Optical investigations of CDOM-rich coastal waters in Parnu Bay/Optilised mootmised lahustunud orgaanilise aine rikastes Parnu lahe rannikuvetes. (2024)

INTRODUCTION

Nowadays problems connected with the estimation of the ecologicalstate of seas and inland waters have become especially topical due toincreasing industrial and human impact on the aquatic environment. For acomprehensive survey of some water body complex investigationsconsisting of chemical, hydrophysical, optical and biologicalmeasurements are necessary. However, rather essential conclusions may bedrawn also on the basis of a certain group of in situ measurements:quite often the complex of data, containing the concentrations ofoptically significant substances (phytoplankton, coloured dissolvedorganic matter (CDOM) and suspended matter) as well as the incomingirradiance and diffuse attenuation coefficient, has been collected.These data enable also determination of the underwater irradiance, whichis an important factor in forming the phytoplankton productivity.Optical measurements yield information on temporal and spatial variationin optically significant substances, including certain kinds of waterpollution.

The advances in ocean colour remote sensing over the decades havemade it possible to use remote sensing imagery to produce maps ofproductivity in the world oceans (Platt & Sathyendranath 1988).Considering the seas and coastal waters, the daily MODIS (onboard theAqua platform) overpass covers the whole of the Baltic Sea area, whichmakes it very operative for environmental monitoring. The MODIS/AquaLevel 1 (Top of Atmosphere, TOA) and Level 2 images (includingatmospherically corrected water-leaving radiance and chlorophyll aconcentration) with 250 m spatial resolution are freely available. TheMODIS standard algorithm for the diffuse attenuation coefficient ofseawater at 490 nm, [K.sub.d](490), has been developed using the ratioof water-leaving radiances at 490 and 555 nm (Austin & Petzold 1981;Mueller 2000). ^d(490) gives more accurate results in the Baltic Seathan chlorophyll standard algorithms (Darecki & Stramski 2004;Kratzer et al. 2008). The values of [K.sub.d](490) are influenced by allabsorbing substances in the water including CDOM, which is often thedominant absorbing compound in these bands.

However, coastal waters are optically complex, characterized by alarge variability of optically significant constituents resulting fromdifferent biological, chemical and physical processes. This leads todifficulties in the interpretation of satellite data in coastal andinland waters due to several technical and methodological limitations(spatial, spectral and radiometric resolution of the instrument, and thedetermination of atmospheric correction in multicomponental waters).

Many Estonian coastal regions and inland waters are under stronghuman impact, and the Baltic is a rather polluted internal sea. Thenature of the bays is often influenced by the inflow of rivers thatbring along large quantities of CDOM. Small and semi-enclosed Parnu Bay(in the Eastern Baltic Sea) is under the influence of the town of Parnuwith its 70 000 inhabitants and the high inflow from the Parnu River.This causes the low transparency of Parnu Bay water. The bay has beeninvestigated in situ for some years, but the majority of investigationshave been conducted in the sphere of hydrology and biology. In thepresent study the optically complex Parnu Bay was chosen for in situmeasurements with the aim of comprehensive investigation of thetemporal-spatial variability of the optical properties of water. Weconsidered it also as a test site for assessing the possibilities ofdeveloping remote sensing algorithms for this kind of water. Forbuilding and testing a model for interpretation of remote sensing data,however, simultaneous in situ measurements of optically significantsubstances and radiation characteristics in the water body are needed.

Besides, in situ measurement data have their own, independentvalue. Firstly, the satellite sensor cannot describe the water state inthe conditions of cloudy weather. Secondly, the signal coming into thesensor originates from the surface layer, which is only 20-40 cm thickin case of turbid waters (Secchi disk depth below 0.5 m) (Arst 2003).For this reason the in situ data on the vertical profiles of opticallysignificant substances and underwater irradiance are indispensable.

A database describing the variability of the biooptical waterparameters in Parnu Bay (and partly in the Gulf of Riga) was collectedduring 2006-2007 (11 sampling stations, 10 series of field trips).Several correlation relationships between optically significantsubstances and radiation characteristics were investigated.

[FIGURE 1 OMITTED]

MATERIAL AND METHODS

Description of Parnu Bay

Parnu Bay is a shallow water basin in the northeastern Gulf of Riga(Fig. 1), which could be divided into an inner and an outer basin. Theinner part has approximate measures of 13 km x 14 km, an area of about190 [km.sup.2] and the maximum depth of 7.6 m. The outer part extendsdown to the southern tip of Kihnu Island, having an area of about 500[km.sup.2] and the maximum depth of about 15 m.

The quality of water and quantity of nutrients in Parnu Bay dependon the inflow of fresh water from rivers and on the intrusion of waterfrom the Gulf of Riga due to changes in wind direction and water level.Nutrient concentrations in Parnu Bay are higher than the average levelin the Gulf of Riga, as it is a relatively isolated strip of sea.However, in comparison with the general situation of Estonian coastalwaters, the bay water has a high quantity of phosphate and aparticularly high quantity of nitrogen compounds--consequently,predominantly the environment, which is suitable for primaryproductivity limited by phosphorus and the light conditions(Tervisekaitseinspektsioon 2009).

The Parnu River accounts for approximately 80% of the inflow toParnu Bay, bringing annually 2 [km.sup.3] of fresh water to the bay,although the volume of the inner basin is only 1 [km.sup.3] (Suursaar& Tenson 1998). The average river flow rate is 64 [m.sup.3][s.sup.-1], which varies considerably during the year. Maximum ratesremain in the range of 220-330 [m.sup.3] [s.sup.-1], but minimum ratesare approximately 100 times smaller--in the range of 3.5 to 4.7[m.sup.3] [s.sup.-1] (BERNET 2000). Due to this fact the salinity ofwater in Parnu Bay is low, only 3-5 in comparison with the salinity of4.5-6 of the Gulf of Riga (Tervisekaitseinspektsioon 2009).

The characteristic bay bottom type is fine sand, with onlyoccasional stony areas. Due to the effect of waves and currents thewater always contains particles of soft bottom sediments. Additionalsuspended matter is brought also by the Parnu River waters. Morespecifically, large quantities of peat dust are directed to ditches andto the Sauga River and from there to the Parnu River with drainage waterfrom peat excavation areas. Peat dust reaches the bay also in the courseof loading unpacked peat at the mouth of the river.

Measurement methods

Optical monitoring of Parnu Bay and the Gulf of Riga was carriedout at 11 sampling stations (Table 1 and Fig. 1) during the ice-freeperiod in 2006-2007. The study programme involved both in situmeasurements and collection of water samples for subsequent laboratoryanalyses.

Water samples were collected from the surface layer (0.2 m) with astandard water sampler and stored in the dark and cold for less than 7 hbefore filtering. The concentrations of chlorophyll a and phaeophytin awere analysed in duplicate by filtration of water samples (0.5-1 L)through Whatman GF/F-filters. Pigments were extracted from the filtersin 90% ethanol at 75[degrees]C for 5 min and measured spectrometrically,both before and after acidification with dilute hydrochloride acid (ISO1992). Eventually, the determined absorbance values were converted,respectively, to chlorophyll a and phaeophytin a concentrations. For thesake of simplicity, the sum of concentrations is from now on abbreviatedto [C.sub.ph]. The concentration of total suspended matter, [C.sub.s],was measured gravimetrically after filtration of the same amount ofwater through pre-weighed and precombusted (103-105[degrees]C for 1 h)filters (ESS 1993). The attenuation coefficients of light,[c.sup.*]([lambda]) and [c.sub.f.sup.*]([lambda]), were determined,respectively, from unfiltered and filtered water samples. The variable[c.sup.*]([lambda]) was obtained as the differencec([lambda])-[c.sub.d]([lambda]), where c([lambda]) and[c.sub.d]([lambda]) are the beam attenuation coefficients for naturaland distilled water, respectively.

Both fresh and saline waters contain also varying concentrations ofdissolved organic material (DOM), the optically active fraction ofwhich, known as CDOM, plays a great role in the attenuation ofirradiance in the water. Due to the fact that in natural waters CDOM isa rather indeterminate mixture of dissolved organic substances, it isextremely difficult to determine individual organic compounds therein byanalytical methods (Dera 1992). In the present study the amount of CDOMwas characterized by its absorption coefficient at 380 nm,[a.sub.CDOM](380). Unfortunately, with spectrometers such as HitachiU1000 we cannot directly measure [a.sub.CDOM]([lambda]), but attenuationcoefficient spectra of filtered water, [c.sup.*.sub.f]([lambda]). Thevariable [c.sup.*.sub.f]([lambda]) is not identical to[a.sub.CDOM]([lambda]) because some very small inorganic particles andcolloids also pass through the filter and the water may remain ascattering medium even after filtration. However, the differences aresmall, about 2-8% (Sipelgas et al. 2003).

The depth profiles of planar downwelling irradiance ([g.sub.d](z))in the water column were measured using a LI-192 SA sensor (LI-COR,Inc., 1984). This device has an almost ideal quantum response over400-700 nm (photosynthetically active region of the spectrum, PAR). Weused the results of the underwater quantum irradiance for estimating thewidely used diffuse attenuation coefficient, [K.sub.d](PAR) (Dera 1992;Kirk 1994; Arst 2003). The coefficient [K.sub.d](PAR) characterizes theaveraged (over a water column) vertical decrease in natural light in thePAR. For these calculations irradiance values were plotted against depth(for the 0.1-3 m layer) and [K.sub.d](PAR) was found as the exponent ofthe least-squares regression line through these point.

A semi-empirical model described in Arst et al. (2002) and Arst(2003) allows estimation of the spectra of the diffuse attenuationcoefficient, [K.sub.d]([lambda]), on the basis of measured[c.sup.*]([lambda]). From these results the spectra of the attenuationdepth ([z.sub.att]([lambda])) can be determined using the relationship[z.sub.att]([lambda]) = 1/[K.sub.d]([lambda]). This parameter shows thethickness of the surface layer from which 90% of radiation received bysatellite sensors originates. Precisely, the spectral values of[z.sub.att] are needed due to remote sensing sensors working in separatewavebands.

RESULTS

In situ and laboratory measurements

The minimum and maximum values of the optical characteristicsmeasured in Parnu Bay in 2006 and 2007 are presented in Table 2. Thespatial variations in optically active substances (except for totalsuspended matter, whose concentrations are connected with ship traffic,i.e. resuspensions of sediments form the bottom) measured in thetransect from the Parnu River mouth towards the open parts of Parnu Bayare shown in Fig. 2. Some examples on the seasonal change in [C.sub.ph],[C.sub.s] and [a.sub.CDOM](380) for stations PB5 and PB11 in Parnu Bayduring 2007 are shown in Fig. 3. The spectral distributions of theattenuation depth at two stations, PB5 and PB12, are described in Fig.4.

Our database showed both spatial and temporal variation in thewater properties in Parnu Bay. The Parnu River brings large amounts ofcDOM into the bay from the surrounding peat excavation areas, whichmakes the water close to river inflow brownish and usually lesstransparent than the water in deeper parts of Parnu Bay. At the presentstudy typical Secchi depth values near the coast (between stations PB1and PB6) were below 1 m, while in the outer basin and in the Gulf ofRiga [z.sub.SD] was always higher than 1.4 m (Table 2). That spatialbehaviour of the optical parameters [c.sup.*](PAR) and [K.sub.d](PAR)had an almost similar pattern--decrease from the northeastern part ofthe bay towards its southwestern part. The concentrations of [C.sub.ph]generally ranged from 4 to 12 mg [m.sup.-3], however, during the vernalalgal bloom in April 2007 the exceptional maximums (25.5-31.8 mg[m.sup.-3]) were observed at stations PB6, PB7, PB11 and PB12 (Fig. 2A).The amount of suspended matter in the water was mostly between 8 and 14g [m.sup.-3]. Extremely high values of [C.sub.s] (from 23.2 to 49.0 g[m.sup.-3]) at stations PB5-PB7 were influenced by the loading ofunpacked peat at the mouth of the river on 2 August 2007. However, thevariations in the concentration of total suspended matter were alsocaused by ship traffic, i.e. resuspension of sediments form the bottom.As expected, there was also a very pronounced decrease in CDOM from theriver mouth towards the offshore stations in Parnu Bay and the Gulf ofRiga. In April and May, when freshwater discharge by the Parnu River washighest, the values of [a.sub.CDOM](380) were between 4.6 and 31.8[m.sup.-1], while in September [a.sub.CDOM](380) varied only from 2.52to 10.2 [m.sup.-1] (Fig. 2B).

The seasonal behaviour of water parameters at different stationswas to some extent even opposite (Fig. 3). The higher vernal[a.sub.CDOM](380) values at station PB5 compared to those at offshorestation PB11 resulted from intensive inflow from the Parnu River to thebay. The chlorophyll a (including its metabolite phaeophytin a)concentration at both stations increased towards late summer, however,at station PB11 intensive algal bloom occurred also in April. Duringspring and summer the concentrations of the total suspended matter atstation PB5 varied generally between 4.7 and 5.6 g [m.sup.-3], but bythe beginning of September the value of [C.sub.s] had risen up to 13.5 g[m.sup.-3]. The slightly higher values of [C.sub.s] (10.2 g [m.sup.-3])on 20 June 2007 were caused by yacht racing in the transect PB1-PB6. Inthe Parnu Bay outer basin the concentrations of total suspended matterwere somewhat higher--from 8.6 to 12.7 g [m.sup.-3].

In order to estimate the values and variability of the attenuationdepth [z.sub.att], first the [K.sub.d]([lambda]) spectra were calculatedfrom measured [c.sup.*]([lambda]) by using a semi-empirical modeldescribed in Arst et al. (2002). Subsequently, the correspondingattenuation depth spectra were determined by using the relationship[z.sub.att]([lambda]) = 1/[K.sub.d]([lambda]). Our computation resultsfor stations PB5 and PB12 (respectively, in the fluvial region and inthe Gulf of Riga) are shown in Fig. 4. The data of [z.sub.att]([lambda])permit us to draw some important conclusions about the possibilities ofusing optical remote sensing. We can see that at station PB12 remotesensing yields information from the surface layer down to 2.3 m, but atstation PB5 only from the upper 0.34-0.8 m (Fig. 4). As the thickness ofthe informative layer varies also with wavelength, the colour indicesand other water optical parameters of remote sensing spectra would bemost effective if they were chosen at the wavelengths corresponding tomaximum values of attenuation depth.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

As is known, the MODIS standard algorithm for the diffuseattenuation coefficient of seawater at 490 nm ([K.sub.d](490)) is widelyused for describing the spatial variation in the water properties. Table3 shows the results of in situ measurements of [K.sub.d](490) at most ofthe sampling stations.

Regression analysis

We used a linear regression program Microsoft Data Analysis. Thedetermination of coefficients and significance of various combinationsof parameters are presented in Table 4. It turned out that for[c.sup.*](PAR) vs [z.sub.SD] and [K.sub.d](PAR) vs [z.sub.SD] the bestregression was nonlinear:

[c.sup.*](PAR) = 3.72[z.sup.1.02.sub.SD], (1)

where [R.sup.2] = 0.793 and p < 0.0001;

[K.sub.d](PAR) = 1.48[z.sup.-0.75.sub.SD] (2)

where [R.sup.2] = 0.655 and p < 0.0001.

According to the results, the main factor influencing the lightattenuation in water was CDOM. It overshadows the relationships betweenthe radiation characteristics and organic/inorganic particles. Asurprisingly weak relationship between the concentrations of [C.sub.ph]and [C.sub.s] could probably be explained by a significant contributionof mineral particles in the total suspended matter. There were someadditional negative regressions, all containing the suspended mattermentioned above ([C.sub.s] vs [a.sub.CDOM], [c.sup.*](PAR) vs [C.sub.s]and [K.sub.d](PAR) vs [C.sub.s]). The reason for this is temporal andspatial irregularity of [C.sub.s], caused by loading unpacked peat atthe mouth of the Parnu River, undulation and ship traffic.

Due to the facts that ocean colour sensors SeaWiFS and MODISprovide [K.sub.d](490) as standard Level 2 product(http://oceancolor.gsfc.nasa.gov/) and this parameter is also widelyused for describing the water properties, we studied its regressionswith [a.sub.CDOM](380), [C.sub.Chl] and [C.sub.s] (all obtained from insitu measurements in 2006-2007). The numerical values of [K.sub.d](490)were calculated from measurements of c*(490) by using a special modeldeveloped in Arst et al. (2002). The results of the regression analysisare presented in Tables 5 and 6. For recognition of the cases withnegative correlation in these tables the values of R are presented(instead of [R.sup.2]). Despite the small number of regression points,these results allow of some useful conclusions. It is clear that inParnu Bay the values of [K.sub.d](490) are strongly influenced by CDOMthat overshadows the actual relationship between Kd(490) and Cph. Insome cases it has led even to negative correlation coefficients for[K.sub.d](490) vs [C.sub.ph] (Tables 5 and 6). These results support theopinion that (a) in CDOM-rich coastal waters the spatial distribution of[K.sub.d](490) allows determination of the corresponding values of[a.sub.CDOM](380) with high accuracy and (b) in these waters[K.sub.d](490) is unsuitable for estimating [C.sub.ph] and [C.sub.s].However, in case of remote sensing measurements we can derive[a.sub.CDOM](380) from [K.sub.d](490) only when the satellite values ofKd(490) are reliable enough.

MODIS level 1 data with 250 m resolution were used for illustrativecomparison of spatial and temporal variations in the water properties inParnu Bay and the Gulf of Riga (Fig. 5). According to these results, thewater properties in Parnu Bay are very variable and differ markedly fromthose in the open area of the Gulf of Riga. Obviously, opticallysignificant substances vary in higher levels inside the bay and allrelative concentrations are decreasing outside the Parnu Bay area. Anattempt to perform quantitative analysis with the purpose of estimatingthe concentrations of different optically active substances separatelygave statistically incorrect results.

DISCUSSION

We compared our in situ measurement data with some others(different time periods, different sampling stations). In Arst et al.(1993, 1994) three field trips to the Parnu Bay inner basin (8 samplingstations) were carried out in May, June and October in 1991. Duringsummer months the values of Cchl were in the range of 3.9-12.9 mg[m.sup.-3]. This means that generally the chlorophyll concentrationswere in a good accordance with the present study, but due to the vernalalgal bloom in April in 2007 our maximum values of [C.sub.chl] were muchhigher. In the present study the concentration of suspended matterranged from 3.7 to 49 g [m.sup.-3] (all measurements), while on 6 June1991 the values of [C.sub.s] were between 11 and 38 g [m.sup.-3]. Thisfact proves once more the temporal and spatial irregularity of suspendedmatter, which is influenced by undulation and ship traffic in the bay.Similarly to our investigation, the beam attenuation coefficientdecreased from the river mouth towards the open parts of Parnu Bay, butthe absolute values of [c.sup.*.sub.PAR] were noticeably higher (atstations close to the coast even up to 20 [m.sup.-1]). This impliesincrease in the water transparency from the year 1991 to 2007.

Another database for describing the bio-optical properties of ParnuBay on 5-6 June 2001 is also available (L. Sipelgas, pers. comm. 2002).The measurements were carried out at two depths (0 and 2 m) at 10stations over the whole Parnu Bay (from the mouth of the Parnu River tothe station with the coordinates 58[degrees]10'N and24[degrees]17'E). The values of [C.sub.chl], [C.sub.s],[c.sup.*](PAR) and [a.sub.CDOM](380) varied, respectively, in the ranges28-133 mg [m.sup.-3], 4.4-14.4 g [m.sup.-3], 2.7-9.4 [m.sup.-1] and5.3-20.5 [m.sup.-1]. Extraordinarily high chlorophyll a concentrations,mainly at the stations in the inner basin of the bay, were obviouslycaused by strong phytoplankton bloom. Secchi depth varied between 0.75and 2.5 m and [K.sub.d](PAR) from 0.8 to 2.1 [m.sup.-1]. The resultslead to a conclusion that during the past decades the optical propertiesof Parnu Bay have been almost unchanged, however, the comparison iscomplicated due to a large variability of these properties in time andspace.

For comparison of the results obtained in the river mouth areas inParnu Bay with those from the other water bodies we chose the Kymijokiand Porvoonjoki estuaries (both rivers discharge into the Gulf ofFinland). However, we had a rather small database collected on 9 and 12August 2005 as well as on 13-14 June 2006. In the Kymijoki themeasurements were carried out in a transect, which starts with thecoordinates 60[degrees]27'N and 26[degrees]28'E and ends atthe station with coordinates 60[degrees]23'N and26[degrees]33'E. In the Porvoonjoki the respective coordinates were60[degrees]22'N and 25[degrees]40'E, and 60[degrees]20'Nand 25[degrees]38'E. As expected, a pronounced decrease in CDOMfrom the river mouths towards the offshore stations was observed also inFinnish estuaries. In August 2005 the values of aCDOM(380) were between3.7 and 7.4 [m.sup.-1] in the Porvoonjoki, while in the Kymijoki the[a.sub.CDOM](380) varied only from 5.2 to 6.4 [m.sup.-1]. In June 2006the absorptions by CDOM in both estuaries were in the range 5.9-11.3[m.sup.-1]. According to these values, Finnish estuaries, are to someextent different from Parnu Bay. They are less 'yellow' (themaximum of [a.sub.CDOM](380) differs about 2.5 times). However,sometimes similar values of [a.sub.CDOM](380) were observed: in June2006 Parnu Bay gave [a.sub.CDOM](380) between 2.4 and 12.6 [m.sup.-1],in June 2005 in the Porvoonjoki estuary this variation was 5.9-11.3[m.sup.-1]. Except for the phytoplankton blooms, the chlorophyllconcentration in Parnu Bay was generally below 12 mg [m.sup.-3]. At thestations of the estuary of the Porvoonjoki it was considerably higher(in June 2006 from 8.7 to 30.4 mg [m.sup.-3] and in August 16.2-19.9 mg[m.sup.-3]). In the Kymijoki estuary the values of [C.sub.ph] variedbetween 10.2 and 14.2 mg [m.sup.-3]. The amount of suspended matter inthe water was almost similar in all three regions--in Parnu Bay thevalues of [C.sub.s] were mostly between 8 and 14 g [m.sup.-3] and inFinnish estuaries they ranged from 8.1 to 17.5 g [m.sup.-3].

The ENVISAT satellite, which carries the medium resolution imagingspectrometer (MERIS) sensor, was launched on 1 March 2002. The MERISsensor characteristics have been developed according to opticallycomplex water properties, making it the first sensor for monitoring themulticomponental waters. MERIS (like other earth observation satellitesensors) has high measurement frequencies, passing over the regions ofinterest each day and greatly increasing the chances of obtaining usefulcloud-free images. The full resolution (~300 m) data provide the highspatial resolution available from a satellite sensor, whereas newdevelopments for optically complex (Case-2 Regional, Boreal Lakes,Eutrophic Lakes) processors are still going on. The new processors haveadditional products such as downwelling irradiance attenuationcoefficient, absorption coefficients of phytoplankton, CDOM, and acoefficient of all particles after bleaching and scattering (Doerffer& Schiller 2008). To illustrate spatial variation in waterproperties, MERIS image from 6 July 2002, which is a good example forcharacterizing the situation in Parnu Bay, is presented in Fig. 6. UsingMERIS standard products, we can observe very complex spatialdistribution of the water properties, which could not be visualized byin situ measurements. Clearly visible is a highly turbid water area(Fig. 6 top left) flowing out of the bay. This feature is notable on thesuspended matter and CDOM products, but not on the algal_2 product. Thehigh amount of CDOM in Parnu Bay overshadows the signal of chlorophylla, which leads to invalid reflectances used for calculating chlorophylla product algal_2. As different times of measurements and differentparameters were used, the quantitative comparison of Figs 5 and 6 isimpossible. However, in both figures the multi-coloured Parnu Bayregion, different from that in the open part of the Gulf of Riga, isclearly seen.

Kutser et al. (2009) analysed the results of remote sensinginvestigations by a satellite sensor Advanced Land Imager (ALI) in theeastern part of the Baltic Sea.

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

The conclusion was that optical properties of the Baltic Sea aredominated by CDOM. Strong absorption of light by CDOM at shorterwavelengths is probably the main reason why standard chlorophyll aretrieval algorithms fail in the Baltic Sea. Thus, according to thisstudy, CDOM dominates not only in Parnu Bay, but also in many coastalregions of the Baltic.

CONCLUSIONS

During the ice-free period in 2006-2007 Parnu Bay was characterizedby marked spatial and seasonal variation in the water properties. A verypronounced decrease in CDOM occurred from the river mouth towards theoffshore stations in Parnu Bay and the Gulf of Riga. In April and May,when freshwater discharge of the Parnu River was highest, the[a.sub.CDOM](380) values were between 4.6 and 31.8 [m.sup.-1], whilst inSeptember [a.sub.CDOM](380) varied only from 2.52 to 10.2 [m.sup.-1].Phytoplankton contributed to light attenuation of seawater primarilyduring algal bloom periods, when [C.sub.ph] increased rapidly from itsgeneral values of 4-12 mg [m.sup.-3] to its vernal peak of 25.5-31.8 mg[m.sup.-3]. The temporal and spatial irregularity of Cs was caused bythe loading of unpacked peat at the mouth of the Parnu River as well asby undulation and ship traffic in the bay.

The results of correlation analysis (data obtained in situ) showedthat the main factor influencing light attenuation in the water is CDOM.It overshadows the relationships between the radiative characteristicsand organic/inorganic particles. In some cases it has led even tonegative correlation coefficients for [K.sub.d](490) vs [C.sub.ph].

Due to the variability of the water properties, Parnu Bay seems tobe an interesting subject of investigations by optical remote sensing.However, in summer the spectral values of the attenuation depth in theinner basin of the bay were mostly below 1 m (especially in theblue--green region of the spectrum) and only in autumn the values[z.sub.att]([lambda]) > 1.5 m were observed. Also the values of[K.sub.d](490) that are often used in satellite results analysis arestrongly influenced by CDOM, which overshadows the actual relationshipbetween [K.sub.d](490) and [C.sub.ph].

The main conclusions concerning the interpretation of the satellitedata in CDOM-rich coastal waters are: (a) the spatial distribution of[K.sub.d](490) allows determination of the corresponding values of[a.sub.CDOM](380) with a high accuracy; (b) in these waters[K.sub.d](490) is unsuitable for estimating [C.sub.Chl] and [C.sub.s].However, in case of remote sensing measurements we can derive[a.sub.CDOM](380) from [K.sub.d](490) only when the satellite values of[K.sub.d](490) are reliable enough. Thus, the remote sensinginvestigation of Parnu Bay is still problematic and new algorithms arenecessary.

doi: 10.3176/earth.2011.2.04

Acknowledgements. We are indebted to Juri Tenson and Ivar Tensonfor their help with field measurements. We thank anonymous referees fortheir valuable comments. Financial support to this investigation wasprovided by the Estonian Ministry of Education and Research (targetedfunding project 0712699s05), Estonian Science Foundation (grant 5594)and Vaisala Foundation (Finland).

REFERENCES

Arst, H. 2003. Optical Properties and Remote Sensing ofMulticomponental Water Bodies. Springer & Praxis-Publishing,Chichester, UK, 231 pp.

Arst, H., Kutser, T., Laesson, L. & Lukk, T. 1993. Opticalmeasurements in Parnu Bay. Proceedings of the Estonian Academy ofSciences, Biology, Ecology, 3, 27-37.

Arst, H., Kutser, T., Laesson, L. & Lukk, T. 1994. Estimationof the quality of Parnu Bay water by means of optical methods. InProceedings of 18th Conference Baltic Oceanographers, St. Petersburg, 1,pp. 90-100. St. Petersburg.

Arst, H., Erm, A., Reinart, A., Sipelgas, L. & Herlevi, A.2002. Calculating irradiance penetration into water bodies from themeasured beam attenuation coefficient II: Application of improved modelto different types of lakes. Nordic Hydrology, 33, 207-226.

Austin, R. W. & Petzold, T. J. 1981. The determination of thediffuse attenuation coefficient of sea water using the coastal zonecolor scanner. In Oceanography from Space (Gower, J. F. R., ed.), pp.239-256. Plenum Press.

BERNET. 2000: Aquatic Monitoring and Assessment. Volume I,Synthesis. BERNET Theme Report. Fyn County, Odense, Denmark, 50 pp.http://www.dwaf.gov.za/projects/ eutrophication/Website%20Survey/Baltic/WG_Report_5_vol_I.pdf [accessed 21 March 2011].

Darecki, M. & Stramski, D. 2004. An evaluation of MODIS andSeaWiFS bio-optical algorithms in the Baltic Sea. Remote Sensing ofEnvironment, 89, 326-350.

Dera, J. 1992. Marine Physics. PWN, Warszawa and Elsevier,Amsterdam, 516 pp.

Doerffer, R. & Schiller, H. 2008. Algorithm Theoretical BasisDocument: MERIS--Lake Water Algorithm for BEAM.http://www.brockmann-consult.de/beamwiki/download/attachments/1900548/ATBD_lake_water _RD20080610.pdf?version=1 [accessed 18 March 2011].

[ESS] Environmental Sciences Section. 1993. ESS Method 340.2: TotalSuspended Solids, Mass Balance (Dried at 103-105 [degrees]C), VolatileSuspended Solids (Ignited at 550 [degrees]C). Environmental SciencesSection, pp. 3-189-3-192.

[ISO] ISO TC 147/SC 2. 1992 (E). ISO 10260:1992, WaterQuality--Measurement of Biochemical Parameters SpectrometricDetermination of the Chlorophyll-a Concentration. Geneva, Switzerland,ISO, 6 pp.

Kirk, J. T. O. 1994. Light and Photosynthesis in AquaticEcosystems. Cambridge University Press, 509 pp.

Kratzer, S., Brockmann, C. & Moore, G. 2008. Using MERIS fullresolution data to monitor coastal waters--a case study fromHimmerfjarden, a fjord-like bay in the northwestern Baltic Sea. RemoteSensing Environment, 112(5), 198-199.

Kutser, T., Paavel, B., Metsamaa, L. & Vahtmae, E. 2009.Mapping coloured dissolved organic matter concentration in coastalwaters. International Journal of Remote Sensing, 30(22), 5843-5849.

Mueller, J. L. 2000. SeaWiFS algorithm for the diffuse attenuationcoefficient, K(490), using water-leaving radiances at 490 and 555 nm. InSeaWiFS Postlaunch Calibration and Validation Analyses, Part 3 (NASATM-2000-206892, ed.), pp. 24-27. Greenbelt, MD: NASA GSFC.

Platt, T. & Sathyendranath, S. 1988. Ocean primary production.Estimation by remote sensing at local and regional scales. Science, 241,1613-1620.

Sipelgas, L., Arst, H., Kallio, K., Oja, P. & Soomere, T. 2003.Optical properties of dissolved organic matter in Finnish and Estonianlakes. Nordic Hydrology, 34(4), 361-386.

Suursaar, U. & Tenson, J. 1998. Hydrochemical regime andproductivity of the Parnu Bay in 1968-1996. EMI Report Series, 9,91-117.

Tervisekaitseinspektsioon. 2009. Parnu keskranna suplusvee profiil[Bathing Water Quality at the Beach of Parnu Keskrand], 36 pp.http://www.terviseamet.ee/fileadmin/dok/Keskkonr^tervis/vesi/suplus/Profiilid/parnu_rand_profiil.pdf [inEstonian; accessed 21 March 2011].

Birgot Paavel (a), Helgi Arst (a), Liisa Metsamaa (a), Kaire Toming(a) and Anu Reinart (b)

(a) Estonian Marine Institute, University of Tartu, Maealuse 14,12618 Tallinn, Estonia; [emailprotected]

(b) Tartu Observatory, 61602 Toravere, Tartumaa, Estonia

Received 28 October 2010, accepted 11 March 2011

Table 1. Sampling stations in Parnu Bay (field tripswere carried out from April to September in 2006-2007)Station Latitude Longitude Number N, [degrees] E, [degrees] of tripsPB1 58.386 24.489 8PB2 58.376 24.476 9PB5 58.367 24.457 10PB6 58.354 24.442 10PB7 58.326 24.410 9PB8 58.306 24.392 2PB9 58.293 24.388 1PB10 58.251 24.342 2PB11 58.216 24.305 8PB12 58.059 24.952 7PB14 58.2 23.4 3Table 2. Minimum and maximum values of water characteristicsmeasured in Parnu Bay during the ice-free period in2006-2007. The denotation N in the last row is the number ofindividual results of each water parameterStation [Z.sub.SD] c * (PAR), [K.sub.d](PAR), m [m.sup.-1] [M.sup.-1]PB1 0.6-1.6 3.0-6.5 1.4-2.3PB2 0.6-1.2 3.2-6.5 1.6-2.3PB5 0.5-1.5 3.1-6.5 1.1-2.0PB6 0.4-1.5 3.3-4.6 0.8-1.8PB7 0.8-1.7 1.7-5.6 0.8-1.9PB8 1.5 1.5-3.7 0.8-1.0PB9 1.7 2.8 0.78PB10 2 1.8-2.0 0.83PB11 1.3-3.1 1.0-2.5 0.5-1.6PB12 1.5-3.5 0.8-2.1 0.6-1.2PB14 2.3-4.3 0.8-1.4 --N 58 60 33 [a.sub. CDOMStation [C.sub.ph], [C.sub.s], (380)], mg [m.sup.-3] g [m.sup.-3] m-1PB1 5.7-12.4 3.9-10.9 7.8-31.8PB2 6.7-17.6 3.8-14.1 7.5-31.8PB5 6.6-15.1 3.7-26.7 5.5-31.4PB6 5.2-31.8 9.9-49.0 2.9-14.1PB7 2.6-30.4 8.7-23.2 2.2-11.4PB8 2.8-6.6 9.4-11.2 2.2-5.0PB9 2.7 8.9 4.2PB10 2.1-6.7 9.4-9.8 2.4-4.0PB11 1.5-25.5 8.0-14.7 2.3-5.8PB12 5.6-28.2 7.8-13.4 2.3-3.9PB14 3.2-12.0 7.4-17.5 2.1-2.6N 67 67 65Table 3. Values of [K.sub.d](490) (in [m.sup.-1])determined in the region of Parnu Bay from in situmeasurements in 2006-2007. The stations where thenumber of field trips was less than 6 were left out StationDate PB1 PB2 PB5 PB631.05.2006 3.83 3.78 3.62 2.2420.06.2006 -- 3.24 1.91 1.9031.07.2006 6.44 -- 2.28 1.9924.04.2007 7.11 7.32 7.20 3.5521.05.2007 -- 6.72 6.43 3.1020.06.2007 4.22 4.17 3.63 2.153.07.2007 2.44 2.52 2.43 1.855.08.2007 2.16 2.42 2.78 --5.09.2007 2.22 2.63 2.35 1.98 StationDate PB7 PB11 PB1231.05.2006 -- -- --20.06.2006 1.60 0.77 --31.07.2006 1.09 -- --24.04.2007 2.46 1.53 1.0621.05.2007 2.70 1.75 0.8820.06.2007 2.53 0.93 0.793.07.2007 1.52 1.33 1.295.08.2007 1.13 0.92 0.705.09.2007 1.89 1.31 0.69Table 4. Determination coefficients ([R.sup.2]) and significance(p) of the linear regressions obtained for severalbio-optical parameters of Parnu BayRegression [R.sup.2] p Comments[C.sub.ph] vs [C.sub.s] 0.077 0.028[C.sub.ph] vs [a.sub.CDOM] 0.022 0.247[C.sub.s] vs [a.sub.CDOM] 0.268 < 0.0001 Negative[c.sup.*](PAR) vs [C.sub.ph] 0.035 0.143[c.sup.*](PAR) vs [C.sub.s] 0.029 0.188 Negative[c.sup.*](PAR) vs [a.sub.cDOM] 0.662 < 0.0001[K.sub.d](PAR) vs [C.sub.ph] 0.110 0.084[K.sub.d](PAR) vs [C.sub.s] 0.029 0.386 Negative[K.sub.d](PAR) vs [a.sub.cDOM] 0.568 < 0.0001[c.sup.*](PAR) vs [K.sub.d](PAR) 0.745 < 0.0001Table 5. Correlation characteristics ([K.sub.d](490) vs differentoptically significant substances) obtained at sampling stations PB5,PB7 and PB11 in 2006-2007. The station PB5 is located close to theshore (near the pier), PB7 in the central part of Parnu Bay andPB11 in the western part of the bay, close to the Gulf of RigaRegression PB5, N = 9 PB7, N = 8 R p R p[K.sub.d] 0.971 <0.0001 0.792 0.019 (490) vs [a.sub.CDOM] (380)[K.sub.d] -0.176 0.650 0.618 0.102 (490) vs [C.sub.chl][K.sub.d] (490) -0.491 0.180 0.683 0.062 vs [C.sub.s]Regression PB11, N = 7 R p[K.sub.d] 0.799 0.031 (490) vs [a.sub.CDOM] (380)[K.sub.d] 0.486 0.268 (490) vs [C.sub.chl][K.sub.d] (490) 0.244 0.597 vs [C.sub.s]Table 6. Correlation characteristics ([K.sub.d](490) vs differentoptically significant substances) obtained on 27 April 2007and 5 September 2007 in Parnu Bay (all stations together)Regression 27.04.2007 05.09.2007 R p R p[K.sub.d](490) vs 0.999 <0.0001 0.794 0.01 [a.sub.CDOM](380)[K.sub.d](490) vs [C.sub.Chl] -0.897 0.006 0.741 0.022[K.sub.d](490) vs [C.sub.s] -0.874 0.01 0.366 0.419