remote sensing of environment -...

Remote Sensing of Environment xxx (2011) xxx–xxx

RSE-08081; No of Pages 12

Contents lists available at SciVerse ScienceDirect

Remote Sensing of Environment

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

Relationships between suspended mineral concentrations and red-wavebandreflectances in moderately turbid shelf seas

Claire Neil, Alex Cunningham ⁎, David McKeePhysics Department, University of Strathclyde, 107 Rottenrow, Glasgow G40NG, UK

⁎ Corresponding author. Tel.: +44 141 548 3474; faxE-mail address: [email protected] (A. Cunn

0034-4257/$ – see front matter © 2011 Elsevier Inc. Alldoi:10.1016/j.rse.2011.09.010

Please cite this article as: Neil, C., et al., Relreflectances in moderately turbid shelf seas

a b s t r a c t
a r t i c l e i n f o
Article history:Received 12 August 2010Received in revised form 11 August 2011Accepted 9 September 2011Available online xxxx

Keywords:Shelf seasSuspended mineral particlesAlgorithm derivation

This paper considers the uncertainties that arise in estimating the concentration of suspended minerals byoptical remote sensing in waters which contain unknown concentrations of other optically significant con-stituents. Relationships between suspended mineral concentrations and remote sensing reflectance were cal-culated by radiative transfer modelling using representative specific inherent optical properties (SIOPs) forphytoplankton (CHL), suspended mineral particles of terrigenous origin (MSSter) and coloured dissolved or-ganic matter (CDOM) that were derived from measurements at 173 stations in UK shelf seas. When only sus-pended minerals were present, remote sensing reflectance (Rrs) was related to MSSter by a family ofsaturation curves whose shape depended strongly on wavelength. However the addition of CHL and CDOMmade this relationship considerably more complex. Polynomial expressions were therefore derived for themaximum and minimum values of MMSter consistent with a given Rrs667 in the presence of independentlyvarying concentrations of CHL and CDOM. For CHL ranging from 0 to 10 mg m−3 and CDOM from 0 to1 m−1, for example, an Rrs667 observation 0.01 sr−1 could corresponded to MSSter values between 7 and12 g m−3. The presence of biogenic minerals in the form of diatom frustules, MSSdia had little influence onthe accuracy of MSSter retrieval. The degree of variability in the relationship between MSSter and Rrs667 pre-dicted by the model was confirmed by measurements of radiometric profiles and mineral concentrations at110 Irish Sea stations. Uncertainties in the remote sensing of MSSter in coastal waters are more appropriatelyindicated by upper and lower limits set according to the likely ranges of other optically significant constitu-ents than by percentage errors. Moreover, the influence of these constituents should be eliminated beforevariations in the relationship between MSSter and Rrs are attributed to qualitative changes in mineral particlecharacteristics.

: +44 141 552 2891.ingham).

rights reserved.

ationships between suspended mineral conce, Remote Sensing of Environment (2011), doi:1

© 2011 Elsevier Inc. All rights reserved.

1. Introduction

Variations in the optical reflectance of shelf seas are generated bychanges in the concentrations of three classes of optically significantconstituents (OSCs): phytoplankton cells, suspended mineral parti-cles, and dissolved coloured organic matter (Kirk, 1994). Optical re-mote sensing has great potential value for mapping water-qualityindicators in these waters (Hu et al., 2004), but attempts to deviseempirical algorithms that are capable of simultaneously deriving theconcentrations of all three constituents have encountered consider-able difficulties (Robinson, 2008; Sathyendranath, 2000). Semi-analytical approaches based on neural networks (Doerffer & Schiller,2007; Schiller & Doerffer, 1999; Tanaka et al., 2004) or spectralmatching (Kuchinke et al., 2009a, b; Lee et al., 1999; Maritorenaet al., 2002; van der Woerd & Pasterkamp, 2008; Zawada et al.,2007) offer substantially improved performance, but rely heavily on

the quality of the spectral libraries which they employ in look-up tables(in the case of spectral matching) or as training data sets (for neuralnetworks). The recovery of suspended mineral concentrations withoutaccurate estimates of the other optically significant constituentswould appear to be a simpler problem, and single-band reflectance al-gorithms operating in the visible region have been proposed for thispurpose (Binding et al., 2003, 2010; Nechad et al., 2010; Sipelgaset al., 2009; Tzortziou et al., 2007). It is necessary, however, to draw adistinction between waters whose optical properties are dominatedby suspended mineral particles and those where suspended mineralsexert a significant influence but are not present in sufficiently high con-centrations tomask the contribution of the other components. Methodsfor quantifying suspended mineral concentrations in the first, highlyturbid, category have recently been reviewed by Hommersom et al.(2010). To date, however, there have been no systematic studies ofthe performance of suspendedmineral algorithms inmoderately turbidwaters where the interfering effects of non-mineral seawater constitu-ents are likely to be significant. Themoderately turbid category includesa large proportion of the world's shallow shelf seas (Shi &Wang, 2010).This paper therefore presents a case study of a semi-enclosed sea and

ntrations and red-waveband0.1016/j.rse.2011.09.010

http://dx.doi.org/10.1016/j.rse.2011.09.010

mailto:[email protected]

Unlabelled image


Unlabelled image

http://www.sciencedirect.com/science/journal/00344257


2 C. Neil et al. / Remote Sensing of Environment xxx (2011) xxx–xxx

adjacent estuary (the Irish Sea and Bristol Channel) whose waters ex-hibit marked spatial and seasonal variability in remote sensing reflec-tance which is largely driven by changes in the concentrations ofsuspended mineral particles (Binding et al., 2003; Bowers, 2003). Theaim is not to propose a definitive algorithm for recovering suspendedmineral concentrations, but to investigate how the optical interactionsbetween pure seawater and its constituents constrain the performanceof such algorithms in moderately turbid coastal waters.

2. Methods

2.1. Background theory

Morel and Prieur (1977) suggested that, in cases where inelasticscattering can be neglected, irradiance reflectance (R) just belowthe surface can be related to the backscattering coefficient (bb) andabsorption coefficient (a) by an expression of the general form

R ¼ Eu wð ÞEd wð Þ ¼ fL;β

bba

ð1Þ

where Eu(w) is the upward planar irradiance, Ed (w) the downwardplanar irradiance and the factor fL,β is a variable function of the radi-ance distribution and volume scattering function. Values for fL,β aredifficult to derive from first principles, but early studies usingMonte Carlo techniques (Kirk, 1981, 1994; Morel & Prieur, 1977) sug-gested that Eq. (1) could be approximated by

R≈C μ 0ð Þ bba

ð2Þ

where C(μ0) is a function of the mean cosine of the photon distribu-tion just below the surface. The value of C(μ0) is therefore stronglydependent on the zenith angle of the refracted solar beam, and isalso influenced by the volume scattering function. Converting subsur-face irradiance reflectance, R, to remote sensing reflectance, Rrs, re-quires the introduction of additional terms:

Rrs ¼Lw að ÞEd að Þ ¼

ΓQR ð3Þ

where Lw(a) and Ed (a) are the vertical water-leaving radiance anddownward irradiance in air just above the surface, Q is the ratio of up-welling irradiance to radiance just below the surface, and Γ representsthe optical processes involved in light crossing the air-water interface(Mobley, 1999). Combining Eqs. (2) and (3) gives

Rrs≈C μoð ÞΓ

Q

� �bba

ð4Þ

For a sufficiently restricted set of conditions, which includes afixed solar angle, cloud-free sky, and limited range of IOP values, theterms enclosed in brackets in Eq. (4) are relatively invariant (Kirk,1994; Mobley, 1999). The notation can then be simplified by aggre-gating them as a single parameter:

Rrs≈κbba

ð5Þ

However κ is not strictly a constant of proportionality, and the de-gree to which it can be treated as such varies from case to case andhas to be established numerically. The values of the backscattering andabsorption coefficients (bb and a) depend on the concentrations of themain optically significant constituents in the water column, and theseare commonly quantified by measuring three proxy variables: the con-centration of solvent-extracted chlorophyll a (CHL) for phytoplankton;the absorption (relative to pure water) of membrane-filtered seawater

Please cite this article as: Neil, C., et al., Relationships between suspendreflectances in moderately turbid shelf seas, Remote Sensing of Environm

at 440 nm (CDOM) for coloured dissolved organic matter; and thetotal dry weight of material retained on a glass fibre filter after combus-tion (MSST) for suspended minerals. Total inherent optical propertiescan be calculated by multiplying the concentration of each constituentby the appropriate specific inherent optical property (SIOP), so that fora given wavelength (λ)

bb λð Þ ¼ b�b W λð Þ þ b�bMSSTλð ÞMSST þ b�b CHL λð ÞCHL ð6Þ

a λð Þ ¼ aW λð Þ þ a�MSSTλð ÞMSST þ a�CHL λð ÞCHLþ a�CDOM λð ÞCDOM ð7Þ

where asterisks indicate SIOPs, and subscripts indicate constituents(with W for pure seawater). Combining Eqs. (5), (6) and (7), anddropping the notation of wavelength dependence for brevity:

Rrs≈κbb W þ b�bMSST

MSST þ b�b CHLCHL

aW þ a�MSSTMSST þ a�CHLCHLþ a�CDOMCDOM

ð8Þ

Eq. (8) emphasises the point that the relationship between Rrs andthe OSM concentrations depends on the ratio of the total backscatter-ing and absorption coefficients, not the backscattering to absorptionratios of the individual constituents, and illustrates the importanceof the SIOPs in formulating this relationship. Unfortunately, very fewcomplete sets of the relevant SIOPS exist and none have been pub-lished for European shelf seas.

2.2. Study location

Measurements of constituent concentrations and inherent opticalproperties were carried out at 173 stations, occupied between 2000and 2006, at the positions plotted in Fig. 1a. Since chlorophyll concen-trations above 4 mg m−3 were rarely encountered in the Irish Sea andBristol Channel during this survey, data from adjacent Scottish westcoast waters (Fig. 1b.) were included to obtain adequate estimatesof phytoplankton optical properties. Radiometer casts for validatingthe results obtained by radiative transfer modelling were made at asubset of the stations occupied, and these stations are indicated byfilled symbols in Fig. 1.

2.3. In situ measurements

Temperature and salinity profiles were measured using a SeaBirdSBE 19 CTD. Non-water absorption and beam attenuation coefficients(anw and cnw) were measured in nine wavebands (412, 440, 488, 510,532, 555, 650, 676 and 715 nm) using WET Labs ac-9 or ac-9 plus in-struments which were calibrated using Milli-Q ultra pure water. Theac-9 data were corrected for differences in temperature and salinitybetween the seawater samples and the pure water used for calibra-tion according to the procedures described in the in ac-9 protocolmanual (WET Labs) using coefficients from Sullivan et al., 2006 andCTD profiles obtained at the same time as the ac-9 deployments. Ab-sorption measurements were subsequently corrected for the incom-plete collection of scattered light using the ‘proportional’ method ofZaneveld et al., 1994, and non-water scattering coefficients (bnw)were derived by subtracting anw from cnw. The Zaneveld correctionmethod sets anw to zero at 715 nm, and assumes that absorption bysuspended particles and dissolved substances can be neglected inthe near infra-red waveband. Whilst some early studies supportedthis assumption (Babin & Stramski, 2002, 2004; Stramski & Wozniak,2004), its validity is increasingly being questioned (Leymarie et al.,2010; Tassan & Ferrari, 2003; Tzortziou et al., 2006). Monte Carlomodelling of the effect of significant infra-red absorption by mineralparticles on ac-9 correction procedures (Leymarie et al., 2010) sug-gested that anw could be under-estimated by up to 5% at 412 nmand 50% at 676 nm in waters whose optics were dominated bythese particles. However this analysis was based on the assumption

ed mineral concentrations and red-wavebandent (2011), doi:10.1016/j.rse.2011.09.010


Fig. 1. Maps of station positions in the Irish Sea and Bristol Channel (a.) and off the west coast of Scotland (b.): filled symbols indicate stations where in profiling radiometry wascarried out.

3C. Neil et al. / Remote Sensing of Environment xxx (2011) xxx–xxx

that the absorption coefficient of non-algal particles at 750 nm was25% of its value at 440 nm, which is at the upper end of the rangereported by Tassan and Ferrari (2003) for marine sediments and con-siderably higher than the 5% suggested by Tzortziou et al. (2006). It isclear that errors in anw arising from the Zaneveld ac-9 correction pro-cedure are most pronounced in the red and near infra-red wave-bands, but there is continuing uncertainty regarding theirmagnitude. The impact of these errors on calculations of remote sens-ing reflectance is mitigated by the dominant role played by water ab-sorption in this spectral region (see Section 3.3).

Backscattering coefficients (bb) in two wavebands centred on470 nm and 676 nmwere measured using a HOBI Labs HS-2 backscat-tering meter. Compensation for absorption and scattering was ap-plied using the ‘sigma correction method’ recommended by themanufacturer with coefficients calculated from the anw and bnw valuesrecorded by the ac-9 during each deployment. Particulate backscat-tering coefficients (bbp) were obtained by subtracting the water con-tribution, which was taken to be half the total scattering valuesmeasured by Smith and Baker (1981). Since the HS-2 measured back-scattering in only two wavebands, it was necessary to extrapolate be-tween them to obtain values for radiative transfer modelling.Deployment of the HS-2 alongside a Wet Labs ECO BB9 instrumentis subsequent cruises confirmed that linear extrapolation betweenthe HS-2 channels at 470 nm and 676 nm gave values within+/−20% of those measured by the BB9 within this wavelength range.

In situ radiometric data were collected using a Satlantic SPMR pro-filing radiometer which measured downwards planar irradiance (Ed)and upward radiance (Lu) in 10 nm wavebands centred on 412, 443,490, 510, 554, 665 and 700 nm. SPMR casts were carried out at least20 m from the ship to avoid shadow effects, and two casts were aver-aged for each station. It was not possible to make all the measure-ments under conditions suitable for satellite remote sensing, andstations with significant cloud cover and occasionally low solar angleswere included in the data set. Values for water-leaving radiance andabove-surface downward irradiance, which are required for calcula-tions of remote sensing reflectance, were derived by extrapolatingSMPR profiles through the air-sea interface using the expressionsquoted in the Satlantic ProSoft manual. However since the available


version of this code (ProSoft 7.7) contained an error in handlingnear-surface data, profiles were processed using Matlab scripts writ-ten for the purpose.

2.4. Concentrations of optically significant constituents

For CHL determinations, which were carried out in triplicate,water samples were filtered through 25 mm GF/F filters and immedi-ately frozen. Phytoplankton pigments were extracted by soaking thefilters for 24 h in neutralised acetone at 4 °C, followed by centrifuga-tion andmeasurement of the absorbance spectrum of the supernatantrelative to an acetone blank in 1 cm path length cuvettes. The mea-surement was repeated after acidification with dilute hydrochloricacid. Chlorophyll a concentrations were then calculated using the tri-chromatic equations of Jeffrey and Humphrey (1975). Suspended par-ticles were collected by filtering 5 l of seawater through pre-weighed90 mm GF/F filters and rinsing with 500 ml of distilled water. The fil-ters were stored frozen until returned to the laboratory, where theywere dried to a constant weight in an oven at 100 °C and reweighedto obtain the concentration of total suspended solids (TSS). Values oftotal mineral suspended solids (MSST) were obtained by reweighingthe filters after they had been placed in a furnace at 500 °C for 3 h, atwhich point it was assumed that all organic materials had been com-busted. For determinations of CDOM, seawater samples were filteredthrough 0.2 μm membrane filters, with the filtrate being collected inacid-rinsed glass bottles and stored under refrigeration. Absorption byCDOM was measured in a spectrophotometer using 10 cm cuvetteswith UV treated ultrapure water as a reference, with care being takento allow the sample and reference to reach the same temperature beforethe measurements were carried out.

2.5. Specific scattering and backscattering coefficients

There is no currently available method for physically separatingthe contributions of different types of particles to the overall scatter-ing coefficients of natural hydrosols. However by plotting MSSTagainst CHL for all stations (Fig. 2), it was possible to identify groupsof points characterised by relatively high CHL and lowMSST (Group 1)


image of Fig.�1


Fig. 2. Total suspended mineral concentrations (MSST) plotted against chlorophyll con-centrations (CHL) for all stations in Fig. 1. Data were classified as Cluster 1 (high CHL:lowMSST, indicated by triangles), Cluster 2 (low CHL: highMSST indicated by open circles)and Cluster 3 (mixed composition, indicated by filled circles).

Fig. 4. Stacked bar charts showing bimodal distributions for the ratios of the coefficients ofparticulate backscattering to total particulate scattering at 676 nm (bbp676/bp676, panela.) and of non-water absorption at 650 nm to 676 nm (anw650/anw676, panel b). Stationswith high diatom numbers, indicated by filled symbols in Fig. 3., are plotted as shadedbars.


and by low CHL accompanied by high MSST (Group 2). Stationsfalling into these groups were selected by setting thresholds ofCHL N4 mg m−3 for Group 1 and MSST N4 g m−3 for Group 2. Group1 stations were located mainly in deep Scottish sea lochs, which gen-erally have low concentrations of suspended sediment, whilst thosein Group 2 were from relatively shallow, coastal areas of the IrishSea and Bristol Channel with high suspended sediment concentra-tions. It was assumed therefore that the scattering properties ofGroup 1 were primarily determined by phytoplankton cells (mea-sured as CHL) and those of Group 2 by mineral particles of terrigenousorigin (MSSter). Specific scattering coefficients, b*MSSter(λ) and b*CHL(λ) and backscattering coefficients, bb*MSSter(λ) and bb*CHL(λ), forthese two classes of material were estimated from the ratio of IOPsto OSCs for the stations in the two groups.

The remaining stations (Group 3) were not classifiable as beingdominated by phytoplankton or suspended sediment in Fig. 2.However a plot of bbp676 against MSST (Fig. 3.) showed that most ofthese Group 3 stations were located in a distinct cluster with verylow bbp values. These stations were located in areas of the Irish Seawith depths over 30 m and their distinctive optical characteristics in-cluded relatively low backscattering ratios, suggesting the presence oflarger particles (Fig. 4a.), and absorption coefficients which werehigher at 676 nm than at 650 nm, indicating the presence of a signif-icant chlorophyll absorption peak (Fig. 4b.). Examination of samples

Fig. 3. Particulate backscattering at 676 nm(bbp676) plotted against total suspendedmin-eral concentrations (MSST). Stations with a high proportion of diatoms are indicated byfilled symbols and labelledMSSdia, others asMSSter.


from 23 of these stations by optical microscopy, carried out by col-leagues at the University of Wales Bangor, showed the presence of ahigh proportion of colonial diatoms: mainly Leptocylindricus spp. andRhizosolenia spp. in August 2001 and Bacillaria paxillifer, Paralia sulcataand Melosira spp. in April 2002. Cell concentrations were generally inthe range 1×104 cells l−1 to 5×104 cells l−1.We conclude that a signif-icant fraction of theMSSTmeasured at these stations consists of biogenicminerals in the formof diatom frustules (MSSdia), and that the contribu-tion of this material to scattering and absorption should logically be in-clude in the phytoplankton-related CHL component.

2.6. Specific absorption coefficients

Suspended particles were collected by filtering 500 ml of seawaterthrough 25 mm diameter Whatman GF/F filters, which were storedfrozen until their optical density could be measured. The measure-ments were made using a custom-built spectrophotometer with ahighly collimated single beam in which the filters were placed imme-diately adjacent to a PTFE diffuser backed by a photomultiplier detec-tor. The filters were moistened with filtered seawater and mountedon a glass slide so that the particle-loaded side of the filter was thefirst surface encountered by the illuminating beam. A clean filter, sim-ilarly wetted and mounted, was used as a blank. Optical density spec-tra were measured before and after the extraction of pigments inacetone, and subtraction of the two spectra allowed the contributionsof phytoplankton pigments and non-algal particles to be determined.For consistency with the treatment of ac-9 data, optical densitieswere set to zero at 715 nm. The relationship between filter optical


image of Fig.�3

image of Fig.�4


Fig. 5. Apparent absorption coefficients of particles collected on filter pads, af (λ), plottedagainst particulate absorption coefficients measured in situ using an ac-9, asus (λ), for allwavebands listed in Table 1. The line, derived by geometric mean regression, has a slopeof β=1.73 and coefficient of determination r2=0.9.


density (ODf) and the apparent absorption coefficient (af) for the ma-terial on the filter is given by

af λð Þ ¼ 2:303×ODf λð Þ×AV

ð9Þ

where A is the effective area of the filter and V the volume of samplefiltered. The apparent absorption coefficients of particles collected onglass fibre filters is known to be greater that of an equivalent concen-tration of particles in suspension (asus), the two coefficients being re-lated by a path length amplification factor β

asus λð Þ ¼ af λð Þβ

ð10Þ

whose numerical value varies with measurement methodology and fil-ter loading (Bricaud & Stramski, 1990; Lohrenz, 2000; Tassan & Ferrari,1998). The ‘transmittance plus reflectance’ technique proposed byTassan and Ferrari (1998) can reduce this variability (Tassan et al.,2000) but it does not avoid the requirement for empirically derivedcoefficients. For the work reported here, we obtained a value for β byplotting the absorption coefficients derived with no pathlength com-pensation against the particulate absorption coefficients measuredat the same location in situ using an ac-9. Spectrophotometer mea-surements were averaged over the 10 nm bandwidths covered bythe ac-9 for this purpose. The resulting graph was clearly linear(Fig. 5.), with a gradient derived by geometrical mean regression ofβ=1.73 and a coefficient of determination r2=0.90. This procedureallowed filter pad optical densities to be converted to absorption

Table 1Values used for radiative transfer modelling.

λ aw bw a*CHL a*MSS

nm m−1 m−1 m2 mg−1 m2 g−1

412 0.0046 0.00666 0.06 0.08440 0.0064 0.00490 0.08 0.06488 0.0145 0.00316 0.06 0.04510 0.0325 0.00260 0.04 0.03532 0.0444 0.00218 0.03 0.02555 0.0596 0.00185 0.02 0.015650 0.3400 0.00100 0.02 0.01667 0.4346 0.00080 0.033 0.007676 0.4532 0.00074 0.04 0.005


coefficients for phytoplankton (aph) and non-algal particulates(anap) using Eqs. (9) and (10).

Specific absorption coefficients for phytoplankton and mineralparticles, (a*CHL(λ) and a*MSSter(λ)) were calculated by dividing the ab-sorption coefficients obtained from the analysis of samples on filtersby the measured OSM concentrations. For CDOM, the spectral depen-dence of absorption was described by

a λð Þa 440ð Þ ¼ e−S λ−440ð Þ ð11Þ

(Bricaud et al., 1981) where a(440) is the absorption coefficient at440 nm and the exponent S serves as the equivalent of an SIOP.

2.7. Radiative transfer modelling

Radiative transfer modelling of Rrs(λ) was carried out using Eco-light 5.0 (Mobley, Sequoia Scientific). In order to replicate typical con-ditions for acceptable satellite data acquisition, the solar zenith anglewas set to 45°, the wind speed to 5 ms−1 and cloud cover to zero.

Specific inherent optical propertieswere taken fromTable 1with lin-ear interpolation to cover the MODIS waveband of 667 nm. Particulatevolume scattering functions were estimated by selecting Fournier–Forand phase functions with appropriate bb/bp ratios from the EcoLightlibrary. Raman scattering was included in the calculations, but chloro-phyll fluorescence was set to zero. Appropriate ranges of constituentconcentrations were derived from field observations in the Irish Seaand Bristol Channel (McKee et al., 2007): these were 0–1.0 m−1 a(440)for CDOM, 0–10 mg m−3 for CHL and 0–20 g m−3 for MSSter. Values forthe total inherent optical properties of a given volume of seawaterwere then calculated by summing the contributions of the individualconstituents.

2.8. Satellite images

MODIS data for the Irish Seawas obtained fromGoddard DistributedActive Archive Centre and processed in our laboratory using SeaDAS 6.1,using the default 2-band aerosol model with iterative NIR correction.Level 2 (Rrs) data was mapped to an equidistant cylindrical projectionand further processing was carried out in MatLab.

3. Results

3.1. SIOPs for radiative transfer modelling

The aim of the radiative transfer modelling was to generate resultsthat were broadly representative of relationships between reflectanceand constituent concentrations in coastal waters rather than defini-tive for a particular station. Frequency distributions were thereforeplotted for the SIOPs derived by the methods described above, andtypical examples are shown in Figs. 6 and 7. In all cases, the calculated

a*CDOM b*CHL b*MSS bb*CHL bb*MSS

none m2 mg−1 m2 g−1 m2 mg−1 m2 g−1

1.4 0.12 0.4 0.00165 0.01621 0.12 0.4 0.00160 0.01600.56 0.12 0.4 0.00152 0.01560.43 0.12 0.4 0.00148 0.01540.33 0.12 0.4 0.00144 0.01520.25 0.12 0.4 0.00141 0.01500.08 0.12 0.4 0.00124 0.01420.067 0.12 0.4 0.00122 0.01410.06 0.12 0.4 0.00120 0.0140


image of Fig.�5


Fig. 6. Frequency distributions for the specific absorption coefficients of chlorophyll(a*CHL) and terrigenous minerals (a*MSSter) at 440 nm and for the CDOM absorptionexponent (SCDOM).


SIOPs formed unimodal distributions with well-defined peak values.The most frequently occurring values were selected as representativeSIOPs for modelling purposes, and these are listed in Table 1.

The width of these frequency distributions can be attributed to acombination of measurement error and intrinsic variability in specificoptical properties between stations, and it is difficult to deconvolutethese two factors. However the degree to which the peak values listedin Table 1. captured the variability in the inherent optical propertiesin the Irish Sea was assessed by reconstructing total (non-water)IOPs from SIOPs and constituent concentrations (Eqs. 6 and 7) andcomparing the results with measurements made in situ using theac-9 and HS-2.

Non-water absorption coefficients were reproduced to a reason-able level of accuracy in most wavebands: 412 nm is shown as an ex-ample in Fig. 8a. The relative spread of the results was greatest at676 nm (Fig. 8b), due partly to the much lower absorption valuesmeasured in this waveband. The majority of the reconstructed partic-ulate scattering and backscattering coefficients lay close to the 1:1lines (Figs. 9a and b), with the obvious exception of a cluster of points


in Fig. 9b where the use of the MMSter coefficient in Table 1 over-estimates bbp. This cluster consists of the diatom-rich stations identi-fied in Fig. 4. Analysis of frequency distributions for this group givesrepresentative values of 0.002 m2 g−1 for the specific backscatteringcoefficient and 0.6 mg g−1 for the CHL/MSS ratio, and combiningthese figures gives estimated chlorophyll-specific backscattering co-efficient for diatomaceous material of 0.003 m2 mg−1. This is suffi-ciently close to the bb*CHL value of 0.0012 m2 mg−1 listed inTable 1. to indicate that the contribution of diatoms to backscatteringcan be accounted for an adjustment of the SIOP value used for thephytoplankton component.

3.2. Comparisons of SIOPs with other data sets

Table 1 makes no attempt to represent the spatial and temporalvariability in SIOPs observed in more detailed studies (Babin et al.,2003a; Blondeau-Patissier et al., 2009; Loisel et al., 2010), but thevalues listed are broadly comparable with those reported elsewhere inthe literature. The logarithmic slope for CDOM absorption(0.012 nm−1) is within the range found by Bricaud et al., 1981(0.010–0.018 nm−1) and a little below that reported by Babin et al.,2003a (0.016–0.020 nm−1) and Tzortziou et al., 2007 (0.014–0.018 nm−1). The estimate for a*CHL at 440 nm (0.08 m2 mg−1) is con-sistent with the measurements of Bricaud et al., 1983 for cultured cells(0.04–0.05 m2 mg−1) and of Babin et al., 2003a in European coastal wa-ters (0.03 to 0.1 m2 mg−1). The value of 0.12 m2 mg−1 for b*CHL at440 nm falls within the range measured by Bricaud et al.1983 (0.07–0.18 m2 mg−1) and Weidemann & Bannister, 1986 (0.06 to0.12 m2 mg−1). The method used to estimate bb*CHL included any con-tribution of other particles which covariedwith chlorophyll, whichmayexplain why the value of 1.6×10−3 m2 mg−1 in Table 1 is rather higherthan that measured by Bricaud et al., 1983 in cell cultures (1×10−5 to2×10−4 m2 mg−1). However chlorophyll-specific backscattering inphytoplankton cultures exhibits considerable variability (Whitmireet al., 2010), and Vaillancourt et al., 2004 reported a mean value of8×10−4 m2 mg−1 for 29 species with a range extending over approxi-mately two orders of magnitude. For minerals, the value for a*MSSter(440) of 0.06 m2 g−1 is within the range reported by Babin et al., 2003for non-algal particles (0.033 m2g−1 to 0.067 m2 g−1) and close to theestimate of Bowers & Binding, 2006 (0.045–0.055 m2 g−1). The valuefor b*MSSter (0.45 m2 g−1) is rather higher than the figures given byBowers & Binding, 2006 (0.25–0.27 m2 g−1) but comparable to the re-sults of Stavn & Richter, 2008 (0.35 to 0.65 m2 g−1). Our value of0.036 for the ratio of bb*MSSter to b*MSSter at 650 nm is consistent withthe observation by Loisel et al., 2007 of a particulate backscatteringratio of 0.0417 in mineral-dominated waters.

3.3. Modelled relationships between MSSter and Rrs

Fig. 10 shows the fraction of the total absorption and backscatter-ing coefficients contributed by pure water to seawater containing arange of suspended mineral concentrations (with CDOM and CHL setto zero). The calculations employed the specific inherent opticalproperties for MSSter listed in Table 1, absorption coefficients forpure water from Pope and Fry (1997) and scattering coefficients forpure seawater from Smith and Baker (1981). The backscattering coef-ficients are strongly influenced byMSSter in all wavebands, but the ab-sorption coefficients are dominated by water at longer wavelengths.Consequently the ratio of the two coefficients becomes increasinglydetermined by MSSter-related backscattering at the red end of thespectrum. This explains the relative insensitivity of remote sensingreflectance calculations to errors in the determination of particulateabsorption. For a mineral concentration of 20 g m−3, for example,an under-estimate of 50% in particulate absorption at 676 nm wouldresult in an error of only 10% in the total absorption coefficient inthis waveband. Since remote sensing reflectance can be linked to


image of Fig.�6


Fig. 7. Frequency distributions for the specific scattering and backscattering coefficients at 676 nm for phytoplankton (b*CHL and bb*CHL, derived from Cluster 1) and suspendedmineral particles (b*MSSter and bb*MSSter, derived from Cluster 2).


the ratio of the backscattering and absorption coefficients (bb/a,Eq. 1), the key to effective remote sensing of suspended minerals isto select a waveband where this ratio is sensitive to values of MSSterover an extended dynamic range. Fig. 10 shows that this conditionis best satisfied at the long wavelength end of the spectrum.

When CHL and CDOM are set to zero, Eq. 8 becomes simply

Rrs λð Þ≈κ λð Þ bbW λð Þ þ b�bMSSterλð ÞMSSter

aW λð Þ þ a�MSSterλð ÞMSSter

" #12

The degree of proportionality in the relationship between Rrs(λ)and bb(λ)/a(λ), i.e. the variability of κ, was investigated by carryingout a set of Ecolight modelling runs at ac-9 wavelengths incorporat-ing SIOPs from Table 1, zero cloud cover and a solar angle of 45°.For MSS concentrations between 0.1 and 20 g m−3, these calculationsgenerated values of κ ranging from 0.046 to 0.052, with an overallmean of 0.049. Eq. (12) therefore gives useful insights into the rela-tionship between Rrs and MSS in situations where the contributionsof other optically significant constituents can be neglected. WhenMSSter=0, it gives the reflectance of pure seawater. When MSStervalues are high, the term in brackets tends towards bbMSS*

ter/aMSSter*

and the reflectance is determined by the ratio of backscattering to ab-sorption for suspended minerals. The relationship between Rrs andMSSter at intermediate values takes the form of a set of saturationcurves whose shape is strongly wavelength-dependent. These curvesare shown in Fig. 11, where the symbols indicate the results obtainedby radiative transfer modelling using Ecolight and the lines show Rrs


calculated from Eq. (12), incorporating SIOPs from Table 1 with κ setto 0.049.

This figure shows that the assumption of a single value for κ doesnot replicate the Ecolight results exactly, but it gives a reasonable ap-proximation for most wavebands. It also indicates that the relation-ship between Rrs and MSSter is always curved, even though thecurves are close to linear in the red wavebands for MSSter below20 g m−3. Further calculations indicated that no simple wavebandratio significantly extends the region of near-linearity. Since chloro-phyll fluorescence may contribute to the signal detected at 687 nm,and wavebands beyond 700 nm are used in atmospheric correctionalgorithms (Shi & Wang, 2009), it can be concluded that the SeaWiFSband at 670 nm or the MODIS band at 667 nm are the most suitable inpractice for estimating MSS values using single-band algorithms. Asimilar conclusion was drawn from a more complex analysis of theunderlying optics by Nechad et al. (2010), and is supported in thestudy of the relationship between turbidity and reflectance in tropicalcoastal waters by Ouillon et al. (2008).

3.4. Sensitivity to interfering substances

The analysis of the previous section leaves unanswered the ques-tion of how MSSter retrieval is affected by the presence of other opti-cally significant constituents. The effects of the separate addition ofeither CDOM or CHL on the relationship between remote sensing re-flectance at 667 nm and MSS are shown in Fig. 12, with the symbolsagain indicating Ecolight numerical results and the lines solutions ofEq. (12). Since CDOM affects only the total absorption coefficient


image of Fig.�7


Fig. 8. Comparison between non-water absorption coefficients measured by ac-9 insitu and those calculated from SIOPs (Table 1.) and OSM concentrations for the412 nm and 676 nm wavebands.

Fig. 9. Comparison between particulate scattering (a.) and backscattering (b.) coeffi-cients at 676 nm measured in situ and those calculated using SIOPs (Table 1.) andOSM concentrations. Open symbols indicate stations with a high proportion of diatoms(see Fig. 3).


(the denominator of Eq. 8), increasing CDOM always decreases Rrs fora given MSSter. Adding CHL alters both the total absorption and back-scattering coefficients and affects both the numerator and denomina-tor in Eq. (2), and for the SIOPs in Table 1 increases Rrs667 for MSStervalues below 2 g m−3 and decreases Rrs above this value. Further cal-culations, involving around 20,000 Ecolight runs, indicate that the en-velope enclosed by the maximum and minimum values in Fig. 12contains all the Rrs667 values which are consistent with a givenMSSterwhen both CDOM and CHL are present and allowed to vary indepen-dently over the full range in the model (0–1 m−1 for CDOM, 0–10 g m−3 for CHL).

Fig. 10. Fractional contribution of water to the total absorption coefficients (open sym-bols) and scattering coefficients (filled symbols) of seawater containing the range ofsuspended mineral concentrations (MSSter) indicated.

3.5. Predictive capability of a model incorporating regionallyrepresentative SIOPs

In regions such as the Irish Sea, frequent cloud cover makes matchups with satellite overpasses rarely possible, and spatial heterogeneityreduces the correspondence between point sampling and pixel-averaged remote sensing observations. For this study, therefore, radia-tive transfer results were compared with Rrs 667 observations derivedfrom SPMR deployments. As a preliminary step, errors for Rrs and MSSTobservations were estimated for stations where duplicate sample ana-lyses or casts were carried out. These were taken to be the standard de-viation of the residuals obtained from a linear least squares fit toscatterplots of all duplicate observations for the two variables, andwere +/−0.7 g m−3 for MSST and +/−0.001 sr−1 for Rrs.

Fig. 13 shows Rrs 667 derived from in situ radiometric profiles atthe stations mapped in Fig. 1 plotted against measurements of MSSTfor surface water samples taken at the same stations.


Upper and lower boundaries calculated using Ecolight with SIOPsfrom Table 1 are superimposed.

The classification of stations as being dominated byMSSter orMSSdia(Section 2.5) is indicated by filled and open symbols, and error bars areincluded only for MSSter stations to avoid over-congestion of the plot.Three significant conclusions can be drawn from Fig. 13. The first isthat the radiative transfer calculations generated Rrs667 values thatwere largely consistent with SPMR observations. This result indicatesa considerable degree of consistency between three sets of independentmeasurements (OSCs, IOPs and radiometric profiles), and demonstratesthat the procedures employed for deriving representative SIOPs in


image of Fig.�8

image of Fig.�9

image of Fig.�10


Fig. 11. Functional relationships between remote sensing in the 667 nm MODIS wave-band (Rrs 667) and suspended sediment concentration (MSSter) in the absence of otheroptical significant materials. The symbols indicate results from radiative transfermodelling, and the lines are solutions to Eq. (12) with κ=0.049. SIOP values fromTable 1. were used in both cases.

Fig. 13. Lines show the upper and lower boundaries calculated by radiative transfermodelling for CHL and CDOM varying independently in the range 0–10 mg m−3 and0–1 m−1 respectively. Symbols indicate observations from 110 stations with reflec-tances calculated from radiometric profiles and MSST determined gravimetrically. Thediscrimination between diatomaceous and terrigenous MSS indicated by open andfilled symbols is based on the classification of Fig. 3.


Table 1 have not introduced any systematic bias. The second is that thespread in the Rrs vs MSSter relationship observed for this set of data (atleast for MMSter in the range 4 g m−3 to 14 g m−3) can be accountedfor by the degree of variability introduced by unconstrained CDOMand CHL values evenwhen variability on the SIOPs themselves is not in-cluded in the analysis. The third is that the presence of biogenic silicawith a very low specific backscattering cross section has a significant ef-fect on the direct problemof estimating Rrs fromMSST but little effect onthe inverse problem of estimating MSSter from Rrs. The degree of vari-ability shown in Fig. 13 suggests that, in the absence of definite knowl-edge of CHL and CDOMconcentrations, any algorithm recoveringMSStershould report the maximum and minimum concentrations consistentwith a specified range of these constituents. For the range consideredin this study, the upper (u) and lower (l) boundaries derived from radi-ative transfer calculations can be approximately described by secondorder polynomials of the form:

MSSter uð Þ ¼ 26014× Rrs667ð Þ2 þ 916× Rrs667ð Þ−0:13 ð13Þ

MSSter lð Þ ¼ 2508× Rrs667ð Þ2 þ 768× Rrs667ð Þ−0:77 ð14Þ

This approximation allows maximum and minimum MSSter valuesto be derived from Rrs667 observations without incurring a signifi-cant computational overhead.

Fig. 12. Effect of the independent addition of CDOM and CHL on the relationship betweenRrs667 andMSSter. The symbols indicate results from radiative transfer modelling, and thelines are solutions to Eq. (12) with κ=0.049.


3.6. An illustrative example

Fig. 14 shows a contour map of Rrs667 for the region of interest de-rived from a MODIS image acquired on May 2nd 2007. The map ex-hibits spatial patterns previously reported by Bowers (2003), withhigh reflectances occurring in the major estuaries and in regionswith high tidal velocities off Anglesey (NW Wales) and WicklowHead (on the Irish coast). Transects are drawn on Fig. 14 from theopen waters of St Georges Channel (point A) east into the BristolChannel (point B) and north east into the Irish Sea (point C).

The upper and lower limits for MSSter retrievals along these tran-sects, calculated using Eqs. (13) and (14), are illustrated in Fig. 15. Thetransects start in waters of low reflectance (and, by implication, low

Fig. 14. Remote sensing reflectance in the 667 nm waveband derived from a MODISAqua image acquired on May 2nd 2007, with transects extending from St GeorgesChannel (A) into the Bristol Channel (B) and the Irish Sea (C).


image of Fig.�11

image of Fig.�12

image of Fig.�13

image of Fig.�14


Fig. 15. Upper and lower boundaries of possible MSSter values for the transects drawnon Fig. 13, calculated using Eq. (13) on the assumption that CHL and CDOM concentra-tions fall within the ranges 0–10 mg m−3 and 0–1 m−1 respectively.


MSSter values). Transect A-B shows a steep rise in suspended mineralconcentrations starting at 140 km, which is where it enters the turbidSevern Estuary. Transect A-C shows a more gentle rise in MSSter levelsat 100 km from the starting point, which is where it crosses the frontbetween the seasonally stratified waters of St Georges Channel andthe tidally mixed waters of the central Irish Sea (Simpson & Hunter,1974).

The figure shows that it is possible to locate major features andtransitions in mineral particle distributions to within a few kilometresin spite of the calculated uncertainties in absolute values. This capa-bility is potentially important for validating sediment transportmodels using ocean colour imagery.

4. Discussion

Many practical applications of optical remote sensing involve themapping of suspended sediment concentrations in regions of moder-ate turbidity. Examples include monitoring river outflow and plumedispersal (D'Sa & Ko, 2008; Lahet & Stramski, 2010; Warrick et al.,2004, 2007), assessing the environmental footprint of dredging oper-ations (Islam et al., 2007; Sipelgas et al., 2006, 2009), and determiningthe geographical extent of the effects of extreme events such as hur-ricanes and tsunamis on the coastal zone (Chen et al., 2009; Lohrenzet al., 2008; Yan & Tang, 2009; Zawada et al., 2007). In addition, quan-titative estimates of relatively low concentrations of suspended min-erals play a key role in predicting light penetration in coastal seas(Tian et al., 2009), modelling underwater visibility (Zaneveld &Pegau, 2003), and tracking the transport of pollutants such as heavymetals and radionuclides (Battle et al., 2008; Charlesworth et al.,2006). Given the nature of these applications, and their potential eco-nomic and legislative implications, it is important that remotelysensed suspended sediment concentrations are accompanied by anexplicit assessment of the associated uncertainties. The aim of thepresent study is to quantify the extent to which these uncertaintiesoriginate in the sensitivity of remote sensing reflectance to the pres-ence on non-mineral materials in moderately turbid waters. The radi-ative transfer calculations employed for this purpose incorporatedSIOPs which are broadly characteristic of the Irish Sea and adjacentwaters, and generally consistent with values reported for other shelfsea areas. The results confirmed that the relationship between remotesensing reflectance and suspended sediment concentration saturateswith increasing concentration in all wavebands (and waveband ra-tios) in the visible region of the spectrum, and that the range of ap-proximate linearity is greatest for red wavelengths. However sincechlorophyll fluorescence can contribute significantly to water-leaving radiance above 670 nm and wavelengths above 700 nm arefrequently employed for atmospheric correction, signals in the 660–


670 nm band are the obvious choice for visible-band MSS algorithms.Taking the MODIS Rrs667 signal as an example, it is possible to drawthree main conclusions. First, as noted by Bowers et al., 1998; Eleveldet al., 2008; Nechad et al., 2010 and Zhang et al., 2010, the relation-ship between red-waveband reflectance and MSSter becomes notice-ably curved at concentrations above 20 g m−3. Second, the effect ofCHL and CDOM over most of the MSSter range considered (2.5–20 g m−3) is to reduce the Rrs676 signal, leading to under-estimatesof MSSter by single waveband algorithms. Third, since the presenceof other optically significant constituents is sufficient to account forthe variability in the relationship between Rrs and MSSer in the IrishSea reported by Binding et al. (2005), caution is required if this vari-ability is to be interpreted in terms of changes in particle size distri-butions (van der Lee et al., 2009).

It is interesting to compare the magnitude of the uncertainties inMSS recovery introduced by the presence of unknown concentrationsof non-mineral seawater constituents with those due to errors of obser-vation. The target accuracy for the recovery of water-leaving radiancesfrom the SeaWiFS and MODIS Aqua satellite radiometers is +/−5%. Vi-carious calibration procedures suggest that this figure is significantlyexceeded in practice, especially at the long wavelength end of the visi-ble spectrum (Franz et al., 2007; Gregg et al., 2009). It is possible, how-ever, that vicarious calibration over-estimates errors in the redwavebands because the reference sites are located in oceanic waterswhere the red reflectance signals are very low. A +/−10% error in Rrs667, which is a reasonable minimum estimate for coastal waters,would generate uncertainties in a single-waveband linear MSS algo-rithm of 2 g m−3 at a concentration of 10 g m−3. This is significantlylower than the figure of 7 g m−3 computed for the modelled range ofinterfering substances using Eqs. (13) and (14).

One conclusion arising from this study is that the products ofMSS al-gorithms in optically complex waters could usefully be reported interms of upper and lower possible values, rather than as percentage er-rors. This procedure has several advantages. It emphasises the factmuch of the uncertainty in MSS retrieval arises from the compositionof the water body being inadequately known rather than from errorsof observation. It reflects the fact that, over most of the relevant range,the effect of interfering substances is not to introduce random errorsbut to cause remote sensing algorithms to systematically under-estimateMSS values. It also allows the range of possible values to be ad-justed in response prior knowledge of seasonal or spatial variations inthe likely range of constituent concentrations. Further refinement ofthe model of shelf sea optics developed in this paper requires a clearerunderstanding of the optical properties of suspended mineral particles.Methodological problems currently prevent the direct determination ofthe scattering and NIR absorption coefficients of these particles in natu-ralmixtures, and remarkably little information is available on their like-ly variability (Bowers & Binding, 2006; Snyder et al., 2008). Sinceremote sensing products are highly sensitive to the SIOPs of mineralparticles, particularly the specific backscattering coefficient, this is anarea which is in need of further study.

References

Babin, M., Morel, A., Fournier-Sicre, V., Fell, F., & Stramski, D. (2003). Light scatteringproperties of marine particles in coastal and open ocean waters as related to theparticle mass concentration. Limnology and Oceanography, 48, 843–859.

Babin, M., & Stramski, D. (2002). Light absorption by aquatic particles in the near-infrared spectral region. Limnology and Oceanography, 47, 911–915.

Babin, M., & Stramski, D. (2004). Variations in the mass-specific absorption coefficientof mineral particles suspended in water. Limnology and Oceanography, 49,756–767.

Babin, M., Stramski, D., Ferrari, G. M., Claustre, H., Bricaud, A., Obolensky, G., et al.(2003). Variations in the light absorption coefficients of phytoplankton, nonalgalparticles, and dissolved organic matter in coastal waters around Europe. Journalof Geophysical Research, 108(C7), 3211.

Battle, J. V. I., Bryan, S., & McDonald, P. (2008). A process-based model for the partition-ing of soluble, suspended particulate and bed sediment fractions of plutonium andcaesium in the eastern Irish Sea. Journal of Environmental Radioactivity, 99, 62–80.


image of Fig.�15



Binding, C. E., Bowers, D. G., & Mitchelson-Jacob, E. G. (2003). An algorithm for the re-trieval of suspended sediment concentrations in the Irish Sea from SeaWiFS oceancolour satellite imagery. International Journal of Remote Sensing, 24, 3791–3806.

Binding, C. E., Bowers, D. G., & Mitchelson-Jacob, E. G. (2005). Estimating suspendedsediment concentrations from ocean colour measurements in moderately turbidwaters; the impact of variable particle scattering properties. Remote Sensing of En-vironment, 94, 373–383.

Binding, C. E., Jerome, J. H., Bukata, R. P., & Booty, W. G. (2010). Suspended particulatematter in Lake Erie derived from MODIS aquatic colour imagery. International Jour-nal of Remote Sensing, 31, 5239–5255.

Blondeau-Patissier, D., Brando, V. E., Oubelkheir, K., Dekker, A. G., Clementson, L. A., &Daniel, P. (2009). Bio-optical variability of the absorption and scattering propertiesof the Queensland inshore and reef waters, Australia. Journal of GeophysicalResearch-Oceans, 114, C05003.

Bowers, D. G. (2003). A simple turbulent energy-based model of fine suspended sedi-ments in the Irish Sea. Continental Shelf Research, 23, 1495–1505.

Bowers, D. G., & Binding, C. E. (2006). The optical properties of mineral suspended par-ticles: A review and synthesis. Estuarine, Coastal and Shelf Science, 67, 219–230.

Bowers, D. G., Boudjelas, S., & Harker, G. E. L. (1998). The distribution of fine suspendedsediments in the surface waters of the Irish Sea and its relation to tidal stirring. In-ternational Journal of Remote Sensing, 19, 2789–2805.

Bricaud, A., Morel, A., & Prieur, L. (1981). Absorption by dissolved organic matter of thesea (yellow substance) in the UV and visible domains. Limnology and Oceanogra-phy, 26, 43–53.

Bricaud, A., Morel, A., & Prieur, L. (1983). Optical efficiency factors of some phytoplank-ters. Limnology and Oceanography, 28, 816–832.

Bricaud, A., & Stramski, D. (1990). Spectral absorption coefficients of living phytoplank-ton and non-algal biogenous matter: A comparison between the Peru upwellingarea and Sargasso Sea. Limnology and Oceanography, 35, 562–582.

Charlesworth, M. E., Chenery, S., Mellor, A., & Service, M. (2006). Isotopic compositionand concentration of Pb in suspended particulate matter of the Irish Sea revealsdistribution and sources. Marine Pollution Bulletin, 52, 81–88.

Chen, S. S., Huang, W. R., Wang, H. Q., & Li, D. (2009). Remote sensing assessment ofsediment re-suspension during Hurricane Frances in Apalachicola Bay, USA. Re-mote Sensing of Environment, 113, 2670–2681.

D'Sa, E., & Ko, D. S. (2008). Short-term influences on suspended particulate matter dis-tribution in the Northern Gulf of Mexico: Satellite and model observations. Sensors,8, 4249–4264.

Doerffer, R., & Schiller, H. (2007). The MERIS case 2 water algorithm. International Jour-nal of Remote Sensing, 28, 517–535.

Eleveld, M. A., Pasterkamp, R., van der Woerd, H. J., & Pietrzak, J. D. (2008). Remotelysensed seasonality in the spatial distribution of sea-surface suspended particulatematter in the southern North Sea. Estuarine, Coastal and Shelf Science, 80, 103–113.

Franz, B. A., Bailey, S. W., Werdell, P. J., & McClain, C. R. (2007). Sensor-independent ap-proach to the vicarious calibration of satellite ocean color radiometry. Applied Op-tics, 46, 5068–5082.

Gregg, W. W., Casey, N. W., O'Reilly, J. E., & Esias, W. E. (2009). An empirical approachto ocean colour data: Reducing bias and the need for post-launch radiometric re-calibration. Remote Sensing of Environment, 113, 1598–1612.

Hommersom, A., Wernand, M. R., Peters, S., & de Boer, J. (2010). A review on substancesand processes relevant for optical remote sensing of extremely turbid marineareas, with a focus on the Wadden Sea. Helgoland Marine Research, 64, 75–92.

Hu, C. M., Chen, Z. Q., Clayton, T. D., Swarzenski, P., Brock, J. C., & Muller-Karger, F. E.(2004). Assessment of estuarine water-quality indicators using MODIS medium-resolution bands: Initial results from Tampa Bay, FL. Remote Sensing of Environ-ment, 93, 423–441.

Islam, A., Wang, L., Smith, C., Reddy, S., Lewis, A., & Smith, A. (2007). Evaluation of sat-ellite remote sensing for operational monitoring of sediment plumes produced bydredging at Hay Point, Queensland, Australia. Journal of Applied Remote Sensing, 1(Article Number: 011506).

Jeffrey, S. W., & Humphrey, G. F. (1975). New spectrophotometric equations for deter-mining chlorophylls a, b, c1 and c2 in higher plants, algae and natural phytoplank-ton. Biochemie und Physiologie der Pflanzen, 167, 191–194.

Kirk, J. T. O. (1981). A Monte Carlo study of the nature of the underwater light field in,and the relationship between optical properties of, turbid yellow waters. AustralianJournal of Marine & Freshwater Research, 32, 517–532.

Kirk, J. T. O. (1994). Light and photosynthesis in aquatic ecosystems (2nd edition). : Cam-bridge University Press.

Kuchinke, C. P., Gordon, H. R., & Franz, B. A. (2009). Spectral optimization for constitu-ent retrieval in Case 2 waters I: Implementation and performance. Remote Sensingof Environment, 113, 571–587.

Kuchinke, C. P., Gordon, H. R., Harding, L. W., & Voss, K. J. (2009). Spectral optimizationfor constituent retrieval in Case 2 waters II: Validation study in the ChesapeakeBay. Remote Sensing of Environment, 113, 610–621.

Lahet, F., & Stramski, D. (2010). MODIS imagery of turbid plumes in San Diego coastalwaters during rainstorm events. Remote Sensing of Environment, 114, 332–344.

Lee, Z., Carder, K. L., Mobley, C. D., Steward, R. G., & Patch, J. S. (1999). Hyperspectralremote sensing for shallow waters: 2. Deriving bottom depths and water proper-ties by optimization. Applied Optics, 38, 3831–3843.

Leymarie, E., Doxaran, D., & Babin, M. (2010). Uncertainties associated to measure-ments of inherent optical properties in natural waters. Applied Optics, 49,5415–5436.

Lohrenz, S. E. (2000). A novel theoretical approach to correct for pathlength amplifica-tion and variable sampling loading in measurements of particulate spectral ab-sorption by the quantitative filter technique. Journal of Plankton Research, 22,639–657.


Lohrenz, S. E., Cai, W. -J., Chen, X., & Tuel, M. (2008). Satellite assessment of bio-opticalproperties of Northern Gulf of Mexico coastal waters following hurricanes Katrinaand Rita. Sensors, 8, 4135–4150.

Loisel, H., Lubac, B., Dessailly, D., Duforet-Gaurier, L., & Vantrepotte, V. (2010). Effect ofinherent optical properties variability on the chlorophyll retrieval from ocean colorremote sensing: An in situ approach. Optics Express, 18, 20949–20959.

Loisel, H., Mériaux, X., Berthon, J. -F., & Poteau, A. (2007). Investigation of the opticalbackscattering to scattering ratio of marine particles in relation to their biogeo-chemical composition in the eastern English Channel and southern North Sea. Lim-nology and Oceanography, 52, 739–752.

Maritorena, S., Siegel, D. A., & Peterson, A. R. (2002). Optimization of a semianalyticalocean color model for global-scale applications. Applied Optics, 41, 2705–2714.

McKee, D., Cunningham, A., & Dudek, A. (2007). Optical water type discrimination andtuning remote sensing band-ratio algorithms: Application to retrieval of chloro-phyll and K-d(490) in the Irish and Celtic Seas. Estuarine, Coastal and Shelf Science,73, 827–834.

Mobley, C. D. (1999). Estimation of the remote-sensing reflectance from above-surfacemeasurements. Applied Optics, 38, 7442–7455.

Morel, A., & Prieur, L. (1977). Analysis of variations in ocean colour. Limnology andOceanography, 22, 709–722.

Nechad, B., Ruddick, K. G., & Park, Y. (2010). Calibration and validation of a genericmultisensor algorithm for mapping of total suspended matter in turbid waters. Re-mote Sensing of Environment, 114, 854–866.

Ouillon, S., Douillet, P., Petrenko, A., Neveux, J., Dupouy, C., Froidefond, J. M., et al.(2008). Optical algorithms at satellite wavelengths for total suspended matter intropical coastal waters. Sensors, 8, 4165–4185.

Pope, R. M., & Fry, E. S. (1997). Absorption spectrum (380–700 nm) of pure water. II.Integrating cavity measurements. Applied Optics, 36, 8710–8723.

Robinson, I. S. (2008). Remote sensing of shelf sea ecosystems. Position Paper 12. Ostend,Belgium: European Science Foundation.

Sathyendranath, S. (2000). Remote sensing of ocean colour in coastal, and other optically-complex, waters. Reports of the Ocean Colour Coordinating Group, 3, Dartmouth, Canada:IOCCG.

Schiller, H., & Doerffer, R. (1999). Neural network for emulation of an inverse model-operational derivation of Case II water properties from MERIS data. InternationalJournal of Remote Sensing, 20, 1735–1746.

Shi, W., & Wang, M. H. (2009). An assessment of the black ocean pixel assumption forMODIS SWIR bands. Remote Sensing of Environment, 113, 1587–1597.

Shi, W., & Wang, M. H. (2010). Characterization of global ocean turbidity from moder-ate resolution imaging spectroradiometer ocean color observations. Journal of Geo-physical Research-Oceans, 115, C11022.

Simpson, J. H., & Hunter, J. (1974). Fronts in the Irish Sea. Nature, 250, 404–406.Sipelgas, L., Ossipova, V., Raudsepp, U., & Lindfors, A. (2009). A bio-optical model for

the calculation of suspended matter concentration from MODIS data in the PakriBay, the Gulf of Finland. Boreal Environment Research, 14, 415–426.

Sipelgas, L., Raudsepp, U., & Kouts, T. (2006). Operational monitoring of suspendedmatter distribution using MODIS images and numerical modelling. Advances inSpace Research, 38, 2182–2188.

Smith, R. C., & Baker, K. S. (1981). Optical properties of the clearest natural waters(200–800 nm). Applied Optics, 20, 177–184.

Snyder, W. A., Arnone, R. A., Davis, C. O., Goode, W., Gould, R. W., Ladner, S., et al.(2008). Optical scattering and backscattering by organic and inorganic particulatesin U.S. coastal waters. Applied Optics, 47, 666–677.

Stavn, R.H., &Richter, S. J. (2008). Biogeo-optics: Particle optical properties and thepartition-ing of the spectral scattering coefficient of ocean waters. Applied Optics, 47, 2660–2679.

Stramski, D., & Wozniak, S. B. (2004). Optical properties of Asian mineral dust sus-pended in seawater. Limnology and Oceanography, 49, 749–755.

Sullivan, J. M., Twardowski, M. S., Zaneveld, J. R. V., Moore, C. M., Barnard, A. H., Donaghay,P. L., et al. (2006). Hyperspectral temperature and salt dependencies of absorption bywater and heavy water in the 400–750 nm spectral range. Applied Optics, 45,5294–5309.

Tanaka, A., Kishino, M., Doerffer, R., Schiller, H., Oishi, T., & Kubota, T. (2004). Develop-ment of a neural network algorithm for retrieving concentrations of chlorophyll,suspended matter and yellow substance from radiance data of the ocean colorand temperature scanner. Journal of Oceanography, 60, 519–530.

Tassan, S., & Ferrari, G. M. (1998). Measurement of light absorption by aquatic particlesretained on filters: Determination of the optical pathlength amplification factor bythe ‘transmittance–reflectance’ method. Journal of Plankton Research, 20,1699–1709.

Tassan, S., & Ferrari, G. M. (2003). Variability of light absorption by aquatic particles inthe near infra-red spectral region. Applied Optics, 42, 4802–4810.

Tassan, S., Ferrari, G. M., Bricaud, A., & Babin, M. (2000). Variability of the amplificationfactor of light absorption by filter-retained aquatic particles in the coastal environ-ment. Journal of Plankton Research, 22, 659–668.

Tian, T., Merico, A., Su, J., Staneva, J., Wiltshire, K., & Wirtz, K. (2009). Importance ofresuspended sediment dynamics for the phytoplankton spring bloom in a coastalmarine ecosystem. Journal of Sea Research, 62, 214–228.

Tzortziou, M., Herman, J., Gallegos, C. L., Neale, P. J., Subramaniam, A., Harding, L. W.,et al. (2006). Bio-optics of the Chesapeake Bay from measurements and radiativetransfer closure. Estuarine, Coastal and Shelf Science, 68, 348–362.

Tzortziou, M., Subramaniam, A., Herman, J. R., Gallegos, C. L., Neale, P. J., & Harding, L.W. (2007). Remote sensing reflectance and inherent optical properties in themid Chesapeake Bay. Estuarine, Coastal and Shelf Science, 72, 16–32.

Vaillancourt, R. D., Brown, C. W., Guillard, R. L., & Balch, W. M. (2004). Light backscat-tering properties of marine phytoplankton: Relationships to cell size, chemicalcomposition and taxonomy. Journal of Plankton Research, 26, 191–212.




van der Lee, E. M., Bowers, D. G., & Kyte, E. (2009). Remote sensing of temporal and spatialpatterns of suspended particle size in the Irish Sea in relation to the Kolmogorov mi-croscale. Continental Shelf Research, 29, 1213–1225.

van der Woerd, H. J., & Pasterkamp, R. (2008). HYDROPT: A fast and flexible method toretrieve chlorophyll-a from multispectral satellite observations of optically com-plex coastal waters. Remote Sensing of Environment, 112, 1795–1807.

Warrick, J. A., Di Giacomo, P. M., Welsberg, S. B., Nezlin, N. P., Mengel, M., Jones, B. H.,et al. (2007). River plume patterns and dynamics within the Southern CaliforniaBight. Continental Shelf Research, 27, 2427–2448.

Warrick, J. A., Mertes, L. A. K., Siegel, D. A., & Mackenzie, C. (2004). Estimating sus-pended sediment concentrations in turbid coastal waters of the Santa BarbaraChannel with SeaWiFS. International Journal of Remote Sensing, 25, 1995–2002.

Weidemann, A. D., & Bannister, T. T. (1986). Absorption and scattering coefficients inIrondequoit Bay. Limnology and Oceanography, 31, 567–583.


Whitmire, A. L., Pegau,W. S., Karp-Boss, L., Boss, E., & Cowles, T. J. (2010). Spectral backscat-tering properties of marine phytoplankton cultures. Optics Express, 18, 15073–15093.

Yan, Z. Z., & Tang, D. L. (2009). Changes in suspended sediments associated with 2004Indian Ocean tsunami. Advances in Space Research, 43, 89–95.

Zaneveld, R. V., Kitchen, J. C., & Moore, C. (1994). The scattering error correction ofreflecting-tube absorption meters. Ocean Optics XII, SPIE, 2258. (pp. 44–55).

Zaneveld, J. R. V., & Pegau, W. S. (2003). Robust underwater visibility parameter. OpticsExpress, 11, 2997–3009.

Zawada, D. G., Hu, C. M., Clayton, T., Chen, Z. Q., Brock, J. C., & Muller-Karger, F. E.(2007). Remote sensing of particle backscattering in Chesapeake Bay: A 6-yearSeaWiFS retrospective view. Estuarine, Coastal and Shelf Science, 73, 792–806.

Zhang, M. W., Tang, J., Dong, Q., Song, Q., & Ding, J. (2010). Retrieval of total suspendedmatter concentration in the Yellow and East China Seas from MODIS imagery. Re-mote Sensing of Environment, 114, 392–403.



remote sensing of environment -...

Documents