Fluorescence measured using a field-portable laser fluorometer as a proxy for CDOM absorption

Estuarine, Coastal and Shelf Science 146 (2014) 33e41

Contents lists available at ScienceDirect

Estuarine, Coastal and Shelf Science journal homepage: www.elsevier.com/locate/ecss

Fluorescence measured using a field-portable laser fluorometer as a proxy for CDOM absorption Peng Chen a, b, Delu Pan b, *, Zhihua Mao b a

State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan 430072, China State Key Laboratory of Satellite Ocean Environment Dynamics, Second Institute of Oceanography, State Oceanic Administration, 36 Bochubeilu, Hangzhou 310012, China

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 15 December 2013 Accepted 15 May 2014 Available online 27 May 2014

The colored fraction of dissolved organic matter, CDOM, directly influences water optical properties and spectral quality, playing an important role in aquatic ecosystems, optical remote sensing and carbon circulation in the sea. Measuring in situ CDOM absorption coefficient, however, is difficult because it requires prefiltration of water samples. A field-portable laser fluorometer for fast acquisition of in situ CDOM fluorescence data without sample filtering was developed. Using the instrument in the Yangtze River estuary, the East China Sea coast, the West Lake, and the Qiandao Lake, good estimates of CDOM absorption were obtained from in situ fluorescence measurements. High correlation (r ¼ 0.94) was observed between CDOM absorption coefficient and fluorescence normalized to Raman scattering in diverse water types. Comparison of different methods for measuring CDOM fluorescence was carried out. Fluorescence measured by in situ laser fluorometer compared very well to the values obtained using more sophisticated laboratory spectrophotofluorometer; a significant relationship (r ¼ 0.95) was identified between the two methods. Spatial distributions of CDOM in the Yangtze River estuary and its adjacent sea compiled from discrete samples analyses were presented. A linear relationship was identified between CDOM fluorescence and salinity in surface water, reflecting the potential of CDOM fluorescence to be used as a passive tracer of freshwater input. The in situ measurement results demonstrated the utility of the laser fluorometer for rapid acquisition of high-resolution profiles, and that the work resulting from sample filtration or storage can be avoided. © 2014 Elsevier Ltd. All rights reserved.

Keywords: lasers fluorescence spectroscopy dissolved organic matter absorption coefficient East China Sea Raman scattering

1. Introduction The colored fraction of dissolved organic matter (CDOM), also known as chromophoric dissolved organic matter, yellow substance or gelbstoff, is an important component in governing light propagation in coastal and open waters (Bricaud et al., 1981). It has a broad application prospect in aquatic ecosystem, optical remote sensing and carbon circulation. The optical properties of CDOM absorption, however, complicate the use of chlorophyll a retrieval algorithms that are based on remotely sensed ocean color (Carder and et al., 1991) and the development of phytoplankton production models (Lee and et al., 1994; Arrigo and Brown, 1996; Keith et al., 2002). Although CDOM strongly influences ocean optical properties, remotely sensed spectra, and biogeochemical processes, dynamics of CDOM in diverse aquatic environments remain largely

* Corresponding author. E-mail address: [email protected] (D. Pan). http://dx.doi.org/10.1016/j.ecss.2014.05.010 0272-7714/© 2014 Elsevier Ltd. All rights reserved.

undefined due partially to a sparse global database of CDOM absorption. Limited availability of such data is related to a lack of highly sensitive, portable optical systems that can provide in situ measurements at sea (Miller and et al., 2002). CDOM is operationally defined as the colored fraction of material passing through a 0.2 um filter and quantified by the absorption coefficient, ag (l), as measured on filtered samples using a spectrophotometer or an absorption meter. In situ determination of ag (l) requires the use of in-line 0.2 um filters fitted to a spectral absorption meter (e.g., WET Labs ac-9). Fouling of these filters prevents their use over an extended period of time, and therefore complicates the acquisition of ag (l) time-series from moorings (Belzile and et al., 2006). The absorption of CDOM may be measured directly using a spectrophotometer or may be inferred from fluorescence measurements (Ferrari and Dowell, 1998). Relative to absorption measurements, fluorescence measurements can be acquired more rapidly with greater sensitivity. The establishment of robust relationships between the level of CDOM fluorescence and absorption may provide a more rapid determination of ag (l). Once

34

P. Chen et al. / Estuarine, Coastal and Shelf Science 146 (2014) 33e41

Fig. 1. Picture of the laser fluorometer.

the reference absorption coefficient ag (l0) is estimated from fluorescence, its spectral ag (l) can be determined by assuming an exponential shape of the form (Bricaud et al., 1981; Twardowski and et al., 2004):


ag ðlÞ ¼ ag ðl0 Þexp  Sg ðl  l0 Þ

(1)

where ag (l0) is the magnitude of absorption at reference wavelength l0, and Sg is the slope parameter describing the wavelength dependence. It has been found that CDOM fluorescence of samples from a wide range of marine, riverine and estuarine waters is linearly related to ag (337) or ag (355), and that the use of a single equation derived from the ratio of fluorescence to absorption at all these sites would incur an error no worse than 100e150% in the retrieval of ag (l) at 337 or 355 nm (Ferrari and Tassan, 1991; Hoge et al., 1993; Green and Blough, 1994; Belzile and et al., 2006). The total absorption coefficient, atot(l), includes the coefficients for total particulate matter of phytoplankton and non-algal particulates, ap (l), ag (l) and pure water aw(l). Because atot(l) can be obtained by field spectral absorption meter(e.g., WET Labs ac-9), if ag (l) also can be derived through in situ fluorescence measurements, ap(l) can be obtained by the expression ap (l) ¼ atot (l)  ag (l)  aw (l). In the past two decades, fluorescence spectroscopy techniques, especially the excitationeemission matrices (EEMs), have been fruitfully used to investigate the composition and dynamics of dissolved organic matter (DOM) in aquatic environments (Coble, 1996; Stedmon et al., 2003; Chen and et al., 2004; Spencer and et al., 2007; Jiang and et al., 2008; Kowalczuk and et al., 2009; Yamashita and et al., 2010). However, current laboratory fluorescence spectrofluorometers (e.g., Shimadzu RF5301 spectrofluorometer, Shimadzu Inc.) (Zhang and et al., 2007; Foden and et al., 2008; Singh et al., 2010) are too bulky for routine use in the field and the measurement scans takes a long time. In recent years, portable or submersible field fluorometers have been employed to acquire rapid, real-time, high-frequency measurements of DOM in aquatic environments (Sivaprakasam and Killinger, 2003; Conmy et al., 2004; Baker and et al., 2004; Killinger and Sivaprakasam, 2006; Suping1a et al., 2010; Tedetti et al., 2010; Chekalyuk and Hafez, 2013, 2008; Tedetti et al., 2013). For instance, the WETStar and ECO Puck submersible fluorometers (WET Labs Inc., USA) use a near-UV light-emitting diode (LED) to measure a fulvic acid-like fluorophore at lEx/lEm of 370/460 nm (Niewiadomska and et al., 2008). The weakness of such an instrument is that most of these field instruments are designed to measure one specific parameter (e.g., Chl-a, CDOM, oil, or variable fluorescence) and do not provide

full and detailed spectral information about other fluorescent constituents for more comprehensive characterization of aquatic environments (Chekalyuk and Hafez, 2008, 2013). This paper presents an original field-portable laser fluorometer (FLF) for rapid, high-resolution fluorescence measurements on unfiltered samples, and for estimating absorptions through fluorescence-absorption relationship. The estimated absorptions inferred from fluorescence measured by the FLF in various aquatic ecosystems are compared to ag(l) measured by ultravioletevisible spectrophotometry on filtered samples. CDOM fluorescence obtained by the FLF is also compared to values measured using more sophisticated scanning spectrofluorometers. Application of the FLF in biogeochemical studies on organic matter is illustrated by in situ profiles obtained in the Yangtze River estuary and its adjacent sea. 2. Materials and methods 2.1. Mechanical and optical configurations of the FLF The FLF is built on a porous aluminium alloy plate with orthogonal Czerny-Turner optical configuration (Fig. 1). A diagram of the FLF is presented in Fig. 2. The laser fluorescence system consists of an excitation source based on a micro blue-violet laser, a measurement cell (MC), collimating lens, two reflecting mirrors, a

Fig. 2. Block diagram of the FLF instrument. MC e measurement cell; M1 and M2 e reflecting mirrors; F1 e long pass filter; L1 and L3e focusing lenses; L2 and L4e collecting lenses.

P. Chen et al. / Estuarine, Coastal and Shelf Science 146 (2014) 33e41

35

Table 1 Main characteristics of the FLF. Optical element

Product model

Product parameter

Transmitter

MM-405-100 solid-state laser (Boson Tech., China) Output average power (mw) Pulse duration (ns) Beam diameter at the aperture (mm) pulsed repetition frequency rate (kHz) USB4000-FL (Ocean Optics Inc., USA) QP600-2-UV-VIS (Ocean Optics Inc., USA) 74-UV (Ocean Optics Inc., USA) SPL-LP-450 (Pu Lei Inc. ., China) INLINE-FH (Ocean Optics Inc., USA) SPLFT-25.4 (Pu Lei Inc., China) GCM-3015M (Da Heng Inc., China)

l ¼ 405 nm 100 ~7 ~2.0 10 360e1000 nm 600 um Premium Fiber, UV/VIS 1-cm focal length, 200e2000 nm l > 450 nm holds filters diameter up to 8 mm R > 0.9 Size: 300  450  12.7 mm

CCD detector Fiber optic Collimating lens Long pass filter In-line filter holder Reflecting mirrors Aluminum alloy porous fixed plate

long pass filter, and a spectroscopy-detecting module. During the measurement, the output of the laser is focused into a quartz sample cell and the emitted fluorescence light is collected through collection quartz lenses and transfers onto the entrance slits of a CCD spectrometer. The CCD detector is interfaced to a personal computer via a USB 2.0 port. A microchip diode-pumped solid-state laser (MM-405-100, Boson Tech) with high-pulsed repetition frequency rate greater than 10 kHz is selected as the excitation source, which emits light at the wavelength of 405 nm. The signal noise ratio (SNR) of fluorescence spectroscopy instrument can be increased by using high pulsed repetition frequency laser (Killinger and Sivaprakasam, 2006). A power supply module (PSU-H-FDA, Boson Tech) is used to provide stable power and control temperature of the laser source. The sample pool is made from quartz, of the size 12.5 mm  12.5 mm  45 mm. The length of the internal optical path of sample space is 10 mm. Its four sides are polished and transparent. The output of the laser is focused into one side of the quartz cell and the emitted light is collected from an orthogonal side, to decrease the influence of scattering interference. Two reflecting mirrors, M1 and M2 (74-msp, Ocean Optics), can reflect fluorescence beams from the other two-pass sides to increase fluorescence signals. A long pass filter (SPL-LP-450, Pu Lei Inc.) is employed for eliminating scattering interference. The fluorescence light is collected with a high-sensitivity hyper-spectral fluorescence spectrometer (USB4000-FL, Ocean Optics), preconfigured with an L4 Detector Collection Lens to increase light-collection efficiency and reduce stray light. The spectrometer covers the wavelength range of 360e1000 nm, with high spectral resolution of 0.2 nm. The laser fluorometer system provides both discrete sample analysis and continuous underway shipboard measurements. During measurements, the pump sucks water into a surge flask from the ocean surface; the water flows into the sample cell by a peristaltic pump. After measurements, the water exits through the waste bottle. The SpectraSuite software (Ocean Optics software Inc.) is used to record the fluorescence spectra. It gives you complete control of setting the parameters for all system functions such as acquiring data, designing the graph display, and using spectra overlays. In addition, it provides both graphical and numeric representation of spectra and can operate on a Windows, Linux, or Macintosh operating system. Signal-processing functions such as electrical dark-signal correction, stray light correction, boxcar pixel smoothing, signal averaging and peak finding are also included. During measurements, an operator can preset the fluorescence spectral integration time (typically, 100 ms to 3 s), boxcar width, their repetition rate and the number of acquisitions (e.g., 5e25) to average the fluorescence induction and to optimize the sampled parameters. The spectral data are displayed and analyzed in real time, and stored on the computer along with the screen captures useful for documentation and data analysis. Each measurement cycle includes two

steps: measuring ambient and background noises (BN) with the laser beam blocked and then measuring laser stimulated spectra of the samples, BN is then subtracted from laser stimulated spectra during each measurement. Water Raman scattering is used to correct laser fluorometer data for normalization of the laser power and effects of water optical attenuation. (Bristow et al., 1981; Exton and et al., 1983; Babichenko and et al., 1993; Determann and et al., 1994). The main characteristics of the FLF are summarized in Table 1. 2.2. Study area Sample data from four geographical regions are presented here: (1) the East China Sea (ECS) coast; (2) the Yangtze River estuary; (3) the Qiandao Lake; and (4) the West Lake. A total of 17 discrete samples were taken from the offshore surface water in the ECS and seven water samples were obtained from the Yangtze River estuary in April 2013 (Fig. 3). The water samples from the West Lake (N ¼ 25) were collected in December 2012 nearby a central site at 30.246 N, 120.146 E. Six other sites were sampled from the Qiandao Lake in May 2013, located in the area between 29.560 N, 118.923 E and 29.615 N, 119.018 E. All together, the data set includes 55 pairs of CDOM absorption and fluorescence measurements. The absorption coefficients of the samples were analyzed in laboratory by spectrophotometer after the cruises, approximately three days after filtering. The fluorescence data was obtained by both FLF and laboratory scanning spectrofluorometer. PH, temperature, turbidity, and salinity were also measured by a multiparameter water quality monitor (Manta, Eureka Inc.) simultaneously during cruises. 2.3. Water sampling and filtration Water samples were collected from surface using a 1.5-L Plexiglas hydrophore, storaged in 1-L ambered glass bottles and immediately transported to the filtering platform on the shipboard for filtration. The hydrophore and ambered bottles were first rinsed copiously with in situ sea water before collecting samples. Prior to filtration, all lab ware and storage bottles were acid-cleaned with 10% HCl and rinsed copiously with Milli-Q water. These samples were then filtered through 0.2-micron nucleopore membrane filters into 60-ml ambered glass bottles (calcination under 450  C for 5 h in advance), and was kept refrigerated in these pre-combusted amber-colored glass bottles for latter optical absorption analyses. 2.4. CDOM absorption measurements Before measuring, the frozen samples were warmed in dark environment until all the samples returned to room temperature. A PerkinElmer Lambada35 dual beam spectrophotometer

36

P. Chen et al. / Estuarine, Coastal and Shelf Science 146 (2014) 33e41

Fig. 3. Distribution of stations measured in April 2013: (1) Yangtze River estuary, and (2) ECS coast.

(PerkinElmer Lambada35, Shimadzu Inc.) was used to measure the absorbance spectra. The Lambda35 was scanned over the wavelength range 250e900 nm, with a spectral resolution of 1 nm. Prior to measurement, the spectrophotometer was preheated above for more than 30 min. The samples were measured in 10 cm quartz cells and baseline corrected with Milli-Q water as the blank to eliminate any temperature effects, because absorption by water itself is temperature-dependent, particularly in the near-infrared (Pegau and Zaneveld, 1993; Stedmon et al., 2000). The baseline stability was verified every five samples by running blank samples, and baseline spectrum curve should be flat in the range from 400 nm to 700 nm, within ±0.0005 v units (absorbance). Absorbance was then converted into absorption coefficient (m1) using the expression ag(l) ¼ 2$3  A(l)/L (Stedmon and Markager, 2001), where A(l) is the absorbance (optical density) and L ¼ 0.1 m is the quartz cells path length in meters. 2.5. CDOM fluorescence measurements by scanning spectrofluorometer Emission spectra of the water samples in the West Lake were measured using a Shimadzu RF-5301 spectrofluorometer (Shimadzu Inc.). Prior to measurements, the instrument was warmed up for 15e30 min. We then opened the software and checked the parameter settings (e.g., saving path, spectral number, scanning speed, excitation wavelength, emission wavelength). The samples were measured in a 1-cm quartz cell and the instrument was scanned over the range 300e900 nm, with a spectral resolution of 1 nm. A Milli-Q water blank correction was applied to eliminate water Raman peaks. 3. Results 3.1. Fluorescence measured by the FLF The FLF system can obtain in situ fluorescence measurements in vavious aquatic environments. An example of fluorescence

Fig. 4. Fluorescence measured by the FLF: (A) An example of water fluorescence emission spectrum by the FLF; (B) spatial distribution of CDOM fluorescence (RU) measured by the FLF in the East China Sea; (C) baseline drift of fluorescence peak intensity at 470 nm: left panel-baseline drift per minute (%); and right panel e baseline drift per hour (%); (D) left panel e fluorescence spectra of quinine sulfate at concentrations from 3 e 15 ug/L; right panel e Calibration curve of quinine sulfate.

P. Chen et al. / Estuarine, Coastal and Shelf Science 146 (2014) 33e41

37

Fig. 4. (continued).

emission spectrum measured by the FLF is presented in Fig. 4A. The emission spectra with the blue-violet excitation at 405 nm can be regarded as a total spectral integrated by the water Raman band, the CDOM constituent band and the different pigments bands. It revealed the capability of the FLF for measuring a variety of water quality parameters. CDOM fluorescence was normalized in water Raman units (RU) in this study. Spatial distribution of CDOM fluorescence measured by the FLF in the ECS is shown in Fig. 4B. We can see the gradual downward trends of CDOM fluorescence magnitudes (RU), as water flows down the Yangtze River estuary into the ECS coast. In order to verify the performance of the laser instrument, some relevant experiments of baseline drift were studied. The baseline drift of fluorescence peak intensity at 470 nm per minute and per hour (%) are presented in Fig. 4C, they are respectively 0.15%/minute and 0.5%/hour. It reveals good stability of the laser fluorescence system. The baseline drift may be affected by laser instability, temperature fluctuations and other factors. To quantify measurement values of the instrument, standard stock solutions of a fluorescence standard reference material (quinine sulfate) (Velapoldi and Mielenz, 1980) in sulfuric acid were prepared with approximate concentrations: 3, 6, 9, 12 and 15 ug/L. Replicates measurements were taken to create calibration curves. Calibration curve of quinine sulfate at a series of concentrations and corresponding fluorescence spectra of quinine sulfate are presented in Fig. 4D. There was a good linearity on the calibration curve for the small concentrations of quinine sulfate and the R2 value for linearity of the curve was 0.994. Calibration curve of standard CDOM fluorescence samples can be used for fluorescence

quantification and for comparison with other fluorescence instruments (Eaton, 1988). 3.2. FLF CDOM fluorescence vs. ag (440) In this study, ag (440) was chosen as an index for CDOM abundance, because the maximum absorption of phytoplankton

Fig. 5. Correlation between FLF CDOM fluorescence and ag (440) from all data.

38

P. Chen et al. / Estuarine, Coastal and Shelf Science 146 (2014) 33e41

pigment can be found at the wavelength of 440 nm, which can then be used in remote sensing inversion of phytoplankton. ag (440) varies in different aquatic environments, ranging from 0.10 to 0.37 m1 (average 0.19 m1) in the ECS coast, from 0.55 to 0.64 m1 (average 0.59 m1) in the Yangtze River estuary, from 0.18 to 0.25 m1 (average 0.21 m1) in the Qiandao Lake, and from 0.19 to 0.49 m1 (average 0.39 m1) in the West Lake (Fig. 5). The CDOM fluorescence measured using the FLF also varies widely, with fluorescence intensity (FL) normalized in water Raman units ranging from 0.40 to 0.96 RU (average 0.71 RU) in the coastal waters of the ECS, from 1.05 to 1.18 RU (average 1.10 RU) in the Yangtze River estuary, from 0.52 to 0.71 RU (average 0.61 RU) in the Qiandao Lake, and from 0.17 to 0.25 RU (average ¼ 0.71 RU) in the West Lake. Despite the wide diversity in aquatic environment types, there are still strong correlations between FL and ag (440) in all samples (Fig. 5; FL ¼ 1.088 ag (440) þ 0.466; r ¼ 0.940, N ¼ 55). The overall average ag (440) of all samples from the four study regions is 0.37 m1 and varies from 0.10 to 0.64 m1; the overall average of FL is 0.82 RU, and varies from 0.40 to 1.18 RU. The root mean square error (RMSE) of fitting curves between model inversion and measured fluorescence is 0.06 RU, and the mean absolute deviation is 0.04 RU. 3.3. FLF measurements vs. spectrofluorometer measurements For the West Lake samples, a strong correlation is found between FLF fluorescence measurements and laboratory spectrofluorometer measurements (Fig. 6; r ¼ 0.950, N ¼ 25). The average FLF fluorescence of all 25 samples is 0.71 RU and varies from 0.17 to 0.25 RU; the average fluorescence measured by the Shimadzu RF5301PC spectrofluorometer is 0.48 RU and varies from 0.36 to 0.53 RU. The different fluorescence intensity by the two instruments was due to different spectral response of detectors. The significant linear correlation suggests the FLF should work as well as more sophisticated laboratory spectrophotofluorometer. 3.4. CDOM optical properties in the Yangtze River estuary and the ECS coast The optical properties of CDOM in the Yangtze River estuary are different from those in the ECS coast according to this study. The CDOM absorption and fluorescence values in the ECS coast, with ag (440) in the range from 0.10 to 0.4 m1 and CDOM-FL in the range

from 0.5 to 1.0 RU, are both lower than those observed in the Yangtze River estuary, where ag (440) is in the range from 0.5 to 0.7 m1 and CDOM-FL is in the range from 1.0 to 1.2 RU (Fig. 7). The absorption coefficients measured in the ECS coast are mainly in the range from 0.1 to 0.2 m1 (nearly 65% of all the ECS coast samples is in this range), compared to the values in the range of 0.5e0.6 m1 measured in the Yangtze River estuary (nearly 86% in this range). The CDOM fluorescence measured in the ECS coast is mainly in the range from 0.5 to 0.7 RU (nearly 59% of the ECS coast samples is in this range), compared to the values with a uniform distribution in the range from 1.0 to 1.2 RU measured in the Yangtze River estuary. Distributions of CDOM-FL and ag (440) reveal similar gradual downward trends of magnitudes, as water flowed down the Yangtze River estuary into the ECS coast. A dilution process was detected as water flowed down the Yangtze River and into the ECS coast. The higher absorption coefficients and CDOM-FL from the cruise data in the estuary were likely greatly influenced by the terrestrial sources; the lower absorption coefficients and CDOM-FL appeared in the offshore area were likely greatly influenced by the mixing of terrestrial humic sources and sea water. The in situ FLF fluorescence versus salinity and ag (440) versus salinity in the surface water along the transect from the inner Yangtze River estuary to the ECS coast are shown in Fig. 8A. The salinity varied from zero to 29.2 as water flowed down the estuary into the coast. Both the magnitudes of ag (440) and CDOM-FL in zero-salinity region are nearly two times larger than the values in the higher salinity region (salinity >25). In the zero-salinity region, both the ag (440) versus salinity plot and CDOM-FL versus salinity plot show a sharp drop. In the medium salinity (3e20) region, fluorescence and absorption are respectively linearly correlated with salinity (Fig. 8B and C), CDOM-FL and ag (440) decrease gradually as the salinity increase. This indicates CDOM was under conservative mixing in the mid-salinity region. At higher salinity (salinity >25), both the fluorescence versus salinity relationship and ag (440) versus salinity relationship are no longer linear. The correlation of absorption and salinity has a similar trend as that of fluorescence and salinity, as water flowed down the Yangtze River estuary into the ECS coast. The cause of this phenomenon feature is likely that water from the Yangtze River estuary is mainly affected by the diluted water of the Yangtze River and the high turbidity water of the Hangzhou Bay, while the water from offshore is mainly affected by the mixed dilution effect of the surface water in the ECS coast and open-sea water. 4. Discussion

Fig. 6. Correlation between CDOM fluorescence measured by the FLF and that measured by a spectrofluorometer.

The FLF analytical instrument was designed for characterization of CDOM in various waters. Using the instrument, high-resolution CDOM fluorescence profiles could be obtained rapidly. The FLF system requires no sample preparation (e.g., filtration or sample storage), during which the structure of natural high molecular compounds may undergo irreversible changes. The high correlation between FLF fluorescence measurements and laboratory spectrofluorometer measurements (Fig. 6; r ¼ 0.950, N ¼ 25) indicates that the FLF instrument could work as well as the more sophisticated laboratory spectrofluorometers. In addition, compared to most other in situ fluorometers with only one or several special detection channels, the field-portable laser fluorometer could provide broad and high-resolution fluorescence spectra as the same as the laboratory spectrofluorometer, which may actually make it sensitive to a broader array of organic molecules. An xenon lamp or a LED was often used in commercial in situ fluorometers as the light source. Broadband excitation light source would result in respective broadening of the Raman spectral band (Desiderio and et al., 1993), thus significantly complicating its discrimination from the

P. Chen et al. / Estuarine, Coastal and Shelf Science 146 (2014) 33e41

39

Fig. 7. Frequency histograms of ag (440) and CDOM-FL at sites in the ECS coast and the Yangtze River estuary: A) distribution of ag (440); B) distribution of CDOM-FL measured by the FLF.

constituent fluorescence. With the narrow-band laser stimulation, it has the characteristic, relatively narrow spectral band (e.g., laser), which allows its reliable discrimination in the overlapped fluorescence spectral patterns (Chekalyuk and Hafez, 2008). Development of micro and cost-efficient lasers would promote the advancement of in-situ laser fluorescence systems. The correlationship between CDOM-FL and ag (440) of samples from a wide range of aquatic environments confirms the strong linear relationship found in previous studies (Hoge et al., 1993; Green and Blough, 1994; Vodacek and et al., 1995; Nieke and et al., 1997; Ferrari and Dowell, 1998; Belzile and et al., 2006). In this study, a very tight correlation between CDOM-FL measured by the FLF and ag (440) (Fig. 5; r ¼ 0.94) was present and the results suggests that CDOM-FL provides a one rapid way to indirectly estimate ag (l). However, CDOM absorption and fluorescence are often poorly correlated due to the chemical composition of CDOM, fluorescence measurement can provide a useful tool as an analog for CDOM absorption, but will not completely replace measurement of CDOM absorption. Using the linear fitting expression derived from all 55 samples (Fig. 5; Fn (l) ¼ 1.088 ag (440) þ 0.466), we found that part of the estimated ag (l) from fluorescence data in the ECS coast was underestimated, by contrast the estimated ag (l) in the Qiandao Lake was overestimated. This is possible likely because that inland water and marine water have different optic properties, which may influence the absorption-fluorescence relationship. Considering that CDOM absorption and fluorescence varies in different aquatic environments, we recommend to study

CDOM fluorescence-absorption relationship at a regional scale, and to build estimated models for various waters according to different aquatic environments. In the actual measure, only a small part of fluorescence measurements from abundant measurements (e.g., continuous measurements) was chosen for correlation analysis between CDOM-FL and ag (440),we then used the derived correlation equation for rapid estimate of ag (440) of all samples. Riverine CDOM passes through estuarine systems as it is transported to the ocean. Therefore, estuaries can potentially play a large role in the fate of riverine CDOM (Callahan and et al., 2004). In the Yangtze River estuary and its adjacent sea, CDOM absorption and fluorescence were found to have similar gradual downward trends of magnitudes as water flowed down the Yangtze River estuary into the ECS coast (Figs. 4B and 8). The distribution trend relates to regional water mass properties and its influence scope. Because the estuarine area is commonly influenced by both terrestrial and marine forces. Thereby, knowledge of CDOM distributions and spectral characteristics is important for identifying CDOM sources, compositions, and fates (Hong and et al., 2005). Especially in highly dynamic and oceanographically complex environments like the estuary and its adjacent sea, features revealed by CDOM-FL profiles are likely to be missed by conventional analysis of discrete bottle samples, and high-resolution CDOM-FL profiles could provide insights into CDOM sources (terrestrial and marine) and water masses circulation at a finer scale (Belzile and et al., 2006). With the progress of industrialization, the influence of human activities on the estuary is more and more obvious. Using

Fig. 8. Salinity vs. ag (440) and salinity vs. FLF CDOM-FL in the Yangtze River estuary and its adjacent sea areas. (A) FLF CDOM-FL vs. salinity and ag (440) vs. salinity from all samples; (B) ag (440) vs. salinity from samples in medium-salinity region; (C) FLF CDOM-FL vs. salinity from samples in medium-salinity region.

40

P. Chen et al. / Estuarine, Coastal and Shelf Science 146 (2014) 33e41

existing laboratory or in-situ instruments cannot fully meet the increased need for acquisition of high frequency, long-term and high-resolution data series. This study illustrated an original fieldportable laser system, which could provide rapid and highresolution in situ measurements. It may be a good tool to help increase our understanding of biogeochemical processes in various waters at high-resolution, high-frequency and longer terms. Coupled with other kinds of monitoring instruments (e.g., CTD, Wetlabs ac9, ADCP, etc.), it works well in aquatic environments. 5. Conclusion An original field-portable fluorometer for rapid CDOM fluorescence measurement is developed. In the course of this study, we observed significant correlations between the measurements by the laser fluorometer and those by conventional measuring methods of absorption spectrophotometer and fluorescence spectrofluorometer. The results illustrated that the laser fluorometer can be a good tool for assessing CDOM optical properties. Using the laser fluorometer, good estimates of CDOM absorption can be obtained from in situ fluorescence data, and the work resulting from sample filtration or storage can be avoided. In addition, compared with most other in situ fluorometers, the new fluorometer provides a wider, high-resolution fluorescence emission spectrum, make it sensitive for reliable assessment of the fluorescent constituents in spectrally complex natural waters. The laser fluorometer can be a good supplement of conventional measuring methods, to help increase our understanding of biogeochemical processes in nature various waters. It can also be used for monitoring phytoplankton and oil spills, which is our on-going study. Acknowledgments This study is supported by the Public Science and Technology Research Funds Projects of Ocean (Grant No. 201005030), the National Science Foundation of China (Grant No. 41321004) and the National Key Technology Research and Development Program of China (Grant Nos. 2012BAH32B01). We thank our colleagues from the Second Institute of Oceanography at the State Oceanic Administration, who took part in the cruises, for on taking in-situ measurements. References Arrigo, K.R., Brown, C.W., 1996. Impact of chromophoric dissolved organic matter on UV inhibition of primary productivity in the sea. Oldendorf Mar. Ecol. Prog. Ser. 140 (1), 207e216. Babichenko, S., et al., 1993. Remote sensing of phytoplankton using laser-induced fluorescence. Remote Sens. Environ. 45 (1), 43e50. Baker, A., et al., 2004. Measurement of protein-like fluorescence in river and waste water using a handheld spectrophotometer. Water Res. 38 (12), 2934e2938. Belzile, C., et al., 2006. Fluorescence measured using the WETStar DOM fluorometer as a proxy for dissolved matter absorption. Estuar. Coast. Shelf Sci. 67 (3), 441e449. Bricaud, A., Morel, A., Prieur, L., 1981. Absorption by dissolved organic matter of the sea (yellow substance) in the UV and visible domains. Limnol. Oceanogr. 26 (1), 43e53. Bristow, M., et al., 1981. Use of water Raman emission to correct airborne laser fluorosensor data for effects of water optical attenuation. Appl. Opt. 20 (17), 2889e2906. Callahan, J., et al., 2004. Distribution of dissolved organic matter in the Pearl River Estuary, China. Mar. Chem. 89 (1), 211e224. Carder, K.L., et al., 1991. Reflectance model for quantifying chlorophyll a in the presence of productivity degradation products. J. Geophys. Res. Oceans (1978e2012) 96 (C11), 20599e20611. Chekalyuk, A.M., Hafez, M.A., 2008. Advanced laser fluorometry of natural aquatic environments. Limnol. Oceanogr. Methods 6, 591. Chekalyuk, A., Hafez, M., 2013. Next generation Advanced Laser Fluorometry (ALF) for characterization of natural aquatic environments: new instruments. Opt. Express 21 (12), 14181e14201.

Chen, R.F., et al., 2004. Chromophoric dissolved organic matter (CDOM) source characterization in the Louisiana Bight. Mar. Chem. 89 (1), 257e272. Coble, P.G., 1996. Characterization of marine and terrestrial DOM in seawater using excitation-emission matrix spectroscopy. Mar. Chem. 51 (4), 325e346. Conmy, R.N., Coble, P.G., Castillo, C.E.D., 2004. Calibration and performance of a new in situ multi-channel fluorometer for measurement of colored dissolved organic matter in the ocean. Cont. Shelf Res. 24 (3), 431e442. Desiderio, R.A., et al., 1993. Microstructure Profiles of Laser-induced Chlorophyll Fluorescence Spectra: Evaluation of Backscatter and Forward-scatter Fiber-optic Sensors. Determann, S., et al., 1994. Fluorescent matter in the eastern Atlantic Ocean. Part 1: method of measurement and near-surface distribution. Deep Sea Res. Part Oceanogr. Res. Pap. 41 (4), 659e675. Eaton, D.F., 1988. Reference materials for fluorescence measurement. Pure Appl. Chem. 60 (7), 1107e1114. Exton, R., et al., 1983. Laboratory analysis of techniques for remote sensing of estuarine parameters using laser excitation. Appl. Opt. 22 (1), 54e64. Ferrari, G., Dowell, M., 1998. CDOM absorption characteristics with relation to fluorescence and salinity in coastal areas of the southern Baltic Sea. Estuar. Coast. Shelf Sci. 47 (1), 91e105. Ferrari, G.M., Tassan, S., 1991. On the accuracy of determining light absorption by yellow substance through measurements of induced fluorescence. Limnol. Oceanogr. 36 (4), 777e786. Foden, J., et al., 2008. Spatial and temporal distribution of chromophoric dissolved organic matter (CDOM) fluorescence and its contribution to light attenuation in UK waterbodies. Estuar. Coast. Shelf Sci. 79 (4), 707e717. Green, S.A., Blough, N.V., 1994. Optical absorption and fluorescence properties of chromophoric dissolved organic matter in natural waters. Limnol. Oceanogr. 39 (8), 1903e1916. Hoge, F.E., Vodacek, A., Blough, N.V., 1993. Inherent optical properties of the ocean: retrieval of the absorption coefficient of chromophoric dissolved organic matter from fluorescence measurements. Limnol. Oceanogr. 38 (7), 1394e1402. Hong, H., et al., 2005. Absorption and fluorescence of chromophoric dissolved organic matter in the Pearl River Estuary, South China. Mar. Chem. 97 (1), 78e89. Jiang, F., et al., 2008. The application of excitation/emission matrix spectroscopy combined with multivariate analysis for the characterization and source identification of dissolved organic matter in seawater of Bohai Sea, China. Mar. Chem. 110 (1), 109e119. Keith, D., Yoder, J., Freeman, S., 2002. Spatial and temporal distribution of coloured dissolved organic matter (CDOM) in Narragansett Bay, Rhode Island: implications for phytoplankton in coastal waters. Estuar. Coast. Shelf Sci. 55 (5), 705e717. Killinger, D., Sivaprakasam, V., 2006. How water glows: water monitoring with laser fluorescence. Opt. Photonics News 17 (1), 34e39. Kowalczuk, P., et al., 2009. Characterization of dissolved organic matter fluorescence in the South Atlantic Bight with use of PARAFAC model: Interannual variability. Mar. Chem. 113 (3), 182e196. Lee, Z., et al., 1994. Model for the interpretation of hyperspectral remote-sensing reflectance. Appl. Opt. 33 (24), 5721e5732. Miller, R.L., et al., 2002. Determining CDOM absorption spectra in diverse coastal environments using a multiple pathlength, liquid core waveguide system. Cont. Shelf Res. 22 (9), 1301e1310. Nieke, B., et al., 1997. Light absorption and fluorescence properties of chromophoric dissolved organic matter (CDOM), in the St. Lawrence Estuary (Case 2 waters). Cont. Shelf Res. 17 (3), 235e252. Niewiadomska, K., et al., 2008. Submesoscale physical-biogeochemical coupling across the Ligurian current (northwestern Mediterranean) using a bio-optical glider. Limnol. Oceanogr. 53 (5), 2210. Pegau, W., Zaneveld, J.R., 1993. Temperature-dependent Absorption of Water in the red and Near-infrared Portions of the Spectrum. Singh, S., D'Sa, E.J., Swenson, E.M., 2010. Chromophoric dissolved organic matter (CDOM) variability in Barataria Basin using excitationeemission matrix (EEM) fluorescence and parallel factor analysis (PARAFAC). Sci. Total Environ. 408 (16), 3211e3222. Sivaprakasam, V., Killinger, D.K., 2003. Tunable ultraviolet laser-induced fluorescence detection of trace plastics and dissolved organic compounds in water. Appl. Opt. 42 (33), 6739e6746. Spencer, R.G., et al., 2007. Diurnal variability in riverine dissolved organic matter composition determined by in situ optical measurement in the San Joaquin River (California, USA). Hydrol. Process. 21 (23), 3181e3189. Stedmon, C., Markager, S., 2001. The optics of chromophoric dissolved organic matter (CDOM) in the Greenland Sea: an algorithm for differentiation between marine and terrestrially derived organic matter. Limnol. Oceanogr., 2087e2093. Stedmon, C., Markager, S., Kaas, H., 2000. Optical properties and signatures of chromophoric dissolved organic matter (CDOM) in Danish coastal waters. Estuar. Coast. Shelf Sci. 51 (2), 267e278. Stedmon, C.A., Markager, S., Bro, R., 2003. Tracing dissolved organic matter in aquatic environments using a new approach to fluorescence spectroscopy. Mar. Chem. 82 (3), 239e254. Suping1a, Y., Yinpinga, H.H.Y.J.F., Pingshenga, S., 2010. Design and Research of Analysis Instrument Based on Q-switch Micro-crystal UV Laser-induced Fluorescence Spectroscopy. in Proc. of SPIE Vol.

P. Chen et al. / Estuarine, Coastal and Shelf Science 146 (2014) 33e41 Tedetti, M., Guigue, C., Goutx, M., 2010. Utilization of a submersible UV fluorometer for monitoring anthropogenic inputs in the Mediterranean coastal waters. Mar. Pollut. Bull. 60 (3), 350e362. Tedetti, M., Joffre, P., Goutx, M., 2013. Development of a field-portable fluorometer based on deep ultraviolet LEDs for the detection of phenanthrene-and tryptophan-like compounds in natural waters. Sens. Actuators B Chem.. Twardowski, M.S., et al., 2004. Modeling the spectral shape of absorption by chromophoric dissolved organic matter. Mar. Chem. 89 (1), 69e88. Velapoldi, R.A., Mielenz, K., 1980. In: A Fluorescence standard reference material: quinine sulfate dihydrate, vol. 260. Department of Commerce, National Bureau of Standards.

41

Vodacek, A., et al., 1995. The use of in situ and airborne fluorescence measurements to determine UV absorption coefficients and DOC concentrations in surface waters. Limnol. Oceanogr. 40 (2), 411e415. Yamashita, Y., et al., 2010. Fluorescence characteristics of dissolved organic matter in the deep waters of the Okhotsk Sea and the northwestern North Pacific Ocean. Deep Sea Research Part II Top. Stud. Oceanogr. 57 (16), 1478e1485. Zhang, Y., et al., 2007. Chromophoric dissolved organic matter (CDOM) absorption characteristics in relation to fluorescence in Lake Taihu, China, a large shallow subtropical lake. Hydrobiologia 581 (1), 43e52.