Photochemical bleaching of oceanic dissolved organic matter and its effect on absorption spectral slope and fluorescence

Marine Chemistry 155 (2013) 81–91

Contents lists available at ScienceDirect

Marine Chemistry

j ourna l homepage: www.e lsev ie r .com/ locate /marchem

Photochemical bleaching of oceanic dissolved organic matter and itseffect on absorption spectral slope and fluorescence

John R. Helms a,⁎, Aron Stubbins b, E. Michael Perdue c,1, Nelson W. Green c,1,Hongmei Chen a, Kenneth Mopper a,⁎⁎a Old Dominion University, Department of Chemistry and Biochemistry, 4402 Elkhorn Ave., Norfolk, VA 23529, United Statesb Skidaway Institute of Oceanography, 10 Ocean Science Circle, Savannah, GA 31411, United Statesc School of Earth and Atmospheric Sciences, Georgia Institute of Technology, Atlanta, GA 30332, United States

Abbreviations: A, absorbance; a, Napierian absorptiobiological index; CDOM, chromophoric dissolved orgorganic carbon; DOM, dissolved organic matter; EEM, flumatrix spectroscopy; FDOM, fluorescent dissolved orgindex, often McKnight's fluorescence index; HIX, fluorpath length; M:C, ratio of EEM Peak M to Peak C; NELHof Hawaii; NPOC, non-purgeable organic carbon; ROwith electrodialysis; Sxxx–xxx, spectral slope, subscriptsspectral slope ratio, S275–295:S350–400; SUVA, specific UVcoefficient divided by [DOC]; UV, ultraviolet light, b400⁎ Correspondence to: J.R. Helms, University of North Ca

of Chemistry and Biochemistry, 601 S. College Rd., WilmiTel.: +1 757 581 8003.⁎⁎ Corresponding author. Tel.: +1 757 683 4094; fax:

E-mail addresses: [email protected] (J.R. Helms), km1 Present address: Department of Chemistry, Ball State

United States.

0304-4203/$ – see front matter © 2013 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.marchem.2013.05.015

a b s t r a c t

Article history:Received 21 September 2012Received in revised form 23 May 2013Accepted 30 May 2013Available online 14 June 2013

Keywords:Colored dissolved organic matterCarbonLight absorptionFluorescencePhotochemistryExcitation emission matrixSpectral slope

Photobleaching of open-ocean dissolved organic matter (DOM) is typically treated as a removal mechanism;however, photobleaching also encompasses a poorly characterized suite of transformative processes. To ex-amine the qualitative changes to DOM optical properties during photobleaching, 674 m N. Pacific DOM, con-centrated and desalted by reverse osmosis with electrodialysis (RO/ED), was subjected to 68 days ofcontinuous irradiation in a UV solar simulator. Approximately 84% of chromophoric and fluorescent DOM(CDOM and FDOM respectively) and 38% of dissolved organic carbon (DOC) were lost during the irradiation.Based on these results the concentration of photochemically refractory DOC in the surface pacific is estimatedto be 27 μmol of carbon per liter. In addition, the spectra of the remaining CDOM and FDOM were shifted to-wards shorter wavelengths, a result which has important implications for the interpretation of fluorescenceexcitation emission matrix (EEM) spectra because the relative positions of fluorescence maxima are often at-tributed to differences in FDOM source. Qualitative indices derived from CDOM and FDOM spectra for the ir-radiated deep DOM sample resembled those for surface waters of the North Pacific Ocean indicating thatphotobleaching has a significant influence upon the optical properties of DOM in the open ocean.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Chromophoric dissolved organic matter (CDOM) is the main UVlight-absorbing substance in the open ocean (Zepp, 2002; Kitidis etal., 2006). In spite of the large source of allochthonous CDOM fromrivers and coastal wetlands (Blough and Del Vecchio, 2002; DelVecchio and Blough, 2004b; Spencer et al., 2009), the dominant sourceof CDOM in theopen ocean is thought to be the heterotrophically altered

n coefficient; BIX, fluorescenceanic matter; DOC, dissolvedorescence excitation emissionanic matter; FI, fluorescence

escence humification index; L,A, Natural Energy Laboratory

/ED, reverse osmosis coupleddenote wavelength range; SR,absorption, decadal absorptionnm; WRU, Water Raman Units.rolina Wilmington, Departmentngton, NC 28403, United States.

+1 757 683 [email protected] (K. Mopper).University, Muncie, IN 47306,

rights reserved.

autochthonous material that is produced in and immediately beneaththe euphotic zone and also released at depth from sinking particles(Chen and Bada, 1992; Hayase and Shinozuka, 1995; Rochelle-Newalland Fisher, 2002; Nelson et al., 2004, 2007; Yamashita and Tanoue,2004; Swan et al., 2009). CDOM plays an important role in protectingorganisms in the upper water column from harmful UV radiation(Williamson et al., 2001; Zepp, 2002) and represents a significant inter-ference for remote sensing of ocean chlorophyll (Carder et al., 1989).Further, the products of photochemical degradation of CDOM representa source of nutrients for phytoplankton (Tarr et al., 2001; Stedmon et al.,2007; Vähätalo and Järvinen, 2007), metabolic substrates for hetero-trophs (Kieber et al., 1989; Miller et al., 2002), and climatologically rele-vant gasses (Johannessen andMiller, 2001; Cutter et al., 2004; Stubbinset al., 2006; Toole et al., 2006). The loss of UV and visible light absorptionand fluorescence that occurs during photochemical degradation ofCDOM is referred to as photobleaching (Zika, 1980; Kieber et al.,1990Chen and Bada, 1992; Skoog et al., 1996). The basin-scale impactof CDOM photobleaching in surface waters of the remote Pacific Oceanhas recently been highlighted by large-scale surveys (Swan et al.,2009; Yamashita and Tanoue, 2009), which show that CDOM absor-bance and fluorescence are strongly depleted in the surfacemixed layer.

Although there is substantial evidence for widespread CDOMphotobleaching from its distribution in the ocean (Hayase et al.,1988; Chen and Bada, 1992; Nelson et al., 2007; Swan et al., 2009;

82 J.R. Helms et al. / Marine Chemistry 155 (2013) 81–91

Jørgensen et al., 2011) and direct observations of photobleaching ofCDOM in terrestrially-impacted waters (Skoog et al., 1996; DelVecchio and Blough, 2002, 2004a; Twardowski and Donaghay, 2002;Helms et al., 2008), photobleaching of open-ocean CDOM has notbeen widely examined (Opsahl and Benner, 1998; Ortega-Retuertaet al., 2010; Stubbins et al., 2012), and little has been reported withrespect to the changes in the spectral distribution of absorption andfluorescence during photodegradation of open-ocean CDOM (Hayaseet al., 1988; Chen and Bada, 1992).

The aim of this study was to explore how deep ocean DOMchanges qualitatively (i.e., relative wavelength dependence of ab-sorption and fluorescence) during photodegradation as an indicationof how the optical properties of this same DOM would be alteredwhen transported from depth to the euphotic zone. Photobleachingwas assessed for a deep ocean DOM sample concentrated anddesalted using reverse osmosis coupled with electrodialysis (RO/ED).RO/ED has higher extraction efficiencies than commonly reported forsolid phase extraction and ultrafiltration, and the optical and chemicalproperties of the RO/ED DOM closely resemble those of the unextractedDOM (Koprivnjak et al., 2006, 2009). The use of deep ocean DOM fromthe Pacific Ocean provided starting material with minimal direct contri-bution from terrestrial environments and broadly representative of theDOC pool of the deep ocean, which is one of the longest lived andmost abundant carbon stores on Earth (Hansell, 2013). In previousstudies irradiating natural deep ocean seawater it took less than28 days before CDOM light absorption at wavelengths greater than350 nm was reduced below the detection limits of a standard spectro-photometer with a 10 cm path length cuvette (Stubbins et al., 2012).In the current study, the seventeen-fold increase in CDOM light absorp-tion in the RO/ED concentrate compared to seawater allowed photo-bleaching to be studied over a longer irradiation period withoutreducing CDOM optical properties below instrument detection limits,thus providing novel insight into the optical properties of extensivelyphotodegraded CDOM.

2. Material and methods

2.1. Cleaning procedures

All glassware was cleaned with ~1 M HCl and rinsed with MilliQUV ultrapure grade water (Millipore), designated below as “MilliQwater”. Glassware was combusted at 450 °C for ≥4 h. Plasticwarewas cleaned with ~1 M HCl and rinsed with MilliQ water. Stainlesssteel equipment was cleaned with mild detergent and copiouslyrinsed with MilliQ water. All containers were rinsed several timeswith sample prior to filling. Filter capsules (0.1 μm pore size,Whatman PolyCap) were rinsed briefly with acetonitrile, flushedwith >20 L of MilliQ water, and conditioned with approximately1 L of sample prior to use.

2.2. Sample collection, handling, and irradiation

Four North Pacific Oceanwater sampleswere collected at 23°N, 158°W (station ALOHA) aboard R/V Kilo Moana. Sampling depths were 5 m(surface subtropical gyre), 125 m (chlorophyll a maximum), 770 m(oxygenminimum layer), and 3500 m (lower circumpolarwater). Sam-ples were gravity-filtered through 0.1 μm capsule filters and stored fro-zen in pre-combusted glass containers. Two of the four ALOHA samples(5 m and 3500 m) were concentrated by reverse osmosis and desaltedby electrodialysis aboard the ship. The isolation and characterization ofthese samples are described elsewhere (Helms, 2012). Nine large-volume water samples were also collected from the 674 m depthpumping system at the Natural Energy Laboratory of Hawaii, Authority(NELHA) in Kona, Hawaii. The water was pumped through 0.1 μmcapsule filters into the 200 L polyethylene tank of the RO/ED system.The sample used for the irradiation study was reduced from 220 L to

8.3 L (concentration factor of 26.6) and had a final conductivity of47.1 μS cm−1. The isolated sample was frozen and shipped back to thelaboratory in Norfolk, VA, where it was immediately re-frozen andstored at −20 °C until used in the experiments described below.

Aliquots of the RO/ED concentrate (NELHA 674 m) were trans-ferred to ~550 mL quartz flasks and placed inside a solar UV simula-tor containing 12 Q-Panel UVA340 bulbs, which provided a spectralshape similar to that of natural sunlight from approximately300–365 nm (Q-Panel), but under-represented solar irradiance atwavelengths greater than about 365 nm. The light output from thesolar simulator was monitored during the course of the irradiationexperiment using a Biospherical PUV 2510 radiometer. A descriptionof the solar simulator can be found inMinor et al. (2007). The sampleswere irradiated constantly during the experiment. Sub-samples werecollected and analyzed at exposure intervals ranging from 5 days to68 days. Based on the measured intensity of the lamps and the factthat samples were irradiated continuously, each 24-hour day oflight exposure is approximately equivalent to 4 days of natural sun-light at 35° latitude (Leifer, 1988; Kieber et al., 2006; Minor et al.,2007; Helms et al., 2008).

At the end of each irradiation period, an aliquot of sample wastested for microbial activity using 3H labeled thymidine (TrD) incor-poration (Fuhrman and Azam, 1982; Smith and Azam, 1992). Afresh 5 μm filtered water sample was collected from a pond adjacentto the laboratory at Old Dominion University and used as a“live-control” to test the effectiveness of the assay. All irradiated sam-ples yielded radioactivity measurements less than 1% of killed con-trols; thus, all measured changes to DOM optical properties wereprincipally due to photochemical reactions because microbial activitywas negligible.

2.3. UV–visible absorption and fluorescence spectroscopy

UV–visible absorption spectra (190–900 nm) were measuredusing an Agilent 8453 diode array spectrophotometer with a 5 cmor 10 cm quartz cuvette. MilliQ water was used as the blank. Absor-bance values were corrected for instrument baseline drift, refractiveindex, and temperature variations according to Green and Blough(1994) and converted to Napierian absorption coefficients using theformula:

a ¼ 2:303A=L ð1Þ

where a = Napierian absorption coefficient (m−1), A = absorbance,and L = path length (m) (Green and Blough, 1994). CDOM absorp-tion spectra are modeled using an exponential function of decreasingabsorption with increasing wavelength:

aλ ¼ aλrefe−S λ−λrefð Þ ð2Þ

where a = Napierian absorption coefficient (m−1), λ = wavelength(nm), λref = reference wavelength (nm), and S = spectral slope(nm−1) (Helms et al., 2008). Specific UV absorbance (SUVA) was de-termined by dividing the absorbance (at 254 nm and at 300 nm) bythe DOC concentration.

The spectral slope over the wavelength range of 300–700 nm wascalculated using non-linear regression of the absorption spectra(Twardowski et al., 2004). All other spectral slopes were calculatedby linear regression of the natural log-transformed absorption spectra(Helms et al., 2008). Spectral slope ratio (SR) was calculated by divid-ing S275–295 by S350–400 (Helms et al., 2008). Slope spectra (Loiselle etal., 2009) were generated using linear regression slopes of the naturallog absorption spectrum over a sliding 21 nm interval (i.e., centralvalue ± 10 nm) with 1 nm resolution. First and second derivativespectra were obtained for absorption and natural log absorption usinglinear regression over 21 nm intervals (i.e., central value ± 10 nm).

83J.R. Helms et al. / Marine Chemistry 155 (2013) 81–91

The second derivative spectra reveal features within the absorbancespectrum that are not easily identified otherwise (e.g., peaks that appearas shoulders).

Fluorescence excitation emission matrix (EEM) spectra were mea-sured using a FluoroMax-2 spectrofluorometer (Jobin Yvon/Spex Indus-tries) and a 1 cm × 1 cm quartz fluorescence cuvette. For the ALOHAsamples, emission spectra (300–550 nm; 2 nm intervals) were mea-sured over a range of excitation wavelengths (240–440 nm; 5 nmintervals). The data integration time was 0.5 s. EEMs were correctedfor inner-filter effects and scatter peaks were removed (Lakowicz,1999; Zepp et al., 2004; Cory and McKnight, 2005). Fluorescenceintensities were normalized to the integrated water Raman peak atλex = 350 nm; λem = 370–428 nm (Lawaetz and Stedmon, 2009).For the NELHA samples, emission spectra (300–550 nm; 2 nmintervals) were measured over a range of excitation wavelengths(200–455 nm; 6.375 nm intervals). The data integration time was1.0 s. EEMs were corrected for inner-filter effects and scatter peakswere removed. Fluorescence intensities were normalized to the inte-grated water Raman peak at λex = 346.625 nm; λem = 366–422 nmand corrected for the RO/ED volumetric concentration factor and DOCrecovery.

Themajor peaks in the EEMswere identified using the wavelengthranges identified by Coble (1996, 2007): Peak A (λex = ~260 nm;λem = 380–460 nm), Peak C (λex = ~350 nm; λem = 420–480 nm),Peak M (λex = ~312 nm; λem = 380–420 nm), Peak B (λex =~275 nm; λem = ~310 nm), and Peak T (λex = ~275 nm; λem =~340 nm). The ratio of Peak M to Peak C fluorescence (M:C), whichshould correlate with the relative source strength of marine-derivedfluorescent DOM (FDOM) vs. terrestrial FDOM, was calculated by divid-ing the intensity at 310ex/410em by the intensity at 345ex/445em(Burdige et al., 2004; Para et al., 2010; Romera-Castillo et al., 2011).M:C values have been reported ranging from 1.1 in marine sedimentpore waters (Burdige et al., 2004) to 3.11 in coastal Mediterranean sur-face waters (Para et al., 2010). Fluorescence index (FI), which indicatesthe relative influences of terrigenous vs. microbial DOM by comparingthe intensity of two commonly observed humic-related fluorescencemaxima, was calculated by dividing the intensity at 372ex/470em bythe intensity at 372ex/520em (Battin, 1998; McKnight et al., 2001).Typical values for FI in microbially sourced fulvic acids are 1.5–3 whileFI values of terrestrially sourced fulvic acids are b1.5 (McKnight et al.,2001). Fluorescence humification index (HIX), which compares twobroad aromatically dominated fluorescence maxima, was determinedat 253 nm excitation by dividing the integrated emission from 434 to480 nm by the integrated emission from 300 to 346 nm (Kalbitz et al.,1999). High HIX values are associated with high C/H ratios, but maybe significantly impacted by concentration effects (Zsolnay, 2003).The fluorescence biological index (BIX), which positively correlates

Table 1DOC and UV–visible measurements of unextracted seawater samples collected from stationtheses are adjusted for volumetric concentration factor and DOC recovery for comparison tolated because long wavelength aCDOM values were below instrument detection limits.

Sample DOC (μM C) a254 (m−1) a300 (m−1)

Unconcentrated seawater samplesALOHA — 5 m 69 1.13 0.225ALOHA — 125 m 78 1.21 0.327ALOHA — 770 m 45 1.26 0.373ALOHA — 3500 m 52 1.27 0.368NELHA — 674 m 44 1.02 0.469

Concentrated and desalted by RO/EDALOHA — 5 m 1325 14.1 (0.730) 3.6 (0.186)ALOHA — 3500 m 1254 19.1 (0.791) 7.2 (0.298)NELHA-0 daya 790 17.9 (1.00) 8.00 (0.449)NELHA-5 daysa 616 15.6 (0.875) 5.30 (0.297)NELHA-26 daysa 576 9.56 (0.537) 2.37 (0.133)NELHA-68 daysa 488 5.56 (0.312) 1.29 (0.0724)

a Photobleaching experiment (NELHA 674 m RO/ED isolate).

with the contribution of autochthonous or microbially derived DOM,was determined by dividing intensity at 308ex/380em by intensity at308ex/430em (Huguet et al., 2009). BIX values reported for urbanstormwater samples ranged from 0.3 to 1.3 (Ghervase et al., 2010),with similar ranges (0.64–1.02) reported for cloud waters (Birdwelland Valsaraj, 2010), and high values (up to 4.95) reported for microbialmat associated waters (Birdwell and Engel, 2010).

2.4. Dissolved organic carbon

Non-purgeable total organic carbon (NPOC) was measured usinghigh temperature (720 °C) catalytic combustion on a ShimadzuTOC-VCPH carbon analyzer. The instrument was calibrated usingdry primary standard grade potassium hydrogen phthalate (AlphaAesar) dissolved in MilliQ water and evaluated by measuring deepseawater consensus reference material (http:/yyy.rsmas.miami.edu/groups/biogeochem/CRM.html). As all samples were filtered to lessthan 0.1 μm prior to RO/ED isolation and irradiation, all NPOC mea-surements are referred to as DOC (Stubbins and Dittmar, 2012).

3. Results

3.1. RO/ED isolation of open ocean DOM

The DOM concentrated and desalted by RO/ED from nine replicateNELHA samples (674 m depth) gave recoveries of 78 ± 3% (based onDOC). The NELHA 674 m sample used in the irradiation experimentsdescribed below was desalted to a conductivity of 47.1 μS cm−1 byED and concentrated 26.6-fold using RO. The fully isolated andde-salted sample contained 67% of DOC (790 μM), and an additional9% of DOC (168 μM) was in the NaOH rinse solution (overallyield of 76%). Only the fully isolated and de-salted sample was usedin this study. When corrected for volumetric concentration factorand DOC recovery, absorption measurements (250–500 nm) of theNELHA 674 m isolate agree with the predicted absorption within10% (Table 1; Fig. 1A). UV absorption attributable to inorganic species(NO3

−, NO2−, Br−) was removed by electrodialysis (Fig. 1B). Fluores-

cence in the EEM Peak C and M regions was recovered by RO/EDroughly in proportion with DOC (Table 2), while Peak A, B, and T re-gions showed lower recoveries (Table 2; Fig. 1C).

3.2. Photobleaching of RO/ED isolated DOM from the open ocean

Prior to irradiation, the RO/ED isolate had an a254 value of 17.9 m−1,an a300 value of 8.0 m−1, and a DOC value of 790 μM C. Photobleachingcaused aCDOM to decrease over a broad range ofwavelengths, particular-ly at ~325 nm and ~500 nm (Fig. 2G and H). During the irradiation,

ALOHA and RO/ED isolated DOM from NELHA 674 m water samples. Values in paren-non-isolated samples. Not determined (n.d.) denotes parameters that were not calcu-

SUVA254 (m2 g-C−1) S300–700 (nm−1) S275–295 (nm−1) SR

0.596 n.d. 0.0375 6.510.564 n.d. 0.0307 2.681.01 n.d. 0.0281 1.640.880 n.d. 0.0275 1.610.843 0.0103 0.0151 1.67

0.866 0.0125 0.0351 3.151.22 0.0123 0.0235 1.850.820 0.0095 0.0191 2.140.916 0.0088 0.0273 3.560.595 0.0112 0.0372 6.300.413 0.0092 0.0388 8.92

Wavelength (nm)

Abs

orpt

ion

Coe

ffic

ient

(m

-1)

20

10

5

0

15

250 300 400 500350 450

190 250210 230

Wavelength (nm)

1200

800

400

0

A

B

250 550350 450

Wavelength (nm)

Fluo

resc

ence

(W

RU

)

0.03

0.02

0.02

0.00

CNELHA-SW0d-DOC corr.

λem=500nm λex= 250nm

Abs

orpt

ion

Coe

ffic

ient

(m

-1)

Fig. 1. Comparison of (A) CDOMUV–visible absorption spectra and (B) nitrate, nitrite, andbromide UV spectra measured before (NELHA-SW) and after isolation by RO/ED(NELHA-0 day). The absorption calculated by correcting the NELHA-SW spectrum forRO/ED concentration factor and DOC recovery (predicted) is also shown. (C) Fluorescenceexcitation (λem = 500 nm) and emission (λex = 250 nm) spectra of NELHA-SW,and NELHA-0 day corrected for volumetric concentration factor and DOC recovery(0 day-DOC corr.).

Table 2Summary of EEM peak intensities and locations for non-isolated seawater (SW) as wellas dark control (0 day) and irradiated (68 days) RO/ED isolates from NELHA 674 mwater samples. Maximum peak intensity values are reported as Water Raman Units(WRU) and corrected for RO/ED concentration factor and DOC recovery.

NELHA NELHA NELHA674 m 674 m 674 m

SWa 0 dayb 68 daysb

Peak A (WRU) 0.025 0.015 0.004λex max (nm) 246 253 246λem max (nm) 424 426 420

Peak C (WRU) 0.011 0.011 0.002λex max (nm) 313 313 326λem max (nm) 414 420 404

Peak M (WRU) 0.014 0.012 0.003λex max (nm) 286 286 286λem max (nm) 412 410 410

Peak B (WRU) 0.011 0.007 0.004λex max (nm) 280 280 280λem max (nm) 328 326 318

Peak T (WRU) 0.013 0.009 0.002λex max (nm) 275 280 275λem max (nm) 346 376 378

a Unconcentrated seawater samples.b Concentrated and desalted by RO/ED.

84 J.R. Helms et al. / Marine Chemistry 155 (2013) 81–91

decreaseswere observed for aCDOM at all wavelengths as exemplified bya254 and a300 (Fig. 3). DOC and SUVA also decreased (Fig. 3).

The spectral slope became shallower with increasing wavelengthand increasing irradiation time (Fig. 2B). Themost significantminimumin the second derivative absorption spectra (Fig. 2C) is at ~280 nmwithtwo additional bands centered at ~375 nmand ~425 nm. The ~280 nmband shifted to shorter wavelengths during irradiation (Fig. 2C).

The initial natural log absorption spectrum was approximatelylinear; however, after extensive photobleaching, several features de-veloped (Fig. 2D). The spectra became steeper at ~270–300 nm and425–575 nm and shallower at ~330–380 nm (Fig. 2D). These changesin the spectral shape are clearly reflected as aminimum at 280–290 nmand amaximumat 350–370 nm, respectively, in thefirst derivative nat-ural log absorption spectra (Fig. 2E) and as a corresponding maximumat 280–290 nm and minimum at 350–370 nm in the spectral slopespectra (Fig. 2I). Interestingly, the spectral slope again became steeperat ~450–500 nm, which is clearly seen as a minimum at 470–480 nmin the first derivative spectra (Fig. 2E) and a corresponding maximum

in the spectral slope spectra (Fig. 2I). The second derivative naturallog spectra (Fig. 2F) show that photobleaching led to the emergenceof an additional peak at ~330 nm. The 275 nm and 375 nm minimaalso intensified during photobleaching.

Prior to irradiation, the NELHA 674 m RO/ED isolate UV spectrumyielded a S275–295 value of 0.019 nm−1 and a SR value of 2.14. Upon irra-diation, S275–295 increased to 0.0388 nm−1, S350–400 decreased, whileS300–700 remained relatively unchanged (Table 1; Figs. 4 and 5). SR in-creased to 8.92, which is slightly higher than the surface valuemeasuredat station ALOHA during this study (Table 1; Fig. 6). The relationship be-tween various spectral slope coefficients and photobleaching wasstrongly wavelength dependent. S275–295 increased linearly with in-creased removal of a300, while S350–400 decreased linearly (Fig. 5A).However, S300–700 changed very little over the course of the irradiation(Fig. 5A). SR increased exponentially with respect to photobleaching ofa300 (Fig. 5B). While photochemical remineralization of DOC correlatedwith loss of a300 (Fig. 5C), SUVA254 remained unchanged or increasedslightly during the earliest phase of photobleaching and decreased forthe remainder of the experiment (Figs. 3C and 5D).

Like aCDOM, fluorescence decreased upon irradiation (Table 2;Fig. 6). The largest decrease occurred at Peak C (~84%; Table 2),while the smallest decrease occurred at Peak B (~47%; Table 2).Peaks A, M and T each decreased by 76% ± 2% (Table 2) and the M:C ratio increased from 1.17 to 1.43. After the 68 day irradiation,Peak A remained the dominant peak (Fig. 6); however, the locationof its maximum shifted from 253ex/426em to 246ex/420em. Peak Cshowed the most significant shift, from 313ex/420em to 326ex/404em,while the remaining peaks exhibited relatively minor shifts. The minorshift in λex observed for Peak B (Table 3) may be partly due to thephotobleaching of the broader and more intense fluorescence in the vi-cinity of Peaks T and M, thereby allowing a better resolution of Peak B.

3.3. Vertical distribution of CDOM and FDOM

CDOM absorption of the unconcentrated water samples from stationALOHAwas lowest at the surface (Fig. 7A)while SRwas highest (Fig. 7B).This contrasts with DOC (Fig. 7A inset), which was highest in the 125 msample (chlorophyll maximum) and lowest in the 770 m sample(oxygen minimum). The depth distribution of FDOM varied dependingupon which excitation–emission signal is examined (Tables 2 and 3;Figs. 7 and 8). Peaks A, C, and M exhibit similar profiles with maximum

da/d

λ

d2a/

dλ2

dln(

a)/d

λ

d2ln

(a)/

dλ2

% P

hoto

blea

ched

Spec

tral

Slo

pe (

nm-1

)

Phot

oble

ache

d (m

-1)

ln (

a)A B C

D E F

G H I

0 d5 d26 d68 d

1st Derivative 2nd Derivative

Wavelength (nm) Wavelength (nm)Wavelength (nm)

1st Derivative Natural log

2nd DerivativeNatural log

Natural log

Spectral Slope

30 0.25 0.010

0.005

0.000

0.002

0.001

0.000

0.05

0.04

0.03

0.02

0.01

0

-0.001

-0.005

0

-0.25

-0.5

0

-0.01

-0.02

-0.03

-0.04

100

75

50

25

0

20

10

0

3

2

1

0

-1

-2

-3

15

12

9

6

3

0

250 300 350 400 450 500 250 300 350 400 450 500 250 300 350 400 450 500

250 300 350 400 450 500 250 300 350 400 450 500 250 300 350 400 450 500

250 300 350 400 450 500 250 300 350 400 450 500 250 300 350 400 450 500

Abs

orpt

ion

Coe

ffic

ient

(m

-1)

Fig. 2. (A) UV–visible absorption spectra, (B) first derivative absorption spectra, (C) second derivative absorption spectra, (D) natural log absorption spectra, (E) first derivativenatural log absorption spectra, (F) second derivative natural log absorption spectra, (G) amount of absorption removed by photobleaching, (H) percentage of absorption removedby photobleaching, and (I) spectral slope spectra, obtained for NELHA RO/ED isolates irradiated for 0 day, 5 days, 26 days, and 68 days.

85J.R. Helms et al. / Marine Chemistry 155 (2013) 81–91

fluorescence at depth and minimal fluorescence at the surface (Fig. 7C).Peaks B and T were highest at 125 m and lowest at 3500 m and 770 mrespectively (Fig. 7C). This variability is also reflected in the differencesbetween the depth profiles for M:C (maximum at 5 m; Table 3;Fig. 7D) and BIX (maximum at 125 m; Table 3; Fig. 7D).

Prior to irradiation, FDOM in the NELHA 674 m Pacific Ocean RO/EDisolate exhibited a dominant peak at 253ex/426em (Peak A) (Figs. 6and 8) and had a FI value of 1.53, a HIX value of 3.9, and a BIX value of0.74 (Tables 2 and 3). The M:C, FI, and BIX values of the NELHA 674 mRO/ED isolate (prior to irradiation) were similar to those found ina non-irradiated surface water RO/ED isolate from the same region(Station ALOHA; Table 3), while HIX was much higher than the surfacewater isolate (3.90 vs. 1.82 for 5 m ALOHA isolate; Table 3) and lowerthan the value measured for DOM isolated from 3500 m (HIX = 4.6).This trendwith depth is also evident in the non-isolated (non-irradiated)seawater values for HIX (Table 3; Fig. 7), though all the values are lowerthan the DOM isolates. Prior to irradiation, Peak Mwas slightly more in-tense than Peak C for the 674 m RO/ED isolate (Table 2; Fig. 8) with anM:C value of 1.17 (Table 3). In comparison, the M:C value was 1.21 inthe 5 m ALOHA RO/ED isolate and 1.13 in the 3500 m ALOHA isolate.Again, this trend with depth is similar to the M:C values measured forthe non-isolated ALOHA samples (Fig. 7D).

4. Discussion

4.1. Chemical/physical fractionation of DOM due to RO/ED

Concentration independent optical parameters such as SUVA254,spectral slope, and fluorescence indices suggest that fractionation of

chromophoric and fluorophoric material as a result of the DOM isola-tion/desalting procedure was minor. The slight increases in S275–295and SR (Table 1), the decreases in Peak A fluorescence (Table 2;Fig. 1C), M:C, and FI (Table 3) relative to the non-isolated sample sug-gest that the material that was incompletely recovered likely includ-ed high molecular weight DOM (Helms et al., 2008). The differencesin Peak B and Peak T fluorescence and BIX observed between the iso-lated and non-isolated samples suggest that up to 36% of protein-likeFDOM was lost during RO/ED processing or storage (Table 2). WhileEEMs were not measured for the NaOH rinses of the RO/ED systemas part of this study, we speculate that peptides and amino acids re-versibly adsorbed to the RO membrane and ion exchange membranesof the ED (N. Green and E.M. Perdue, unpublished results). The in-crease in HIX during isolation (Table 3) was possibly due to a concen-tration artifact. Zsolnay (2003) previously described this artifact inwhich high DOC straw extracts were shown to exhibit red shiftedfluorescence maxima and much higher HIX values relative to ten-fold diluted extracts. Zsolnay (2003) hypothesized that this was dueto concentration related differences in DOM physicochemical proper-ties or tertiary structure (presumably formation of aggregates, mi-celles, or hydrophobic micro-environments at high concentration).

4.2. Photobleaching of CDOM and its effect on spectral slope and sloperatio

The steepness of the first derivative absorption spectrum de-creases with increasing wavelength and increasing irradiation time(Fig. 2B), the latter as a result of photobleaching. The second deriva-tive spectra (Fig. 2C) indicate that the short wavelength absorbing

DO

C (

μM)

0 20 40 60

25

20

10

0

A

900

600

500

400

700

800B

Irradiation Time (d)

1.2

0.4

0

0.8

C

15

0 20 40 60

0 20 40 60

a254

a300

SUV

A (

m2 g

-C-1

)A

bsor

ptio

n C

oeff

icie

nt (

m-1

)

Fig. 3. The trends of decreasing (A) UV–visible absorption, (B) DOC, and (C) specific UVabsorbance (SUVA) with increasing irradiation time. Error bars were calculated as twotimes the standard deviation (n ≥ 3) plus the blank (absorption coefficient error barswere smaller than the circles).

Slop

e R

atio

0 20 40 60

0.05

0.04

0.02

0

A

10

4

2

0

6

8B

Irradiation Time (d)

0.03

0 20 40 60

0.01

S275-295

S350-400

S300-700

Spec

tral

Slo

pe (

nm-1

)

Fig. 4. (A) Spectral slope coefficients for 300–700 nm, 275–295 nm, and 350–400 nmshowing different trends with irradiation. (B) Slope ratio (SR) increased duringirradiation.

86 J.R. Helms et al. / Marine Chemistry 155 (2013) 81–91

chromophores particularly at ~275–280 nm are preferentially lost.The ~275–280 nm chromophores in the second derivative spectrum(Fig. 2C) are likely related to aromatic (e.g., π → π*) transitions(Summers et al., 1987). The small blue shift (i.e. ~5 nm shift toshorter wavelengths) in the minimum in the second derivative spec-trum at ~280 nm during the course of the irradiation, suggests thatthe degree of conjugation and/or molecular size decreased (Pavia etal., 1979; Helms et al., 2008). In contrast, the apparent emergenceor intensification of negative peaks at ~325, ~375, and ~430 nm inthe second derivative natural log spectra (Fig. 2F) could be due tothe formation of absorbing photoproducts and/or the preferentialdegradation of material with absorbance maxima adjacent to thoseof photo-refractory chromophores. In support of the latter, decreasinga254 (Fig. 3) indicates degradation of aromatic and/or highly conju-gated DOM moieties, which may give rise to selective losses at longerwavelengths via coupled charge transfer reactions (Del Vecchio andBlough, 2004a). Decreases in SUVA indicate that these moieties arebleached preferentially compared to the bulk DOC pool (Traina etal., 1990; Weishaar et al., 2003).

The observed increases in UV S and SR (Fig. 4) were similarto those reported for unconcentrated seawater upon irradiation(Stubbins et al., 2012), which further supports that changes occurringdue to irradiation of isolated DOM samples are similar, at least quali-tatively, to changes that occur in seawater. Notably, the current ex-perimental setup allowed us to bleach samples for 68 days andretain enough chromophoric material for spectrophotometric analy-sis using a conventional spectrophotometer and a 10 cm path length

cell. Whereas when natural deep waters were irradiated for 28 daysusing the same solar simulator aCDOM had already fallen below detec-tion limits at wavelengths >350 nm (Stubbins et al., 2012). Irradia-tion of the RO/ED concentrates in the current study thus enabled usto prolong irradiations until bleaching substantially exceeded that inStubbins et al. (2012) where the maximum recordable SR was ~4compared to the far more degraded 8.92 observed here (Table 1).Mechanistically, it has been suggested that the cross spectrum ofnatural sunlight and aCDOM results in the optical properties of photo-bleached CDOM as seen in rainwater and open ocean gyre surface wa-ters (Reche et al., 2000; Kieber et al., 2007) and is also consistent withthe optical properties observed at the end of our photodegradationexperiments.

Increases in UV S and SR have been shown to correlate with de-creases in average CDOM molecular weight for terrestrial waters(Helms et al., 2008). For our unextracted samples, S and SR decreasedwith depth at the North Pacific ALOHA site (Fig. 7.) In the same regionthe fraction of high molecular weight DOC recoverable by 1000 Dal-ton cutoff ultrafiltration membranes decreases with depth (Benner,2002). Thus the molecular weight information obtained from opticalparameters must be confined to the CDOM pool and not extrapolatedto the bulk DOC pool, since the distribution and cycling of the twopools in the open ocean are largely decoupled (Swan et al., 2009;Stedmon and Alvarez-Salgado, 2011). It is also possible that the rela-tionship between S and molecular weight observed elsewhere isdominated by the relative abundance of terrestrial DOM and there-fore has limited relevance to the open ocean. Other studies haverevealed CDOM optical properties to be strongly correlated withCDOM aromaticity (Weishaar et al., 2003), dissolved lignin (Spenceret al., 2009; Fichot and Benner, 2012), and dissolved black carbon(Stubbins et al., 2012). Condensed aromatics (dissolved black carbon)in the oceans are a significant fraction of the DOM pool and have sim-ilar distributions to those for CDOM, i.e., with highest concentrationsin terrestrially influenced surface waters, moderate values in the deepoceans, and lowest values in the photobleached subtropical gyres(Dittmar and Paeng, 2009; Nelson and Siegel, 2013). Trends indissolved black carbon match those for CDOM as both have a strongphotochemical sink (Stubbins et al., 2012). Further work is requiredto determine which pools of aromatic carbon, lignin, dissolved black

Loss of a300 (m-1) Loss of a300 (m-1)

Loss of a300 (m-1) Loss of a300 (m-1)

Slop

e R

atio

(S27

5-29

5:S 3

50-4

00)

275-295 nm350-400 nm 300-700 nm

y=0.003x+0.019

y=0.0001x+0.0093

y=-0.0006x+0.0092

A

y=2.084e0.2075x

B

y=-0.063x+0.92

D

DO

C (

μM)

SUV

A25

4 (m

2 g-C

-1)

y=-40.4x+770

C

0.04 10

8

6

4

2

0

1.2

0.8

0.4

0.0

0.03

0.02

0.01

00 1 2 3 4 5 6 7

0 1 2 3 4 5 6 7

0 1 2 3 4 5 6 7

0 1 2 3 4 5 6 7

1000

750

500

250

Spec

tral

Slo

pe (

nm-1

)

Fig. 5. (A) S300–700 changed little (r2 = 0.1003; p = 0.7), while S275–295 correlated positively (r2 = 0.994; p = 0.003) with photobleaching of a300 and S350–400 correlated negative-ly (r2 = 0.959; p = 0.02). (B) SR increased exponentially with photobleaching of a300 (r2 = 0.993; p = 0.004). (C) DOC (r2 = 0.924; p = 0.04) and (D) SUVA254 (r2 = 0.704;p = 0.161) also correlated with photobleaching of a300.

300 550500450400350

400

350

200

300 550500450400350

400

350

200

A) 0 d

B) 68 d450

450

300

300

250

250

Emission Wavelength (nm-1)

<0.001

0.002-0.005

0.007-0.01

>0.012

Exc

itatio

n W

avel

engt

h (n

m-1

)E

xcita

tion

Wav

elen

gth

(nm

-1)

Fig. 6. Fluorescence excitation emission matrix spectra shown for (A) non-irradiated674 m NELHA DOM isolate and (B) 674 m NELHA DOM that was irradiated for 68 days.Intensity scale is reported asWater Raman Units corrected for RO/ED concentration factorand DOC recovery.

87J.R. Helms et al. / Marine Chemistry 155 (2013) 81–91

carbon or others play quantitatively significant roles in determiningthe optical properties of deep ocean waters.

Irrespective of the physicochemical cause of the shift of absorbancetowards shorter wavelengths during photobleaching, 275–295 nm and350–400 nm are clearly identified as the most dynamic regions of thenatural log (Fig. 2D), 1st derivative natural log (Fig. 2E), and spectralslope spectra (Fig. 2I). Similar observations in estuarine systems under-lie thederivation, i.e. basis, of the SR concept (Helms et al., 2008). S275–295and SR are particularly sensitive to photobleaching (Figs. 4 and 5) andmay be useful indicators of photochemical histories of water masses ortracers of recently stratified surface oceanwaters, assuming that appro-priate end-members (e.g., strongly photobleached surface water vs.unbleached aged deep water) can be identified.

4.3. Photobleaching of FDOM

The shift of fluorescence to shorter wavelengths during irradiation(Table 2) suggests that molecular size, degree of conjugation, struc-tural rigidity, or capacity for intramolecular charge/energy transferwithin the constituents of the humic-like FDOM pool decreasedas a result of photobleaching (Pavia et al., 1979; Lakowicz, 1999;

Table 3Fluorescence indices (unitless) calculated based on EEMs.

Sample M:C FI HIX BIX

UnconcentratedALOHA — 5 m 1.70 1.55 0.92 0.98ALOHA — 125 m 1.42 1.54 1.38 1.38ALOHA — 770 m 1.17 1.77 1.80 0.88ALOHA — 3500 m 1.29 1.72 1.68 0.88NELHA — 674 m-SW 1.30 1.72 1.68 0.87

RO/ED isolatedALOHA — 5 m 1.21 1.59 1.82 0.89ALOHA — 3500 m 1.13 1.48 4.60 0.75NELHA — 674 m-0 daya 1.17 1.53 3.90 0.74NELHA — 674 m-68 daysa 1.43 1.61 1.40 0.97

a Photochemical bleaching experiment.

Slope Ratio (SR; unitless)

Dep

th (

m)

Fluor. Index (unitless)

M:CFIHIXBIX

Fluor. Intens. (WRU)

Dep

th (

m)

Dep

th (

m)

DOC (μM)

A B

C D

ACMBT

a300 (m-1)

Fig. 7. Depth profile of (A) Napierian absorption coefficient at 300 nm and DOC (inset),(B) spectral slope ratio, (C) fluorescence intensity, and (D) fluorescence indices mea-sured for non-irradiated 0.2 μm filtered water samples collected at station ALOHA(not isolated by RO/ED). WRU = Water Raman Units.

Emission Wavelength (nm-1)

300 550500450400350

400

350

300

A) ALOHA 5m

B) NELHA 670m

C) ALOHA 3500m

450

300 550500450400350

400

350

300

450

300 550500450400350

400

350

300

450

0.004-0.006

0.006-0.008

0.008-0.010

0.010-0.012

0.002-0.004

Exc

itatio

n W

avel

engt

h (n

m-1

)E

xcita

tion

Wav

elen

gth

(nm

-1)

Exc

itatio

n W

avel

engt

h (n

m-1

)

Fig. 8. Fluorescence excitation emission matrix spectra obtained for RO/ED isolatedDOM samples from (A) a surface sample collected at station ALOHA, (B) a samplepumped from the 674 m water supply at NELHA, and (C) a sample collected at3500 m depth at station ALOHA. Intensity is reported in Water Raman Units correctedfor RO/ED concentration factor.

88 J.R. Helms et al. / Marine Chemistry 155 (2013) 81–91

Del Vecchio and Blough, 2004a; Cory and McKnight, 2005). Whencompared to the absorbance photobleaching results, three fluores-cence trends are of interest. First, three of the four fluorescence indi-ces in the original 674 m RO/ED isolate shifted to values that are moreconsistent with those observed in surface waters (i.e., less consistentwith samples from the oxygenminimum layer or deep ocean). Specif-ically, M:C and BIX increased, while HIX decreased significantly(Table 3).

Second, it is interesting to note that the photobleached BIX valuewas comparable to the surface ALOHA sample, but remained consid-erably lower than the 125 m (chlorophyll max) ALOHA sample(Table 3). This result suggests that BIX may be a more robust sourceindicator in surface waters than the other indices because it is less im-pacted by photobleaching.

Third, interpreting the photobleaching increase observed in FI iscomplicated by the fact that it is intended as an indication of DOMsource (terrestrial vs. microbial fulvic acids) within terrestrial aquaticsystems (McKnight et al., 2001) and thus may have little meaning inremote marine waters. For example for unconcentrated ALOHA sea-water (Table 3: Fig. 7), the FI is markedly lower at 5 m and 125 mthan it is at 770 m and 3500 m. Thus, this difference does not appearto reflect recent autochthonous microbial sources, which are presum-ably much greater in surface waters and at the chlorophyll maximum.In addition, the FI increased upon irradiation of the 674 m RO/EDisolated NELHA sample (Table 3), which suggests that terrestrial-like fulvic substances may be preferentially destroyed during photo-bleaching (Helms, 2012). These results further support that the FI,

as conventionally applied may not be applicable to open ocean sam-ples and would need to be re-interpreted if it is to be used for thosewaters. We note, however, that other experiments have found, forexample, that mangrove-derived DOM approached a terrestrial end-member FI value (decreased from 1.65 to 1.5) upon simulatedsolar irradiation (Jaffé et al., 2004), that arctic Toolik Lake waterFI values decreased during natural sunlight exposure (Cory et al.,2007), and that subtropical coastal surface waters yielded more mi-crobial fulvic-like FI values than nearby estuarine surface waters(Maie et al., 2006). Thus, based on currently available evidence the

89J.R. Helms et al. / Marine Chemistry 155 (2013) 81–91

response of FI to DOM photobleaching appears to be stronglyinfluenced by the source, optical characteristics, and history of previ-ous light exposure of the starting material.

4.4. Implications for the cycling of CDOM, FDOM and recalcitrant DOM

Swan et al. (2009) found that CDOM absorption and DOC concen-tration do not significantly correlate in the Pacific Ocean, and thatprocesses governing CDOM and DOC distributions appear to be large-ly decoupled. Specifically, studies have found that surface gyre watershave high DOC, high protein-like FDOM, low aCDOM, and littlehumic-like FDOM, while deep and subsurface waters have low DOC,little protein-like FDOM, and higher aCDOM and humic-like FDOM(Swan et al., 2009; Yamashita and Tanoue, 2009; Stedmon andAlvarez-Salgado, 2011). By contrast, terrestrially dominated systemssuch as coastal waters and estuaries often show strong correlationsbetween aCDOM and DOC (Blough and Del Vecchio, 2002). The lightinduced removal of absorption and fluorescence as well as the atten-dant changes in optical properties observed here support the conclu-sion that the depletion of CDOM in the subtropical gyres is largelydue to photochemical degradation (Swan et al., 2009; Stedmon andAlvarez-Salgado, 2011). The residual photo-refractory and photo-produced DOM in the irradiated 674 m sample represented ~62% ofthe initial sample DOC (Table 1). This photochemical mineralizationof DOC suggests that the build-up of DOC in surface waters (Hansellet al., 2009; Hansell, 2013) must involve production of fresh algalDOC, which may, in part, be supported by photochemical release ofnutrients (Tarr et al., 2001; Stedmon et al., 2007). Assuming thatour irradiation experiment represents complete long-term photo-bleaching of mixed-layer DOM, the photo-refractory pool of DOC isroughly 27 μM and that near surface algal production of DOC mustovercome a photochemically driven deficit of nearly 40% of the back-ground refractory DOC (likely an underestimate) before any excesscould be measured. From our results it appears that the spectralslope at 275–295 nm and/or the SR may be used to help deconvolutethe relative proportions of photobleached vs. fresh microbial CDOMin surface ocean waters.

CDOM with optical properties similar to subsurface ocean DOMhas been shown to accumulate in cultures of aquatic microbial assem-blages (Tanoue et al., 1995; Rochelle-Newall and Fisher, 2002; Nelsonet al., 2004). It appears that a significant fraction of this “new” CDOMis biologically semi-labile (Nelson et al., 2004; Yamashita and Tanoue,2004, 2008, 2009; Nieto-Cid et al., 2006; Yamashita et al., 2007;Jørgensen et al., 2011), and based on the findings presented here, itis largely photo-labile as well. Photo-refractory DOM appears to con-sist largely of carbohydrates and carboxylated aliphatic material(Helms, 2012; Stubbins et al., 2010). Further, based on the relativelyslow degradation of EEM Peak B observed here, on the photo-stability of amides (Helms, 2012), and on the observed distributionof amide-nitrogen in the major ocean basins (McCarthy et al., 1997;Aluwihare et al., 2005), there is likely a significant peptide or proteincomponent within the photo-refractory pool (McCarthy et al., 1998;Mopper and Kieber, 2002; Yamashita and Tanoue, 2003a,b; Stubbinset al., 2013). Based on the vertical distributions of CDOM presentedin Nelson et al. (2007) and Swan et al. (2009) and FDOM describedin Yamashita and Tanoue (2008, 2009), Yamashita et al. (2007) andStedmon and Alverez-Salgado (2011), it appears that an importantsink for microbially produced CDOM in surface waters is a coupledphotochemical–biological process (Kieber et al., 1989; Miller andMoran, 1997; Moran et al., 2000), though the results presented hereshow that photobleaching alone can remove most of the CDOM andhumic-like FDOM in the surface mixed layer.

The photo-lability of CDOM depends on its composition. Terrestri-ally produced CDOM typically contains significantly more aromaticcarbon than algal produced CDOM (Sulzberger and Durisch-Kaiser,2009). As shown previously (Hayase et al., 1988; Chen and Bada,

1992; Minor et al., 2007; Dalzell et al., 2009; Spencer et al., 2009), ex-tensively photobleached terrestrial DOM resembles ocean DOM inboth bulk composition and in its optical properties (e.g., lower ab-sorption and fluorescence, higher spectral slope, and EEM Peak M be-coming larger than Peak C). EEM Peak C, while less intense than PeakM, was the most photo-labile fluorophore (Table 2). In fact, the dom-inance of Peak M relative to Peak C is seen in extensively bleachedterrestrial CDOM (Helms, 2012) and the results of the present studyindicate that a more intense Peak M relative to Peak C (e.g., higherM:C ratio) may be an indicator of extensive surface photobleachingas well as an indicator of algal or microbial source (Coble, 1996,2007; Helms, 2012).

4.5. Conclusions and future directions

Our findings showed that concentrating deep ocean DOM prior tolight exposure permitted significantly longer irradiation experimentswithout loss of detection for long-wavelength aCDOM, and thus accu-rate optical characterization of a more completely photodegradedsample than would be possible with seawater. The optical character-istics of extensively bleached DOM, including SUVA254, S, SR, and M:C,were comparable to surface waters. Photodegradation is a probablesink for recalcitrant deep ocean DOM characterized by shallow spec-tral slope and high percentage humic fluorescence. Optical indices,such as S275–295, SR, and M:C, represent potential indicators of previ-ous photochemical exposure in open ocean seawater.

Continued improvement in the molecular level and compositionalcharacterization of marine DOM with accompanying optical spectro-scopic characterization will provide greater understanding of howchanges in DOM composition impact the optical properties of CDOMand FDOM. A better understanding of these relationships has implica-tions for the use of in situ optical measurements (Spencer et al., 2007;Menon et al., 2011) and remote sensing technology (Carder et al.,1989; Del Vecchio and Subramaniam, 2004; Kutser et al., 2005; DelCastillo and Miller, 2008; Griffin et al., 2011) to track biogeochemicallyand ecologically important processes over comprehensive temporaland spatial scales.

The distributions of DOC, CDOM, and FDOM in the oceans have beenextensively investigated (Carlson and Ducklow, 1995; Guo et al., 1995;Aluwihare et al., 1997; Coble et al., 1998; Hansell and Carlson, 1998;Nelson and Siegel, 2002; Nelson et al., 2007; Yamashita et al., 2007;Yamashita and Tanoue, 2008, 2009; Swan et al., 2009), though the pro-cesses of their production and removal remain poorly understood. Theresults presented here help explain the role of photobleaching in thebuildup of low absorbance, high spectral slope DOM in the surfacewaters of the subtropical gyres (Nelson et al., 2007; Swan et al., 2009).This study has also shown (along with Helms, 2012) that, while EEMPeak M is associated with surface and sub-surface marine microbialproduction of FDOM (Rochelle-Newall and Fisher, 2002; Yamashitaand Tanoue, 2004), the photochemically induced shift of Peak C fluores-cence (e.g., from deep or near-shore waters) to shorter wavelengthsmay result in a significant source of Peak M fluorescence in surfaceoceanwaters (Helms, 2012). A greater awareness and further investiga-tion of these photochromic shifts in EEM spectra are needed to ensurethe proper interpretation of FDOM distributions in the upper ~400 mof the ocean.

Acknowledgments

The authors gratefully acknowledge the staff of the Natural EnergyLab of Hawaii (NELHA), and the crew of R/V Kilo Moana for assistancewith sample collection. Thanks also to David Burdige for allowing ususe of the fluorometer and Patrick G. Hatcher for the use of the TOCanalyzer. Patrick G. Hatcher, Jingdong Mao, Richard C. Zimmerman,Colin A. Stedmon, and two anonymous reviewers provided valuablecomments on earlier versions of this manuscript. This research

90 J.R. Helms et al. / Marine Chemistry 155 (2013) 81–91

was supported by the National Science Foundation grant numbersOCE-0728634 and OCE-0728050.

References

Aluwihare, L.I., Repeta, D.J., Chen, R.F., 1997. A major biopolymeric component todissolved organic carbon in surface sea water. Nature 387, 166–169.

Aluwihare, L.I., Repeta, D.J., Pantoja, S., Johnson, C.G., 2005. Two chemically distinctpools of organic nitrogen accumulate in the ocean. Science 308 (5724), 1007–1010.

Battin, T., 1998. Dissolved organic matter and its optical properties in a blackwater trib-utary of the upper Orinoco River, Venezuela. Org. Geochem. 28 (9/10), 561–569.

Benner, R., 2002. Chemical composition and reactivity. In: Hansell, D.A., Carlson, C.A.(Eds.), Biogeochemistry of Marine Dissolved Organic Matter. Academic Press,New York, pp. 59–90.

Birdwell, J.E., Engel, A.S., 2010. Characterization of dissolved organic matter in caveand spring waters using UV–vis absorbance and fluorescence spectroscopy. Org.Geochem. 41, 270–280.

Birdwell, J.E., Valsaraj, K.T., 2010. Characterization of dissolved organic matter infogwater by excitation–emission matrix fluorescence spectroscopy. Atmos. Envi-ron. 44 (27), 3246–3253.

Blough, N.V., Del Vecchio, R., 2002. Chromophoric DOM in the coastal environment. In:Hansell, D.A., Carlson, C.A. (Eds.), Biogeochemistry of Marine Dissolved OrganicMatter. Academic Press, San Diego, CA, pp. 508–545.

Burdige, D.J., Kline, S.W., Chen, W., 2004. Fluorescent dissolved organic matter in ma-rine sediment pore waters. Mar. Chem. 89, 289–311.

Carder, K.L., Steward, R.G., Harvey, G.R., Ortner, P.B., 1989. Marine humic and fulvicacids: their effects on remote sensing of ocean chlorophyll. Limnol. Oceanogr. 34(1), 68–81.

Carlson, C.A., Ducklow, H.W., 1995. Dissolved organic carbon in the upper ocean of thecentral Equatorial Pacific, 1992: daily and fine-scale vertical variations. Deep-SeaRes. II 42, 639–656.

Chen, R.F., Bada, J.L., 1992. The fluorescence of dissolved organic matter in seawater.Mar. Chem. 37 (3–4), 191–221.

Coble, P.G., 1996. Characterisation of marine and terrestrial DOM in seawater using ex-citation emission matrix spectroscopy. Mar. Chem. 51, 325–346.

Coble, P.G., 2007. Marine optical biogeochemistry: the chemistry of ocean color. Chem.Rev. 107 (2), 402–419.

Coble, P.G., Del Castillo, C.E., Avril, B., 1998. Distribution and optical properties of CDOMin the Arabian Sea during the 1995 Southwest Monsoon. Deep-Sea Res. II 45(10–11), 2195–2223.

Cory, R.M., McKnight, D.M., 2005. Fluorescence spectroscopy reveals ubiquitous pres-ence of oxidized and reduced quinones in dissolved organic matter. Environ.Sci. Technol. 39, 8142–8149.

Cory, R.M., McKnight, D.M., Chin, Y.-P., Miller, P., Jaros, C.L., 2007. Chemical character-istics of fulvic acids from Arctic surface waters: microbial contributions and photo-chemical transformations. J. Geophys. Res. Biogeosci. 112 (G4).

Cutter, G.A., Cutter, L.S., Filippino, K.C., 2004. Sources and cycling of carbonyl sulfide inthe Sargasso Sea. Limnol. Oceanogr. 49, 555–565.

Dalzell, B.J., Minor, E.C., Mopper, K.M., 2009. Photodegradation of estuarine dissolvedorganic matter: a multi-method assessment of DOM transformation. Org. Geochem.40, 243–257.

Del Castillo, C.E., Miller, R.L., 2008. On the use of ocean color remote sensing to measurethe transport of dissolved organic carbon by the Mississippi River Plume. RemoteSens. Environ. 112 (3), 836–844.

Del Vecchio, R., Blough, N.V., 2002. Photobleaching of chromophoric dissolved organicmatter in natural waters: kinetics and modeling. Mar. Chem. 78, 231–253.

Del Vecchio, R., Blough, N.V., 2004a. On the origin of the optical properties of humicsubstances. Environ. Sci. Technol. 38, 3885–3891.

Del Vecchio, R., Blough, N.V., 2004b. Spatial and seasonal distribution of chromophoricdissolved organic matter and dissolved organic carbon in the Middle Atlantic Bight.Mar. Chem. 89, 169–187.

Del Vecchio, R., Subramaniam, A., 2004. Influence of the Amazon River on the surfaceoptical properties of the western tropical North Atlantic Ocean. J. Geophys. Res.109, C11001.

Dittmar, T., Paeng, J., 2009. A heat induced molecular signature in marine dissolved or-ganic matter. Nat. Geosci. 2, 175–179.

Fichot, C.G., Benner, R., 2012. The spectral slope coefficient of chromophoric dissolvedorganic matter (S275–295) as a tracer of terrigenous dissolved organic carbon inriver-influenced ocean margins. Limnol. Oceanogr. 57 (5), 1453–1466.

Fuhrman, J.A., Azam, F., 1982. Thymidine incorporation as a measure of heterotrophicbacterioplankton production in marine surface waters: evaluation and field results.Mar. Biol. 66, 109–120.

Ghervase, L., Carstea, E.M., Pavelescu, G., Borisova, E., Daskalova, A., 2010. Fluorescenceevaluation of anthropogenic influence on rivers crossing Sofia. Rom. Rep. Phys. 62(1), 193–201.

Green, S.A., Blough, N.V., 1994. Optical absorption and fluorescence properties of chro-mophoric dissolved organic matter in natural waters. Limnol. Oceanogr. 39 (8),1903–1916.

Griffin, C.G., Frey, K.E., Rogan, J., Holmes, R.M., 2011. Spatial and interannual variabilityof dissolved organic matter in the Kolyma River, East Siberia, observed using satel-lite imagery. J. Geophys. Res. 116, G03018.

Guo, L., Santschi, P.H., Warnken, K.W., 1995. Dynamics of dissolved organic carbon(DOC) in oceanic environments. Limnol. Oceanogr. 40 (8), 1392–1403.

Hansell, D.A., 2013. Recalcitrant dissolved organic carbon fractions. Annu. Rev. Mar. Sci.5, 3.1–3.25.

Hansell, D.A., Carlson, C.A., 1998. Deep-ocean gradients in the concentration ofdissolved organic carbon. Nature 395, 263–266.

Hansell, D.A., Carlson, C.A., Repeta, D.J., Schlitzer, R., 2009. Dissolved organic matter inthe ocean: a controversy stimulates new insights. Oceanography 22 (4), 202–212.

Hayase, K., Shinozuka, N., 1995. Vertical distribution of fluorescent organic matteralong with AOU and nutrients in the equatorial Central Pacific. Mar. Chem. 48,283–290.

Hayase, K., Tsubota, H., Sunada, I., Goda, S., Yamazaki, H., 1988. Vertical distribution offluorescent organic matter in the North Pacific. Mar. Chem. 25 (4), 373–381.

Helms, J.R., 2012. Spectroscopic Characterization of Dissolved Organic Matter: InsightsInto Composition, Photochemical Transformation and Carbon Cycling. DoctoralThesis Old Dominion University, Norfolk, VA 228.

Helms, J.R., et al., 2008. Absorption spectral slopes and slope ratios as indicators of mo-lecular weight, source, and photobleaching of chromophoric dissolved organicmatter. Limnol. Oceanogr. 53 (3), 955–969.

Huguet, A., et al., 2009. Properties of fluorescent dissolved organic matter in theGironde Estuary. Org. Geochem. 40, 706–719.

Jaffé, R., et al., 2004. Source characterization of dissolved organic matter in asubtropical mangrove-dominated estuary by fluorescence analysis. Mar. Chem.84, 195–210.

Johannessen, S.C., Miller, W.L., 2001. Quantum yield for the photochemical productionof dissolved inorganic carbon in seawater. Mar. Chem. 76, 271–283.

Jørgensen, L., et al., 2011. Global trends in the fluorescence characteristics and distribu-tion of marine dissolved organic matter. Mar. Chem. 126, 129–148.

Kalbitz, K., Geyer, W., Geyer, S., 1999. Spectroscopic properties of dissolved humicsubstances: a reflection of land use history in a fen area. Biogeochemistry 47 (2),219–238.

Kieber, D.J., McDaniel, J.A., Mopper, K., 1989. Photochemical source of biological sub-strates in seawater: implications for carbon cycling. Nature 341, 637–639.

Kieber, R.J., Zhou, X., Mopper, K., 1990. Formation of carbonyl compounds from UV-Induced photodegradation of humic substances in natural waters: fate of riverinecarbon in the sea. Limnol. Oceanogr. 35, 1503–1515.

Kieber, R.J., Whitehead, R.F., Skrabal, S.A., 2006. Photochemical production of dis-solved organic carbon from resuspended sediments. Limnol. Oceanogr. 51 (5),2187–2195.

Kieber, R.J., Willey, J.D., Whitehead, R.F., Reid, S.N., 2007. Photobleaching of chromophoricdissolved organic matter (CDOM) in rainwater. J. Atmos. Chem. 58, 219–235.

Kitidis, V., et al., 2006. Variability of chromophoric organic matter in surface waters ofthe Atlantic Ocean. Deep-Sea Res. II 53, 1666–1684.

Koprivnjak, J.-F., Pfromm, P.H., Perdue, E.M., 2006. Coupling reverse osmosis withelectrodialysis to isolate natural organic matter from fresh waters. Water Res. 40,3385–3392.

Koprivnjak, J.-F., et al., 2009. Chemical and spectroscopic characterization of marinedissolved organic matter isolated using coupled reverse osmosis–electrodialysis.Geochim. Cosmochim. Acta 73, 4215–4231.

Kutser, T., et al., 2005. Using satellite remote sensing to estimate the colored dissolvedorganic matter absorption coefficient in lakes. Ecosystems 8, 709–720.

Lakowicz, J.R., 1999. Principles of Fluorescence Spectroscopy, 2nd ed. Kluwer Academic,New York.

Lawaetz, A.J., Stedmon, C.A., 2009. Fluorescence intensity calibration using Raman scat-ter peak of water. Appl. Spectrosc. 63 (8), 936–940.

Leifer, A., 1988. The Kinetics of Environmental Aquatic Photochemistry: Theory andPractice. American Chemical Society, Washington D.C.

Loiselle, S.A., et al., 2009. Optical characterization of chromophoric dissolved organicmatter using wavelength distribution of absorption spectral slopes. Limnol.Oceanogr. 54 (2), 590–597.

Maie, N., et al., 2006. Chemical characteristics of dissolved organic nitrogen in an oligo-trophic subtropical coastal system. Geochim. Cosmochim. Acta 70, 4491–4506.

McCarthy, M., Pratum, T., Hedges, J., Benner, R., 1997. Chemical composition ofdissolved organic nitrogen in the ocean. Nature 390 (6656), 150–154.

McCarthy, M., Hedges, J., Benner, R., 1998. Major bacterial contribution to marinedissolved organic nitrogen. Science 281, 231–234.

McKnight, D.M., et al., 2001. Spectrofluorometric characterization of dissolved organicmatter for indication of precursor organic material and aromaticity. Limnol.Oceanogr. 46 (1), 38–48.

Menon, H.B., Sangekar, N.P., Lotliker, A.A., Vetamony, P., 2011. Dynamics of chromo-phoric dissolved organic matter in Mandovi and Zuari estuaries — a study throughin situ and satellite data. ISPRS J. Photogramm. Remote. Sens. 66 (4), 545–552.

Miller, W.L., Moran, M.A., 1997. Interaction of photochemical and microbial processesin the degradation of refractory dissolved organic matter from a coastal marine en-vironment. Limnol. Oceanogr. 42 (6), 1317–1324.

Miller, W.L., Moran, M.A., Sheldon, W.M., Zepp, R.G., Opsahl, S., 2002. Determination ofapparent quantum yield spectra for the formation of biologically labile photoprod-ucts. Limnol. Oceanogr. 47 (2), 343–352.

Minor, E.C., Dalzell, B.J., Stubbins, A., Mopper, K., 2007. Evaluating the photoalterationof estuarine dissolved organic matter using direct temperature-resolved massspectrometry and UV–visible spectroscopy. Aquat. Sci. 69, 440–455.

Mopper, K., Kieber, D.J., 2002. Photochemistry and the cycling of carbon, sulfur, nitro-gen and phosphorus. In: Hansell, D.A., Carlson, C.A. (Eds.), Biogeochemistry of Ma-rine Dissolved Organic Matter. Academic Press, San Diego, CA, pp. 455–507.

Moran, M.A., Sheldon, W.M., Zepp, R.G., 2000. Carbon loss and optical property changesduring long term photochemical and biological degradation of estuarine organicmatter. Limnol. Oceanogr. 45 (6), 1254–1264.

Nelson, N.B., Siegel, D.A., 2002. Chromophoric DOM in the open ocean. In: Hansell, D.A.,Carlson, C.A. (Eds.), Biogeochemistry of Marine Dissolved Organic Matter. AcademicPress, San Diego, CA, pp. 547–578.

91J.R. Helms et al. / Marine Chemistry 155 (2013) 81–91

Nelson, N.B., Siegel, D.A., 2013. Global distribution and dynamics of chromophoricdissolved organic matter. Annu. Rev. Mar. Sci. 5, 447–467.

Nelson, N.B., Carlson, C.A., Steinberg, D.K., 2004. Production of chromophoric dissolvedorganic matter by Sargasso Sea microbes. Mar. Chem. 89, 273–287.

Nelson, N.B., et al., 2007. Hydrography of chromophoric dissolved organic matter in theNorth Atlantic. Deep-Sea Res. I 54, 710–731.

Nieto-Cid, M., Alverez-Salgado, X.A., Perez, F.F., 2006. Microbial and photochemicalreactivity of fluorescent dissolved organic matter in a coastal upwelling system.Limnol. Oceanogr. 51, 1391–1400.

Opsahl, S., Benner, R., 1998. Photochemical reactivity of dissolved lignin in river andocean waters. Limnol. Oceanogr. 43 (6), 1297–1304.

Ortega-Retuerta, E., Reche, I., Pulido-Villena, E., Austi, S., Duarte, C., 2010. Distributionand photoreactivity of chromophoric dissolved organic matter in the AntarcticPeninsula (Southern Ocean). Mar. Chem. 118, 129–139.

Para, J., et al., 2010. Fluorescence and absorption properties of chromophoric dissolvedorganic matter CDOM in coastal surface waters of the northwestern MediterraneanSea, influence of the Rhone River. Biogeosciences 7, 4083–4103.

Pavia, D.L., Lampman, G.M., Kriz, G.S., 1979. Introduction to Spectroscopy: A Guide forStudents of Organic Chemistry. Saunders, Philadelphia.

Reche, I., Pace, M.L., Cole, J.J., 2000. Modeled effects of dissolved organic carbon andsolar spectra on photobleaching in lake ecosystems. Ecosystems 3, 419–432.

Rochelle-Newall, E.J., Fisher, T.R., 2002. Production of chromophoric dissolved organicmatter fluorescence in marine and estuarine environments: an investigation intothe role of phytoplankton. Mar. Chem. 77, 7–21.

Romera-Castillo, C., Sarmento, H., Alverez-Salgado, X.A., Gasol, J.M., Marrase, C., 2011.Net production and consumption of fluorescent colored dissolved organic matterby natural bacterial assemblages growing on marine phytoplankton exudates.Appl. Environ. Microbiol. 77 (21), 7490–7498.

Skoog, A., Wedborg, M., Fogelqvist, E., 1996. Photobleaching of fluorescence and the or-ganic carbon concentration in a coastal environment. Mar. Chem. 55, 333–345.

Smith, D.C., Azam, F., 1992. A simple economical method for measuring bacterial pro-tein synthesis rates in sea water using 3H-leucine. Mar. Microb. Food Webs 6,107–114.

Spencer, R.G.M., Pellerin, B.A., Bergamaschi, B.A., 2007. Diurnal variability in riverinedissolved organic matter composition determined by in situ optical measurementin the San Joaquin River (California, USA). Hydrol. Process. 21, 3181–3189.

Spencer, R.G.M., et al., 2009. Photochemical degradation of dissolved organic matterand dissolved lignin phenols from the Congo River. J. Geophys. Res. 114, G03010.

Stedmon, C.A., Alvarez-Salgado, X.A., 2011. Shedding light on a black box: UV–visiblespectroscopic characterization of marine dissolved organic matter. In: Jiao, N.,Azam, F., Sanders, S. (Eds.), Microbial Carbon Pump in the Ocean. American Asso-ciation for the Advancement of Science, Washington, DC, pp. 62–63.

Stedmon, C.A., et al., 2007. Photochemical production of ammonium and transforma-tion of dissolved organic matter in the Baltic Sea. Mar. Chem. 104, 227–240.

Stubbins, A., Dittmar, T., 2012. Low volume quantification of dissolved organic carbonand dissolved nitrogen. Limnol. Oceanogr. Methods 10, 347–352.

Stubbins, A., et al., 2006. Open-ocean carbon monoxide photoproduction. Deep-SeaRes. II 53, 1695–1705.

Stubbins, A., Spencer, R.G.M., Chen, H., Hatcher, P.G., Mopper, K., Hernes, P.J., Mwamba,V.L., Mangangu, A.M., Wabakanghanzi, J.N., Six, J., 2010. Illuminated darkness: mo-lecular signatures of Congo River dissolved organic matter and its photochemicalalteration as revealed by ultrahigh precision mass spectrometry. Limnol. Oceanogr.55 (4), 1467–1477.

Stubbins, A., Niggemann, J., Dittmar, T., 2012. Photo-lability of deep ocean dissolvedblack carbon. Biogeosciences 9, 1661–1670.

Stubbins, A., del Giorgio, P., Berggren, M., Lapierre, J.F., Dittmar, T., 2013. What's in anEEM? Molecular signatures associated with dissolved organic fluorophores. ASLO2013 Aquatic Sciences Meeting, New Orleans, LA USA.

Sulzberger, B., Durisch-Kaiser, E., 2009. Chemical characterization of dissolved organicmatter (DOM): a prerequisite for understanding UV-induced changes of DOM ab-sorption properties and bioavailability. Aquat. Sci. 71 (2), 104–126.

Summers, R.S., Cornel, P.K., Roberts, P.V., 1987. Molecular size distribution and spectro-scopic characterization of humic substances. Sci. Total Environ. 62, 27–37.

Swan, C.M., Siegel, D.A., Nelson, N.B., Carlson, C.A., Nasir, E., 2009. Biogeochemical andhydrographic controls on chromophoric dissolved organic matter distribution inthe Pacific Ocean. Deep-Sea Res. I 56, 2175–2192.

Tanoue, E., Nishiyama, S., Kamo, M., Tsugita, A., 1995. Bacterial membranes: possiblesource of a major dissolved protein in seawater. Geochim. Cosmochim. Acta 59(12), 2643–2648.

Tarr, M.A., Wang, W., Bianchi, T.S., Englehaupt, E., 2001. Mechanisms of ammonia andamino acid photoproduction from aquatic humic and colloidal matter. Water Res.35 (15), 3688–3696.

Toole, D.A., Slezak, D., Kiene, R.P., Kieber, D.J., Siegel, D.A., 2006. Effects of solar radia-tion on dimethylsulfide cycling in the western Atlantic Ocean. Deep-Sea Res. I 53,136–153.

Traina, S.J., Novak, J., Smeck, N.E., 1990. An ultraviolet absorbance method of estimatingthe percent aromatic carbon content of humic acids. J. Environ. Qual. 19, 151–153.

Twardowski, M.S., Donaghay, P.L., 2002. Photobleaching of aquatic dissolved materials:absorption removal, spectral alteration and their interrelationship. J. Geophys. Res.107 (C8), 6-1–6-12. http://dx.doi.org/10.1029/1999JC000281 (issn 0148-0227).

Twardowski, M.S., Boss, E., Sullivan, J.M., Donaghay, P.L., 2004. Modeling the spectralshape of absorption by chromophoric dissolved organic matter. Mar. Chem. 89,69–88.

Vähätalo, A.V., Järvinen, M., 2007. Photochemically produced bioavailable nitrogenfrom biologically recalcitrant dissolved organic matter stimulates production ofnitrogen-limited microbial food web in the Baltic Sea. Limnol. Oceanogr. 52 (1),132–143.

Weishaar, J.L., et al., 2003. Evaluation of specific ultraviolet absorbance as an indicatorof the chemical composition and reactivity of dissolved organic carbon. Environ.Sci. Technol. 37, 4702–4708.

Williamson, C.E., Neale, P.J., Grad, G., De Lange, H.T., Hargreaves, B.R., 2001. Beneficialand detrimental effects of UV on aquatic organisms: implications of spectral varia-tion. Ecol. Appl. 11 (6), 1843–1857.

Yamashita, Y., Tanoue, E., 2003a. Chemical characterization of protein-like fluorophoresin DOM in relation to aromatic amino acids. Mar. Chem. 82, 255–271.

Yamashita, Y., Tanoue, E., 2003b. Distribution and alteration of amino acids in bulkDOM along a transect from bay to oceanic waters. Mar. Chem. 82, 145–160.

Yamashita, Y., Tanoue, E., 2004. In situ production of chromophoric dissolved organicmatter in coastal environments. Geophys. Res. Lett. 31, L24302.

Yamashita, Y., Tanoue, E., 2008. Production of bio-refractory fluorescent dissolved or-ganic matter in the ocean interior. Nat. Geosci. 1, 579–582.

Yamashita, Y., Tanoue, E., 2009. Basin scale distribution of chromophoric dissolved or-ganic matter in the Pacific Ocean. Limnol. Oceanogr. 54, 598–609.

Yamashita, Y., Tsukasaki, A., Nishida, T., Tanoue, E., 2007. Vertical and horizontal distri-bution of fluorescent dissolved organic matter in the Southern Ocean. Mar. Chem.106, 498–509.

Zepp, R.G., 2002. Solar ultraviolet radiation and aquatic carbon, nitrogen, sulfur andmetals cycles. In: Helbling, E.W., Zagarese, H. (Eds.), UV Effects in Aquatic Organ-isms and Ecosystems. Royal Society of Chemistry, Cambridge, UK, pp. 137–183.

Zepp, R.G., Sheldon, W.M., Moran, M.A., 2004. Dissolved organic fluorophores in south-eastern US coastal waters: correction method for eliminating Rayleigh and Ramanscattering peaks in excitation–emission matrices. Mar. Chem. 89, 15–36.

Zika, R.G., 1980. The sunlight induced decay of seawater light absorbance and fluores-cence. EOS 61, 1010.

Zsolnay, A., 2003. Dissolved organic matter: artefacts, definitions, and functions. Geoderma113 (3–4), 187–209.