampled stations with a minimum of five valid pixels. With the space window of 3×3 pixels and time window of ±3 h, only six stations were found. The AE and RE of satellite-derived aCDOM(400) are shown in Table 2. For aCDOM(400), mean |AE| was 0.067 m–1, with minimum of 0.000 4 m–1 and maximum of 0.200 3 m–1; mean |RE| was 18.127%, with minimum of 0.507 5% and maximum of 44.773 8% (Table 2). In general, the application results were acceptable.
Moreover, an underway CDOM concentration was also used to evaluate application of the developed algorithm to MODIS/ Aqua data. Underway aCDOM(400) was computed from recorded CDOM fluorescence (QSU) through the fitted linear equation embedded in Fig. 7a. When matching underway measurement and satellite observation, similar screening rules as the in situ measurement were adopted. However, the mean value of underway aCDOM(400) 1 km around the central pixel of the 3×3 pixels array was calculated . For the 65 matched underway aCDOM(400), validation results were also desirable, with RMSE of 0.094 1 m–1, mean |AE| of 0.068 6 m–1, and mean |RE| of 11.241 4% respectively (Fig. 7b). In sum, based on the MODIS/Aqua data and the Eq. (4), distributions and dynamics of CDOM in the ZJE can be monitored in high tempo-spatial resolution from space.
5.2 Tempo-spatial variations of CDOM
Table 3 shows the statistical results of in situ aCDOM(400) from four seasonal cruises. aCDOM(400) was negatively linearly related to salinity (Fig. 4a). With different salinities, aCDOM(400) changed with sampling sites (Fig. 1b). Discharge from the Zhujiang River was lowest in winter, with minimum in situ salinity of 13.904 and mean aCDOM(400) of only 0.212 m–1. When with the highest discharge in summer, the minimum in situ salinity and mean aCDOM(400) were 0.047 m–1 and 0.448 m–1, respectively. Moreover, in situ aCDOM(400) were similar in autumn and spring when with similar discharges (Table 3). In short, in situ aCDOM(400) presented greatly tempo-spatial variations in the ZJE.
Equation (4) was applied on daily satellite water reflectance to calculate daily aCDOM(400) from 2002 to 2014 (Section 3.2). Then, monthly satellite climatology aCDOM(400) was calculated by averaging daily aCDOM(400) of a specific month. The results are shown as Fig. 8. Agreeing with the in situ data (Table 3), satellite-derived seasonal aCDOM(400) intensely varied with season and space. For the same area, aCDOM(400) in summer was significantly higher than those in other seasons, especially for near shore waters (Fig. 8). Surface area percent of waters with aCDOM(400) ranging from 0 to 1 m–1 (with interval of 0.1 m–1) is shown in Fig. 9a. aCDOM(400) in most areas was lower than 0.3 m–1, with maximum area percent of 28.73% from 0.1 to 0.2 m–1 in autumn, 50.37% in winter, 60.94% in spring and 27.77% in summer but from 0.2 to 0.3 m–1 (Figs 8 and 9a). The area percent of waters with seasonal aCDOM(400) higher than 0.3 m–1 was 42.66% in autumn, 24.16% in winter, 16.90% in spring and 43.49% in summer, respectively (Fig. 8).
In terms of space, satellite-derived seasonal aCDOM(400) was low for off shore waters with high salinities (Fig. 8). Along Section P (Fig. 1b), seasonal aCDOM(400) was intertwined in near shore waters, with the maximum in summer and minimum in spring; seasonal aCDOM(400) was close to each other in off shore waters east of 114.0°E (Fig. 9b). The Lingding Bay was the major mixing zone of riverine and saline waters. Seasonal aCDOM(400) was higher in the west part of the Lingding Bay, especially in spring and summer when with higher discharge from the Zhujiang River (Table 3, Fig. 8). However, this distribution pattern was not very obvious in winter and autumn when with low discharge. In the north-south direction, seasonal aCDOM(400) was higher in the Lingding Bay than that in the south of 22°N. In the south of 22°N, most high values of seasonal aCDOM(400) happened in the west near shore waters which were mixed by fresh water from the Zhujiang River (see Figs 1 and 8). Along Section P (Fig. 1b), seasonal aCDOM(400) exponentially decreased from near shore to off shore waters with increasing longitude and salinity (Fig. 8). Moreover, decreases of seasonal aCDOM(400) in all seasons could be fitted with exponential equations (y=aebx, b<0), with R2 of 0.943 1 in autumn, 0.923 1 in winter, 0.753 8 in spring, and 0.971 4 in summer, respectively.
5.3 The effective DOC flux into the SCS by the Zhujiang River
Waters in Regions A and B were with different salinities (Fig. 1b). On the basis of the developed CDOM algorithm (Eq. (4)) and the CDOM conservative mixing equation (Fig. 4a), not only the mean aCDOM(400) in Regions A and B were estimated from the MODIS/Aqua data, but also the mean salinities. Using Eq. (8), mean DOC concentrations in Regions A and B were also calculated from the MODIS/Aqua data. Following the calculation processes (Fig. 3), effective riverine end-member DOC concentrations and fluxes in winter and summer from 2002 to 2014 were calculated (Fig. 10).
In summer from 2002 to 2014, mean Ce was 1.59 mg/L, with minimum of 1.42 mg/L in 2007 and maximum of 1.9 mg/L in 2006. Mean Fe was 2.05×105 t, with minimum of 0.98×105 t in 2011 and maximum 3.3×105 of in 2006 (Fig. 10a). In summer, Ce was positively linearly related to discharge (R2=0.698) except for 2008, when riverine DOC concentration might be diluted by discharge as high as 18.09×1010 m3. It was the same to Fe, with R2 of 0.965 7 (Fig. 10a). In winter except for 2004, 2005 and 2006, mean Ce was 2.64 mg/L, with minimum of 2.02 mg/L in 2008 and maximum of 3.62 mg/L in 2009. Mean Fe was 0.81×105 t, with minimum of 0.46×105 t in 2012 and maximum of 1.35×105 t in 2003 (Fig. 10b). Except for 2004–2006, Ce and Fe were also positively related to discharge in winter, but more weakly than those in summer. With low water discharge of the Zhujiang River from 2004 to 2006, photosynthetic active radiation in Region A might increase (Section 1). Part DOC in Region A might be produced by in situ phytoplankton. As results, the values of Ce in these years might be overestimated.
6 Discussion
Less
6.1 Sources and spectral slope of CDOM
First, massively terrigenous CDOM was discharged into estuary along with riverine fresh water (Ni et al., 2008; Wang et al., 2012). The greatly linear relationship between aCDOM(400) and the salinity (Fig. 4a) indicated the great contribution of terrigenous CDOM in the ZJE (Hong et al., 2005). Secondly, terrestrial nutrient loadings could fuel a high phytoplankton production in estuarine regions (Callahan et al., 2004); and phytoplankton degradation could further produce CDOM. However, there was no correlation between in situ aCDOM(400) and Chl a, so the CDOM contribution by phytoplankton should be limited in the ZJE. Thirdly, microbial community can produce CDOM by transforming non-colored DOM to CDOM. The high fluorescence signature of low salinity waters in the ZJE, reported by Callahan et al. (2004), might be attributed to microbial transformation of non-fluorescent DOM to fluorescent CDOM. In addition, marsh and wetland within the estuarine system can also produce CDOM. However, with little marsh and wetland in the highly engineered estuary, CDOM from these sources was assumed to be small in the ZJE (Callahan et al., 2004). Judging from the conservative mixing (Fig. 4a), CDOM sourced from phytoplankton, microbial communities, and wetlands, etc, just balanced degradation of riverine CDOM; and fresh water was the primary source of CDOM in the ZJE.
Equation (7) was applied to the daily satellite water reflectance of MODIS/Aqua to calculate daily SCDOM from 2002 to 2014 (Section 4.2). Then, monthly satellite climatology SCDOM was calculated by averaging daily SCDOM of a specific month. The results are shown as Fig. 11. SCDOM was reported negative correlation with the CDOM concentration in coastal waters of the Bohai Sea (Xing et al., (2008) and references therein). However, it was not the case in the ZJE, where SCDOM decreased significantly along Section P with decreasing aCDOM(400) (Figs 1b, 9b and 11). There was a positive correlation between SCDOM and aCDOM(400) in the ZJE, with R2=0.689 2. Fulvic acid and humic acid are the primary components of CDOM. The spectral slope of fulvic acid was higher than that of humic acid (Carder et al., 1989). Therefore, SCDOM can serve as an indicator for the fulvic acid fraction of CDOM (Carder et al., 1989). For the Mississippi Plume, Carder et al. (1989) reported a logarithmic relation between SCDOM and the fulvic acid fraction ([fulvic acid fraction]=aln(SCDOM)+b). For the ZJE, fulvic acid fraction decreased with decreasing SCDOM and increasing salinity in both winter and summer (Fig. 11). SCDOM and fulvic acid fraction were high in near shore waters with low salinities. It might be due to high fulvic acid fraction of terrigeous CDOM, flushed from soils with richly fulvic acid (Yu and Bai, 1962). For the near shore waters in winter and summer, the great discrepancy of fulvic acid fraction might be attributed to the fact that more fulvic acid was degraded with the longer residence time in winter (Sun et al., 2014). The remaining CDOM was persistent, so the fulvic acid fraction decreased slightly in the waters with high salinities (Fig. 10).
6.2 Factors influencing CDOM distribution in the ZJE
Many physical-biogeochemical processes, such as water advection, photo-bleaching, bacterial utilization (Amon and Benner, 1996; Hong et al., 2005), phytoplankton growth, etc., can impact the CDOM distribution in estuarine waters. CDOM was majorly sourced from fresh water and was conservatively mixed in the ZJE (Fig. 4a). Moreover, before entering into the SCS, the average residence time of fresh water in the Lingding Bay was short, with about 6 days during the dry season and 3 days in the wet season (Sun et al., 2014). Therefore, river discharge and factors that influence the fresh water diffusion should be the major impact factors on the CDOM distribution in the ZJE.
The discharge of the Zhujiang River, high in summer (wet season) and low in winter (dry season) (Callahan et al., 2004; Chen et al., 2004; Liu et al., 2015), has significant influence on the distribution pattern of aCDOM(400) in the ZJE (Section 5.2). With low discharge and strongly northeast wind during the dry season, the Zhujiang River plume is advected westward by Guangdong coastal current and tidal flood current. Small area with aCDOM(400) higher than 0.3 m–1 (Figs 8 and 9b) was synchronous with the much small plume area in winter (Dong et al., 2004). At high discharge, however, fresh water could diffuse far away. For the peak discharge season of the Changjiang River, the plume area was greatly positively correlated with the water discharge (Bai et al., 2014). With high discharge and eastward Guangdong coastal current under strongly southwest wind, the Zhujiang River plume could intrude eastward even into the Penghu Channel during the wet season (Dong et al., 2004; Bai et al., 2015). The area of low salinity waters with aCDOM(400) larger than 0.3 m–1 in summer (high discharge season) was apparently larger than that in winter (Figs 8 and 9b). Moreover, there was a significantly positive correlation between the discharge of the Zhujiang River and the area percent of waters with high aCDOM(400) in the ZJE, especially for waters with aCDOM(400) larger than 0.25 m–1 in summer (Fig. 12). However, due to strongly vertical mixing, low discharge from the Zhujiang River, and long residence in winter, no obvious relation among them was found.
Besides river discharge, an estuarine underwater topography was another important factor impacting aCDOM(400) distribution in the ZJE. Deep water channels were located in the east and fresh water was discharged into the west (Wang et al., 2013; Liu et al., 2015). When the tide rising, saline waters intruded into the Lingding Bay from the east channels and were advected by fresh water to form a counter-clockwise current inside the Lingding Bay (Chen et al., 2004). Consequently, the CDOM concentration was high in the west of the Lingding Bay, especially in summer when with high water discharge from the Zhujiang River (Fig. 8). The CDOM diffusion along with fresh water was also related to water currents in the northern SCS. Outside the Lingding Bay, surface water flowed southwest along the Guangdong seacoast in winter, but northeast in summer (Dong et al., 2004; Wong et al., 2007; Bai et al., 2015). Therefore, part of high aCDOM(400) waters were transported eastward in summer, but westward in winter (Figs 1 and 8).
6.3 Relationship between CDOM and DOC
Terrigenous CDOM, although resistant to biodegradation, might be degraded to non-colored DOM or CO2 after absorbing light in the estuary (Hernes and Benner, 2003; Hong et al., 2005; Bauer et al., 2013). Moreover, CDOM might also be decomposed through biological activities such as bacterial utilization (Amon and Benner, 1996; Hong et al., 2005), micro-zooplankton ZJEdation (Mannino et al., 2008), and virus hydrolysis (Pan et al., 2012), etc. Judging from the conservative CDOM mixing (Fig. 4a), however, these processes were not significant in the ZJE. Except for CDOM, other non-colored DOM might be labile and therefore had a short residence time (Chen et al., 2004). Furthermore, DOM from other sources might have a different non-colored DOM fraction with that from the Zhujiang River. All of these together shaped the correlation between CDOM and DOC in the ZJE.
Figure 13 shows the scatter plot of aCDOM(400) and the DOC concentration in the ZJE. In summer with larger water discharge, fresh water stayed in the ZJE for shorter time. DOC co-varied with aCDOM(400), with R2 of 0.866 6. For other three seasons, waters of some stations with low salinities had significantly high DOC concentrations (Fig. 13). There were two possible reasons for these deviations. The first one was the input of anthropogenic DOC. Several Chinese metropolises are located around the ZJE, such as the Guangzhou, Shenzhen. Industrial wastewater and domestic sewage with the high DOC concentration might be discharged into the ZJE without removing DOM well. These DOM, mainly protein, were biodegradable and easily degraded soon after being poured into the ZJE. The other one was the transformation from POC to DOC in low salinity waters. The Zhujiang River affected by the East Asian Monsoon was featured by high ratio of POC to DOC (Ni et al., 2008). During the early stage of estuarine mixing, macro-particulate might be transformed into DOM so as to increase the DOC concentration (Dai et al., 2000). These increases not only collapsed the conservative mixing of DOC, but also complicated the estimation of the effective riverine end-member DOC concentration. These might also account for the abnormally high Ce in the winter of 2004, 2005 and 2006 (Fig. 10b).
7 Conclusions and prospects
Less
The estuarine DOM distribution and characteristic are comprehensive functions of various impact factors. A series of physical-biogeochemical processes can influence the estuarine DOM dynamics. For the ZJE, high variation and great importance to water environment make it is necessary to monitor estuarine DOM in high tempo-spatial resolution. Satellite remote sensing is a good choice. This study developed remote sensing algorithms to inverse the CDOM concentration (determined by aCDOM(400)) and its spectral slope (SCDOM) using the in situ Rrs data from four seasonal cruises in the ZJE. Then, the new developed algorithms were applied on the water reflectance of MODIS/Aqua from 2002 to 2014 to calculate seasonal aCDOM(400) and SCDOM in the ZJE.
In the ZJE, CDOM was primarily sourced from fresh water of the Zhujiang River. These terrigenous CDOM was rich in fulvic acid, which accounted for the unexpectedly high fulvic acid fraction of CDOM in the low salinity waters. The CDOM distributions were majorly influenced by water discharge of the Zhujiang River and the underwater topography influencing water mixing. CDOM was conservatively mixed and linearly decreased with increasing salinity in the ZJE. Along Section P vertical to water depth gradient, satellite-derived seasonal aCDOM(400) could be described by decaying exponential functions shaped as y=aebx (b<0), with minimum R2 of 0.753 8 in spring and maximum R2 of 0.971 4 in summer. Combining with water discharge measured at hydrometric stations of the Zhujiang River in summer and winter from 2002 to 2014, the effective riverine end-member DOC concentration and flux into the SCS were first estimated from the MODIS/Aqua data. In summer with great discharge, both effective riverine end-member DOC concentration and flux had significantly positive correlations with the discharge of the Zhujiang River. However, they might be overestimated due to DOM sourced from human activities and POC in winter, especially in years with abnormally low river discharge.
On the basis of conservative mixing of CDOM, the developed CDOM algorithm can further be applied to monitor salinity distribution, the fresh water diffusion, the mass transportation, etc. in the ZJE. However, the conservative mixing of DOC was disturbed by in situ produced DOC in the low salinity waters, significantly in autumn and spring. After figuring out these DOC additions, the satellite estimation of the effective riverine end-member DOC concentration and flux may also be realized in these seasons. Moreover, the developed algorithms employed commonly used bands of ocean color sensors, such as the medium-resolution imaging spectrometer (MERIS), the geostationary ocean color imager (GOCI), visible infrared imaging radiometer suite (VIIRS), etc. Therefore, using global ocean color satellite data they can also be used to monitor DOM dynamics in other estuaries. The developed algorithms can be applied to explore DOM compositions, distributions, and dynamic changes, etc. in highly turbid estuaries.
Acknowledgements
Less
The authors thank the Level 1 and Atmosphere Archive and Distribution System for providing us with the daily level 1 B calibrated radiance products of MODIS/Aqua (MYD021KM, Version 2.16). We are particularly thankful to the crews of the four seasonal cruises conducted in the Zhujiang (Pearl River) Estuary for their help of field sampling.
Funding
Less
  • The National Key Research and Development Progam of China under contract No.2017YFA0603003; the National Basic Research Program (973 Program) of China under contract No. 2015CB954002; the Public Science and Technology Research Funds Project of Ocean under contract No. 201505003; the National Natural Science Foundation of China under contract Nos 41676170, 41676172, 41476155, 41621064 and 41406202; the Project of State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, State Oceanic Administration of China under contract No. SOEDZZ1801; the Research Startup Project of Nanjing Instiute of Geography and Limnology, Chinese Academy of Sciences under contract No. Y7SL051001.
References
Less
Amon R M W, Benner R. 1996. Bacterial utilization of different size classes of dissolved organic matter. Limnology and Oceanography, 41(1): 41–51
Arenz R F Jr, Lewis W M Jr, Saunders J F. 1996. Determination of chlorophyll and dissolved organic carbon from reflectance data for Colorado reservoirs. International Journal of Remote Sensing, 17(8): 1547–1566
Bai Yan, Pan Delu, Cai Weijun, et al. 2013. Remote sensing of salinity from satellite-derived CDOM in the Changjiang River dominated East China Sea. Journal of Geophysical Research: Oceans, 118(1): 227–243
Bai Yan, He Xianqiang, Pan D, et al. 2014. Summertime Changjiang River plume variation during 1998–2010. Journal of Geophysical Research: Oceans, 119(9): 6238–6257
Bai Yan, Huang T H, He Xianqiang, et al. 2015. Intrusion of the Pearl River plume into the main channel of the Taiwan Strait in summer. Journal of Sea Research, 95: 1–15
Bailey S W, Werdell P J. 2006. A multi-sensor approach for the on-orbit validation of ocean color satellite data products. Remote Sensing of Environment, 102(1–2): 12–23
Bauer J E, Cai W J, Raymond P A, et al. 2013. The changing carbon cycle of the coastal ocean. Nature, 504(7478): 61–70
Belanger S, Babin M, Larouche P. 2008. An empirical ocean color algorithm for estimating the contribution of chromophoric dissolved organic matter to total light absorption in optically complex waters. Journal of Geophysical Research: Oceans, 113(C4): C04027
Cai Wei-Jun, Dai Minhan, Wang Yongchen, et al. 2004. The biogeochemistry of inorganic carbon and nutrients in the Pearl River estuary and the adjacent northern South China Sea. Continental Shelf Research, 24(12): 1301–1319
Cai Wei-Jun, Guo Xianghui, Chen C T A, et al. 2008. A comparative overview of weathering intensity and HCO3- flux in the world’s major rivers with emphasis on the Changjiang, Huanghe, Zhujiang (Pearl) and Mississippi Rivers. Continental Shelf Research, 28(12): 1538–1549
Callahan J, Dai Minhan, Chen R F, et al. 2004. Distribution of dissolved organic matter in the Pearl River Estuary, China. Marine Chemistry, 89(1–4): 211–224
Carder K L, Chen F R, Lee Zhongping, et al. 2003. MODIS Ocean Science Team-Algorithm Theoretical Basis Document. ATBD 19, Case 2 Chlorophyll a, Version 7. NASA MODIS
Carder K L, Steward R G, Harvey G R, et al. 1989. Marine humic and fulvic acids: Their effects on remote sensing of ocean chlorophyll. Limnology and Oceanography, 34(1): 68–81
Chen Chuqun, Shi Ping, Zhan Haigang. 2003. A local algorithm for estimation of yellow substance (gelbstoff) in coastal waters from SeaWiFS data: Pearl River estuary, China. International Journal of Remote Sensing, 24(5): 1171–1176
Chen Hao, Meng Wei, Zheng Bing-Hui, et al. 2013. Optical signatures of dissolved organic matter in the watershed of a globally large river (Yangtze River, China). Limnologica-Ecology and Management of Inland Waters, 43(6): 482–491
Chen Zhiqiang, Li Yan, Pan Jianming. 2004. Distributions of colored dissolved organic matter and dissolved organic carbon in the Pearl River Estuary, China. Continental Shelf Research, 24(16): 1845–1856
Dai Minhan, Martin J M, Hong Huasheng, et al. 2000. ZJEliminary study on the dissolved and colloidal organic carbon in the Zhujiang River estuary. Chinese Journal of Oceanology and Limnology, 18(3): 265–273
Dai Minhan, Yin Zhiqiang, Meng Feifei, et al. 2012. Spatial distribution of riverine DOC input to the ocean: an updated global system. Current Opinion in Environmental Sustainability, 4(2): 170–178
Dong Lixian, Su Jilan, Wong L A, et al. 2004. Seasonal variation and dynamics of the pearl river plume. Continental Shelf Research, 24(16): 1761–1777
Dong Qiang, Shang Shaoling, Lee Zhongping. 2013. An algorithm to retrieve absorption coefficient of chromophoric dissolved organic matter from ocean color. Remote Sensing of Environment, 128: 259–267
D’Sa E J, Miller R L. 2003. Bio-optical properties in waters influenced by the Mississippi River during low flow conditions. Remote Sensing of Environment, 84(4): 538–549
Fang Ligang, Chen Shuisen, Li Dong, et al. 2009. Use of reflectance ratios as a proxy for coastal water constituent monitoring in the pearl river estuary. Sensors, 9(1): 656–673
Gao Q, Tao Z, Shen C, et al. 2002. Riverine organic carbon in the Xijiang River (South China): seasonal variation in content and flux budget. Environmental Geology, 41(7): 826–832
Guo Weidong, Yang Liyang, Zhai Weidong, et al. 2013. Runoff-mediated seasonal oscillation in the dynamics of dissolved organic matter in different branches of a large bifurcated estuary-The Changjiang Estuary. Journal of Geophysical Research: Biogeosciences, 119(5): 776–793
He Biyan, Dai Minhan, Zhai Weridong, et al. 2010. Distribution, degradation and dynamics of dissolved organic carbon and its major compound classes in the Pearl River Estuary, China. Marine Chemistry, 119(1–4): 52–64
He Xianqiang, Bai Yan, Pan Delu, et al. 2013. Using geostationary satellite ocean color data to map the diurnal dynamics of suspended particulate matter in coastal waters. Remote Sensing of Environment, 113: 225–239
Hernes P J, Benner R. 2003. Photochemical and microbial degradation of dissolved lignin phenols: Implications for the fate of terrigenous dissolved organic matter in marine environments. Journal of Geophysical Research: Oceans, 108(C9): 3291
Hong Huasheng, Wu Jingyu, Shang Shaoling, et al. 2005. Absorption and fluorescence of chromophoric dissolved organic matter in the Pearl River Estuary, South China. Marine Chemistry, 97(1–2): 78–89
Huang Wei, Chen R F. 2009. Sources and transformations of chromophoric dissolved organic matter in the Neponset River Watershed. Journal of Geophysical Research: Biogeosciences, 114(G4): G00F05
Jaffé R, Boyer J N, Maie N, et al. 2004. Source characterization of dissolved organic matter in a subtropical mangrove-dominated estuary by fluorescence analysis. Marine Chemistry, 84(3–4): 195–210
Knap A H, Michaels A F, Close A R, et al. 1996. Protocols for the joint global ocean flux study (JGOFS) core measurements. Jgofs.
Lee Zhongping, Carder K L, Arnone R A. 2002. Deriving inherent optical properties from water color: a multiband quasi-analytical algorithm for optically deep waters. Applied Optics, 41(27): 5755–5772
Liu Dong, Pan Delu, Bai, Yan, et al. 2015. Remote Sensing Observation of Particulate Organic Carbon in the Pearl River Estuary. Remote Sensing, 7(7): 8683–8704
Liu Qiong, Pan Delu, Bai Yan, et al. 2014. Estimating dissolved organic carbon inventories in the East China Sea using remote-sensing data. Journal of Geophysical Research: Oceans, 119(10): 6557–6574
Mannino A, Russ M E, Hooker S B. 2008. Algorithm development and validation for satellite-derived distributions of DOC and CDOM in the US Middle Atlantic Bight. Journal of Geophysical Research: Oceans, 113(C7): C07051
Mitchell B G, Kahru M, Wieland J, et al. 2000. Determination of spectral absorption coefficients of particles, dissolved material and phytoplankton for discrete water samples, In: Fargion G S Mueller J, eds. Ocean Optics Protocols for Satellite Ocean Color Sensor validation., Goddard Space Flight Space Center, Greenbelt, Md
Mueller J L, Fargion G S, Mcclain C R. 2003. Biogeochemical and bio-optical measurements and data analysis methods, NASA Tech. Mem., NASA TM-2003-211621, Rev. 4, vol. 2, Goddard Space Flight Space Center, Maryland
Ni Honggang, Lu Feng-Hui, Luo Xian-Lin, et al. 2008. Riverine inputs of total organic carbon and suspended particulate matter from the Pearl River Delta to the coastal ocean off South China. Marine Pollution Bulletin, 56(6): 1150–1157
Pan Delu, Liu Qiong, Bai Yan. 2012. Progress in remote sensing of DOC: based on the analysis of conservative behaviors of DOC and CDOM in global large rivers estuaries. Haiyang Xuebao (in Chinese), 34(4): 1–11
Park Y, Ahn Y, Han H, et al. 2014. GOCI level 2 Ocean Color Products (GDPS 1.3) Brief Algorithm Description. 17 November
Ran Lishan, Lu X X, Sun Huiguo, et al. 2013. Spatial and seasonal variability of organic carbon transport in the Yellow River, China. Journal of Hydrology, 498: 76–88
IOCCG. 2000. Remote sensing of ocean colour in coastal, and other optically-complex, waters. Sathyendranath, S. (ed.), reports of the international ocean-colour coordinating group, No. 3, IOCCG, Dartmouth, Canada.
Siegel D A, Maritorena S, Nelson N B, et al. 2002. Global distribution and dynamics of colored dissolved and detrital organic materials. Journal of Geophysical Research: Oceans, 107(C12): 21-1–21-14
Siswanto E, Tang J, Yamaguchi H, et al. 2011. Empirical ocean-color algorithms to retrieve chlorophyll-α, total suspended matter, and colored dissolved organic matter absorption coefficient in the Yellow and East China Seas. Journal of Oceanography, 67(5): 627–650
Sun Jian, Lin Binliang, Li Kaiming, et al. 2014. A modelling study of residence time and exposure time in the Pearl River Estuary, China. Journal of Hydro-environment Research, 8(3): 281–291
Świrgoń M, Stramska M. 2015. Comparison of in situ and satellite ocean color determinations of particulate organic carbon concentration in the global ocean. Oceanologia, 57(1): 25–31
Tiwari S P, Shanmugam P. 2011. An optical for the remote sensing of coloured dissolved organic matter in coastal/ocean waters. Estuarine, Coastal and Shelf Science, 93(4): 396–402
Wang Menghua, Shi Wei. 2007. The NIR-SWIR combined atmospheric correction approach for MODIS ocean color data processing. Optics ExZJEss, 15(24): 15722–15733
Wang Xuchen, Ma Haiqing, Li Ronghua, et al. 2012. Seasonal fluxes and source variation of organic carbon transported by two major Chinese rivers-The Yellow River and Changjiang (Yangtze) River. Global Biogeochemical Cycles, 26(2): GB2025
Wang Xiaozhen, Zhang Huaguo, Fu Bin, et al. 2013. Analysis on the coastline change and erosion-accretion evolution of the Pearl River Estuary, China, based on remote-sensing images and nautical charts. Journal of Applied Remote Sensing, 7: 073519
Wong G T F, Ku T, Mulholland M, et al. 2007. The South East Asian time-series study (SEATS) and the biogeochemistry of the South China Sea—An overview. Deep-Sea Research II: Topical Studies in Oceanography, 54(14–15): 1434–1447
Wu Y, Zhang J, Liu S M, et al. 2007. Sources and distribution of carbon within the Yangtze River system. Estuarine, Coastal and Shelf Science, 71(1–2): 13–25
Xing Xiaogang, Zhao Dongzhi, Liu Yuguang, et al. 2008. Absorption characteristics of de-pigmented particle and yellow substance in Bohai Sea. Marine Environmental Science (in Chinese), 27(6): 595–598
Yu Mengde, Bai Jinquan. 1962. Effect of vegetation on the composition of humus and active mineral substences in soils of Kwangtung and Kwangsi. Acta Pedologica Sinica (in Chinese), 10(1): 29–43
Zhu Weining, Yu Qian, Tian Y Q, et al. 2011. Estimation of chromophoric dissolved organic matter in the Mississippi and Atchafalaya river plume regions using above-surface hyperspectral remote sensing. Journal of Geophysical Research: Atmospheres, 116(C2): C02011
Appendix
Less
Year 2018 volume 37 Issue 7
PDF
36
19
Cite this Article
BibTeX
Article Info
doi: 10.1007/s13131-017-1248-7
  • Receive Date：2017-06-14
  • Online Date：2026-04-14
  • Published：2018-07-25
Article Data
Affiliations
History
  • Received：2017-06-14
  • Accepted：2017-11-02
Funding
The National Key Research and Development Progam of China under contract No.2017YFA0603003; the National Basic Research Program (973 Program) of China under contract No. 2015CB954002; the Public Science and Technology Research Funds Project of Ocean under contract No. 201505003; the National Natural Science Foundation of China under contract Nos 41676170, 41676172, 41476155, 41621064 and 41406202; the Project of State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, State Oceanic Administration of China under contract No. SOEDZZ1801; the Research Startup Project of Nanjing Instiute of Geography and Limnology, Chinese Academy of Sciences under contract No. Y7SL051001.
Affiliations
    1 State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, State Oceanic Administration, Hangzhou 310012, China
    2 Key Laboratory of Watershed Geographic Sciences, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008, China

Corresponding:

References
Share
https://castjournals.cast.org.cn/joweb/aos/EN/10.1007/s13131-017-1248-7
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数
科
Family		 属数
Number of
genus		 种数
Number of
species		 占总种数比例
Percentage of
total species (%)		 属
Genus		 种数
Number of
species		 占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT
Links
Articles
Latest Articles
Most Read
Collections
Updates
Events
News
Multimedia
About
About Us
Contact
No. 86 Xueyuan South Road, Haidian District, Beijing
100081
010-62199257
qkjq@cast.org.cn

Copyright © 2025 China Association for Science and Technology. All rights reserved. For all open access content, the relevant licensing terms apply.
Sponsored by the Office of the Leading Group for Cybersecurity and Informatization of CAST, and supported by Science and Technology Review Publishing House