Particle dynamics in the Eastern Mediterranean Sea - Semantic Scholar

1 downloads 0 Views 3MB Size Report
Nov 23, 2007 - of Psyttaleia Island. Krasakopoulou and Karageor- gis (2005) have reported POC concentrations in the vicinity of the outfall often higher than ...
ARTICLE IN PRESS

Deep-Sea Research I 55 (2008) 177–202 www.elsevier.com/locate/dsri

Particle dynamics in the Eastern Mediterranean Sea: A synthesis based on light transmission, PMC, and POC archives (1991–2001) Aristomenis P. Karageorgisa,, Wilford D. Gardnerb, Dimitris Georgopoulosa, Alexey V. Mishonovc, Evangelia Krasakopouloua, Christos Anagnostoua a

Hellenic Centre for Marine Research, 46.7 km Athens-Sounio Avenue, Mavro Lithari, 19013 Anavyssos, Greece b Department of Oceanography, Texas A&M University, College Station, TX 77843, USA c NODC/NOAA, 1315 East West Highway, Silver Spring, MD 20910-3282, USA Received 28 June 2007; received in revised form 7 November 2007; accepted 12 November 2007 Available online 23 November 2007

Abstract During the last two decades light transmission (LT) data have been collected routinely in the Eastern Mediterranean Sea, within the framework of several research projects. A procedure was developed to obtain beam attenuation coefficient due to particles (cp) at 660–670 nm adjusted for variations in mid-depth ‘clear’ water and instrumental drifts. Data from 3146 stations occupied between 1991 and 2001 were converted to a common format for the analysis of particulate matter (PM) temporal and spatial distribution patterns. The data were separated into ‘wet’ (December–May) and ‘dry’ (June–November) periods. The horizontal distribution of beam cp at various depths revealed clearly higher values in the surface nepheloid layer (SNL) in the vicinity of river mouths during the ‘wet’ period, whilst the increase was negligible during the ‘dry’ period. In contrast, the bottom nepheloid layer (BNL; 1–10 m above bottom) appeared to be turbid throughout the year, particularly on the continental shelves receiving riverine discharge. This feature is attributed to resuspension and advection of recently deposited bottom sediments due to waves and currents. However, the Eastern Mediterranean as a whole is impoverished in PM in the water column, particularly at depths 4200 m. The behavior of surface-water cp revealed a strong relationship to mesoscale dynamic features. Cyclonic eddies, which upwell nutrient-rich waters toward the surface, favor primary production, which was identified as elevated beam cp values. Beam cp was correlated with PM concentration (PMC) and particulate organic carbon (POC) concentration obtained by bottle sampling. Although there were regional differences in the correlations, no significant seasonal variations were observed. Two generic equations were generated that can be used for a first-order estimate of PMC and POC from historical LT measurements conducted in the area, provided that data are handled according to the proposed methodology. r 2007 Elsevier Ltd. All rights reserved. Keywords: Light transmission; Beam attenuation; Particulate matter; Particulate organic carbon; Eastern Mediterranean Sea

Corresponding author. Tel.: +30 2291076369; fax: +30 2291076347.

E-mail addresses: [email protected] (A.P. Karageorgis), [email protected] (W.D. Gardner), [email protected] (D. Georgopoulos), [email protected] (A.V. Mishonov), [email protected] (E. Krasakopoulou), [email protected] (C. Anagnostou). 0967-0637/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.dsr.2007.11.002

ARTICLE IN PRESS 178

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

1. Introduction Ever since the 1970s, numerous particulate matter (PM) studies have been undertaken over the world’s oceans, providing on the temporal and spatial variations of the PM field, aimed at a better understanding of PM optical characteristics and its distribution patterns in the ocean (Eittreim et al., 1976; Biscaye and Eittreim, 1977; Spinrad et al., 1983; Gardner et al., 1985; Spinrad, 1986; Richardson, 1987), shelf/slope exchange processes (Biscaye et al., 1994; McCave et al., 2001; McCave and Hall, 2002), submarine canyon role and dynamics (Drake, 1971; Baker and Hickey, 1986; Hickey et al., 1986; Gardner, 1989a), sediment resuspension mechanisms (Moody et al., 1987; Gardner, 1989b), relation between hydrography and nepheloid layers (Cacchione and Drake, 1986; Durrieu de Madron et al., 1990; Palanques and Biscaye, 1992; Durrieu de Madron, 1994; Puig and Palanques, 1998; Castaing et al., 1999; Durand et al., 2002), and biogeochemical cycles (Monaco et al., 1990; Gardner et al., 1993, 1995). The use of transmissometers and other types of optical instruments measuring light attenuation (scattering and absorption), light scattering, or optical backscatter in seawater greatly supported this research, obtaining data on vertical profiles or occasionally as time series. The inherent problem of such kinds of station-based measurements is relatively poor spatial coverage. In most cases, research projects aim to study specific processes in selected marine regions, whilst very few of them have covered adequately large sectors of the ocean, e.g. the entire Atlantic Ocean (Eittreim et al., 1976; Biscaye and Eittreim, 1977), the Goban Spur (McCave et al., 2001), the Pacific (Kawahata, 2002), and the Yellow Sea (Park et al., 2001). Studies employing optical turbidity measurements in the Western Mediterranean Sea have provided comprehensive information on PM dynamics on the Spanish continental margin (Puig and Palanques, 1998; Puig et al., 2004) and in the Gulf of Lions (Durrieu de Madron et al., 1990; Monaco et al., 1990; Durrieu de Madron, 1994; Lapouyade and Durrieu De Madron, 2001; Frignani et al., 2002). In the Eastern Mediterranean, despite the collection of light transmission (LT) data since the early 1990s, an overall picture of PM distribution patterns is missing. However, results from several small-scale investigations have been published, usually covering coastal areas and regional seas (Durrieu de Madron et al., 1992; De Lazzari et al., 1999; Karageorgis et al., 2000,

2003; Karageorgis and Anagnostou, 2001, 2003; Karageorgis and Stavrakakis, 2005; Krasakopoulou and Karageorgis, 2005), where LT readings were collected routinely as part of CTD casts. Nevertheless, numerous LT data remain unused to date because of lack of expertise in studying such parameters in particular projects where, e.g. only temperature and salinity were required, but the importance of LT data was recognized. This paper aims to compile all available LT measurements conducted in the Eastern Mediterranean at the Hellenic Center for Marine Research (HCMR) (Fig. 1), in order to assess general temporal and spatial distribution trends, as well as their relation to hydrographic features of the region. To accomplish this, a series of methodological steps used for data integration are presented. Apart from the painstaking, tedious task of data collection and handling, there is a specific issue addressed here that deserves particular attention: the merger of LT measurements obtained over a 11-year time span, while ensuring data comparability against instrument drift and any type of shifts or errors generated during sampling. Moreover, PM concentration (PMC) and particulate organic carbon (POC) concentration data have been collected in parallel. The latter parameters are physically linked to light attenuation and their relationships are documented. 2. Regional setting 2.1. Morphology of the Eastern Mediterranean The Eastern Mediterranean Sea is a relict of the ancient Tethys Ocean, and its morphology has been formed by continuous geodynamic processes during 50–70  106 years, largely the subduction of the African tectonic plate under the Aegean microplate (Sakellariou et al., 2005). The seafloor morphology is extremely complex (Fig. 1). The North Aegean is one of the few areas where wide continental shelves have developed, dipping rapidly into deep (1400 m) basins and alternating with shallower plateaus toward the Central Aegean, which is dominated by the Cyclades Island arc. The South Aegean includes the Cretan Sea, where water depths up to 2500 m are found. The Ionian Sea in its northern sector is the continuation of the Adriatic Sea and is characterized by an extensive shelf, while its southern sector is very steep, characterized by deep canyons and basins more than 4000 m deep. Thirty miles off SW Peloponnisos one finds the deepest

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

179

Fig. 1. Simplified bathymetry of the Eastern Mediterranean Sea, country borders (orange line), major rivers (black lines) and light transmission stations (white dots). Rivers: 1—Pineios, 2—Aliakmon, 3—Loudias, 4—Axios, 5—Gallikos, 6—Strymon, 7—Nestos, 8— Evros, 9—Kalamas, 10—Acheron, 11—Acheloos, 12—Arachthos, 13—Louros, 14—Alfeios, 15—Pamisos, 16—Evrotas, 17—Spercheios, 18—Asopos, 19—Mornos, 20—Evvinos, 21—Kara Menderes, 22—Bakirc- ay, 23—Gediz, 24—Ku¨c- u¨k Menderes, 25—Bu¨yu¨k Menderes.

basin of the Mediterranean (Vavilov Deep), where water depth is 5100 m (Sakellariou et al., 2005). The Levantine Basin is the second largest basin of the Eastern Mediterranean Sea (after the Ionian Sea) and is bounded by Asia Minor and the NE African mainland. In this paper, data collected in the Levantine Basin occupy only its NW sector, including the Rhodes Basin and the Cretan Passage (Fig. 1). Overall, the study area (central sector of the Eastern Mediterranean) covers 550,000 km2, with continental shelves (water depth o200 m) covering about 90,000 km2. 2.2. Sources of PM in the Eastern Mediterranean River discharge, biological production and the atmosphere are the main sources of PM in the

Eastern Mediterranean. Aeolian inputs include dust aerosols created by wind erosion of soils in the Balkans peninsula and Turkey, but a considerable dust input originates from the Sahara desert, transported by dust-raising southerlies (SW to SE; see Guerzoni et al., 1999). A 4-year experiment employing 6 dust traps distributed over Crete Island (South Aegean Sea) concluded that most of the dust passing Greece is carried from North Africa; in Crete, the mean annual deposition rate was calculated at 21.3 g m2 yr1 (Nihle´n et al., 1995); annual average input of insoluble particles into the study area (550,000 km2) was therefore estimated at 11.7  106 t yr1. Kubilay and Saydam (1995) also conclude that the Eastern Mediterranean Sea is substantially influenced by Sahara dust aerosols, especially during spring and autumn. In the Western

ARTICLE IN PRESS 180

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

Mediterranean, after comprehensive studies Martin et al. (1989) suggest that atmospheric input of particulates in the basin is equally important as the river input, whilst Guerzoni et al. (1999) argue that if 2/3 of the riverine input remains on the continental shelf (water depth o100 m) then the atmospheric contribution is up to 60% for the Mediterranean as a whole. The major rivers flowing into the Eastern Mediterranean (Fig. 1) discharge into the North Aegean Sea, namely from west to east: Pineios, Aliakmon, Loudias, Axios, and Gallikos (Thermaikos Gulf, NW Aegean Sea), Strymon, Nestos and Evros. In Western Greece, the rivers Kalamas, Acheron, and Acheloos discharge into the Ionian Sea; Arachthos and Louros discharge into the semienclosed Amvrakikos Gulf, which communicates with the Ionian Sea through a narrow strait. The rivers Alfeios, Pamisos, and Evrotas drain Peloponnisos, discharging into the Ionian Sea. A number of smaller rivers discharge into semienclosed gulfs of the Hellenic mainland, namely Spercheios, Asopos, Mornos, and Evvinos and others, whilst numerous streams, with small catchments (including Turkish rivers draining Asia Minor, i.e. Kara Menderes, Bakirc- ay, Gediz, Ku¨c- u¨k Menderes, and Bu¨yu¨k Menderes), contribute to the overall freshwater input to the sea. A common characteristic of the Hellenic rivers is their variable behavior between winter/spring and summer/autumn. High freshwater discharge periods begin generally in December, following the commencement of the rainy season, and culminate around May, when snow melts on the mountains. This seasonality results in relatively constant flow during the ‘wet’ period (December–May) and intermittent flow during the ‘dry’ period (June– November). Because of increasing needs for freshwater during the latter period even large rivers become almost dry during July and August. Freshwater outflow has decreased significantly because thousands of wells now pump water from underground aquifers, mainly for irrigation. Furthermore, dams have been constructed on many rivers over the past 40 years for hydroelectric power production and irrigation/watering purposes. Poulos and Chronis (1997), based on water discharge from 16 large and 8 smaller Greek rivers, have estimated that some 35,000  106 m3 of freshwater flows into the Eastern Mediterranean. Turkish rivers discharge an additional 1250  106 m3 of freshwater into the Aegean Sea (Poulos et al., 1997).

Suspended solids of riverine origin are injected into the sea following the temporal variations of freshwater discharge. For example, PMC near the Axios River mouth in Thermaikos Gulf (NW Aegean Sea) varied from 5 mg l1 (discharge 270 m3 s1) during May 1997 to o1 mg l1 (discharge 33 m3 s1) during July 1997 (Karageorgis and Anagnostou, 2003). Recently, Lykousis et al. (2005) have estimated that the mean annual terrestrial flux of the rivers flowing into the NW Aegean Sea is 5.5  106 t yr1, whilst an overall estimate for the principal Greek rivers, based on measured and estimated sediment loads, amounts to 80–95  106 t yr1 (Poulos and Chronis, 1997). The latter figure is probably greater than present discharges because it is based on terrestrial fluxes measured before the extensive construction of dams in Greece and neighboring countries. In a more recent estimate, Poulos and Collins (2002) suggest that PM fluxes in the Mediterranean have suffered a 54% decrease due to entrapment by reservoirs. Another primary source of particles in the Eastern Mediterranean is the autochthonous primary and secondary production. Phytoplankton, zooplankton, and their detritus are abundant in the euphotic zone (80–100 m in the North Aegean and 110–150 m in the South Aegean (Cretan Sea); Lykousis et al., 2002). The Eastern Mediterranean is one of the well-known basins of low productivity of the world ocean due to limited nutrient supply to its surface waters from upwelling, mixing, and external sources including atmospheric input, riverine and waste discharges (Dugdale and Wilkerson, 1988). The mean annual primary production has been estimated to range regionally and seasonally between 15.0 and 60 gC m2 yr1, being higher during late winter–early spring in upwelling cyclonic regions as well as in areas influenced by land-based sources (Dugdale and Wilkerson, 1988; Psarra et al., 2000; Lykousis et al., 2002; Moutin and Raimbault, 2002). These conditions result in high water transparency and general scarcity of particles, particularly in the sub-surface waters of the Eastern Mediterranean. Although limited data exist on POC distribution for the oligotrophic Eastern Mediterranean, the vertical profiles exhibit coherent peaks within the deep chlorophyll maximum zone near the base of the euphotic layer, implying that algal biomass constitutes a significant fraction of the POC pool in this layer (Abdel-Moati, 1990; Socal et al., 1999; Ediger et al., 2005). However, the high POC/Chl-a ratios observed in the surface mixed

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

layer of the Eastern Mediterranean by different investigators (Rabitti et al., 1994; Socal et al., 1999; Ediger et al., 2005) suggest that the POC pool is possibly dominated by bacteria, protozooplankton, detritus originating from regenerative production and heterotrophic activities, and organic matter of atmospheric origin. 2.3. Hydrological characteristics of the Eastern Mediterranean In the Eastern Mediterranean and the Aegean Sea several water masses can be identified. In the upper thermocline Modified North Atlantic (MAW) and Black Sea (BSW) waters intrude into the area under study from the Sicily Channel and the Straits of Dardanelles, respectively (Fig. 2). Both are characterized by their salinity minimum (Lacombe et al., 1958; Ovchinnikov, 1966; Hopkins, 1978). According to Malanotte-Rizzoli et al. (1997, 1999), using the ‘Physical Oceanography of the Eastern Mediterranean’ (POEM) March–April 1986, October–November 1986 and August–September 1987 data sets, the Atlantic-Ionian Stream (AIS) entering the Sicily channel bifurcates into two main branches advecting the MAW into the Ionian interior (Fig. 2). The first branch turns directly southward following the Ionian Anticyclone (IA) in the southwest Ionian, while the second extends further to the north–northeast where it turns southward, advecting MAW on its left side and Ionian Surface Water (ISW) on its right, finally crossing the Cretan Passage transporting MAW in the Levantine Basin (see also Fig. 1). Along the western coasts of Greece (Fig. 2), a series of two cyclones (one off the western coast of Crete, the Cretan Cyclone (CC) and another one off the southern part of Corfu Island) and the Pelops Anticyclone (PA) off southwest Peloponnisos, modulate the circulation of the Adriatic Surface Water (ASW) and the Levantine Surface Water (LSW). Melding observations from the same period as above with model dynamics, Robinson and Golnaraghi (1994), produced a schematic of the upper thermocline general circulation in the Levantine Basin. The AIS jet crossing the Cretan Passage under the name of Mid Mediterranean Jet (MMJ) meanders on the periphery of a number of permanent, recurrent or transient cyclonic and anticyclonic gyres (Fig. 2). In the northern part of the Levantine along the southern Turkish coast, the

181

Asia Minor Current (AMC) meanders in the periphery of the Rhodes Cyclonic Gyre and the Western Cyprus Cyclone, transporting warm, saline LSW westward (Ozsoy et al., 1989; Theocharis and Kontoyiannis, 1999). Branches of AMC are deflected as they intrude the Aegean from the eastern straits of the Cretan Arc (Rhodes, Karpathos and Kassos straits; see also Fig. 1). The LSW enters the Aegean from the eastern and western straits of the Cretan Arc (Fig. 2, Georgopoulos et al., 1989; The POEM Group, 1992; Theocharis et al., 1993). The AMC entering the Aegean from the eastern straits bifurcates, and one branch moves westward in the Cretan Sea, while the other circulates northward, meeting the BSW over the Limnos Plateau. The BSW (low-salinity and cold water, Fig. 2) intrudes the Aegean Sea from the Strait of Dardanelles, after passing through the Bosphorus Strait and the Sea of Marmara (U¨nlu¨ata et al., 1990). The BSW circulates cyclonically in the North Aegean, affecting first the northern and then the eastern coasts of the Aegean. Branches of lowsalinity BSW enter the Thermaikos Gulf. Arriving at the north barrier of Cyclades Plateau, BSW bifurcates and one branch moves southward towards the Cretan Sea, while the other, following the prevailing cyclonic circulation, moves eastward meandering around the Chios multi-lobe permanent cyclone (Zodiatis, 1994; Georgopoulos, 2002; Zervakis and Georgopoulos, 2002). Under the surface layer, a distinct warm and saline water mass, namely Levantine Intermediate Water (LIW) dominates the entire Mediterranean Sea. The Rhodes Cyclonic Gyre and the Western Cyprus Cyclone, permanent dynamic features of the Northern Levantine basin, are the most important sources of both the LIW and, under extreme meteorological conditions, of the Eastern Mediterranean Deep Water (EMDW) in the area. Waters with similar hydrological characteristics are also formed in the Cretan Sea during extremely cold winters (Theocharis et al., 1998, 1999; Georgopoulos et al., 1989). Details of the circulation of LIW are described in Malanotte-Rizzoli et al. (1997). The Adriatic is historically considered as a source of the Eastern Mediterranean Deep and bottom Waters (EMDW). Distinct lenses from the Aegean Sea are identified at depths between 700 and 1100 m. Data obtained between 1987 (Schlitzer et al., 1991) and 1995 (Roether et al., 1996) cruises revealed that the thermohaline circulation in the Eastern Mediterranean has changed. Waters of high density

Fig. 2. Schematic of the circulation and water mass pathways in the Central and Eastern Mediterranean Sea (synthesis from Robinson and Golnaraghi, 1994; Malanotte-Rizzoli et al., 1997; Theocharis et al., 1999; Georgopoulos, 2002). Bathymetry from IOC, IHO, BODC (2003). Water masses: AIS-AIS, MMJ-Mid-Mediterranean Jet, AMC-Asia Minor Current, MAW-Modified Atlantic Water, ASW-Adriatic Sea Water, LSW-Levantine Surface Water, ISW-Ionian Sea Water, BSW-Black Sea Water, CC-Cretan Cyclone, IA-Ionian Anticyclone, PA-Pelops Anticyclone.

182 A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

ARTICLE IN PRESS

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

(warmer and saltier) formed in the Aegean, outflow from the straits of the Cretan Arc replacing the EMDW of Adriatic origin. This abrupt change, called the Eastern Mediterranean Transient (EMT), was attributed to high Aegean Sea salinity, resulting from changes in either the circulation or the largescale water budget (Roether et al., 1996). In addition, Zervakis et al. (2000), suggest that reduced Black Sea outflow into the North Aegean could facilitate dense water formation during the passage of cold atmospheric fronts in the winter. 3. Methodology 3.1. Transmissometry 3.1.1. Data set description The initial data set comprises LT measurements conducted in parallel with routine CTD casts at 3193 stations, occupied between 1991 and 2001 in the Eastern Mediterranean. The data were obtained from 12 research projects, which conducted a total of 40 oceanographic cruises (Table 1), exclusively on board the R/V AEGAEO. LT measurements have been obtained using two models of transmissometers, both emitting in the red part of the spectrum: (a) a 10-cm path length by SeaTech (670 nm); and (b) a 25-cm path length by Chelsea (660 nm). SeaTech instruments have been used widely for the exploration of turbidity variations in nepheloid layers (e.g. Spinrad et al., 1983; Gardner et al., 1985; McCave et al., 2001) and moreover to relate light attenuation and PMC (Baker and Lavelle, 1984; Gardner et al., 1985; Bishop, 1986; Moody et al., 1987; Gardner, 1989a). One or occasionally both transmissometers were used, and were interfaced with a Seabird Electronics SBE-19+CTD deck unit, sampling at 24 Hz. Usually CTD/LT casts were conducted down to 10 m above bottom (mab), but in several cruises they reached 1–2 mab, when the characteristics of near-bottom waters were being addressed in the research project. During field work, routine cleaning of the transmissometer lenses occurred before each cast and an air calibration was performed at the beginning of each cruise. Periodically, the instruments were sent to the manufacturer for calibration. All CTD/LT raw data were processed by the physical oceanographer in charge of each research project. The measurements underwent consistent filtering and processing, as described in the SBE

183

software manual (available at www.seabird.com); data were 1-dbar bin-averaged after quality control of raw data. Data from the projects CINCS, MATER, and PELAGOS were extracted from the Hellenic National Oceanographic Data Center (HNODC), where they were processed and stored following the MEDATLAS protocol (Maillard et al., 2002). Data formats were extremely variable; therefore, the primary goal at first was to convert metadata (sampling date and time, station location, maximum water depth, type of instrument used, etc.) and LT data, to a common format. A number of projectspecific Visual Basicr codes were developed to convert all data to a single format; i.e. to store data in one ASCII file per sampling station, holding metadata in a header line, followed by the water column profile data. The beam attenuation coefficient (c, m1) was computed at the same time and stored in the file, according to the equation: 1 lnðLT=100Þ, L where L is the transmissometer’s path length (m) and LT is light transmission in percent units. Subsequently, all stations obtained during a single cruise were transferred into a ‘cruise file’ in a format compatible for input to Ocean Data View (ODV; Schlitzer, 2003). Metadata, LT, c, and other measured parameters, i.e., temperature, salinity, dissolved oxygen, fluorescence, etc. were also stored in ODV. c¼

3.1.2. Estimation of beam cp Beam c is the sum of three types of attenuations: (1) attenuation due to particles (cp, m1); (2) attenuation due to particle-free water (cw, m1); and (3) attenuation due to colored dissolved organic matter (cCDOM, m1). However, at the wavelength of the transmissometers employed, the last component can be considered as negligible (Jerlov, 1968; Bricaud et al., 1981); therefore, c ¼ cp+cw. Attenuation due to particle-free water (cw) has different reported values, so the SeaTech company adjusts the electronics of all instruments to yield a cw of 0.364 m1 to achieve standardization. In the field, values of cw may be slightly higher or lower than 0.364 m1 in the cleanest part of the ocean because of a few particles in the water, instability or drift of the light-emitting diode, pressure effects on the instrument, or insufficient cleaning of the transmissometer windows (Gardner et al., 2006).

ARTICLE IN PRESS 184

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

Table 1 Inventory of archived light transmission, PMC and POC data Project

Cruise date

Location

ADIOS ADIOS CINCS CINCS CINCS CINCS INTERREG-ION INTERREG-ION INTERREG-NA INTERREG-NA INTERREG-NA INTERREG-NA INTERREG-NA KEYCOP KEYCOP LIW-EXP MATER MATER MATER MATER MATER MATERTRANSMED METROMED METROMED METROMED METROMED PAGAS PAGAS PAGAS PELAGOS PELAGOS PELAGOS PELAGOS POEM SARONIKOS SARONIKOS SARONIKOS SARONIKOS SARONIKOS SARONIKOS

April, 2001 October 2001 May 1994 November 1994 February 1995 May 1995 March 2000 September 2000 May 1997 February 1998 June 1998 September 1998 March 2000 September 1999 March 2000 April 1995 March 1997 September 1997 February 1998 March 1998 October 1998 July 1999

Ionian Sea, Cretan Straits Ionian Sea Cretan Sea Cretan Sea Cretan Sea Cretan Sea Ionian Sea, Korinthiakos Gulf, Patraikos Gulf Ionian Sea, Korinthiakos Gulf, Patraikos Gulf North Aegean Sea North Aegean Sea North Aegean Sea North Aegean Sea North Aegean Sea NE Aegean Sea NE Aegean Sea Levantine North Aegean Sea, Cretan Sea North Aegean Sea, Cretan Sea North Aegean Sea, Cretan Sea, Cyclades, Myrtoon Sea North Aegean Sea, Cretan Sea Cretan Sea, Ionian Sea, Levantine Levantine, Ionian Sea, Central and Western Mediterranean NW Aegean Sea NW Aegean Sea NW Aegean Sea NW Aegean Sea Pagassitikos Gulf Pagassitikos Gulf Pagassitikos Gulf Cretan Sea and straits Cretan Sea and straits Cretan Sea and straits Cretan Sea and straits Ionian Sea, Levantine Saronikos Gulf Saronikos Gulf Saronikos Gulf Saronikos Gulf Saronikos Gulf Saronikos Gulf

May 1997 July 1997 February 1998 September 1998 April 1999 June 1999 September 1999 March 1994 June 1994 September 1994 December 1994 October 1991 February 1998 May 1998 June 1998 August 1998 December 1998 February 1999

Total

In the measurement of particles in water, there are no ‘standard samples’ that can be prepared and used for calibration as with salinity ‘standards’. Thus, a standard practice with LT measurements in the open ocean has been to set the minimum recorded c of the profiles of a cruise to cw (Bartz et al., 1978; Gardner et al., 2006). To estimate the value of attenuation due to clear water for each one of the cruises, we first classified

No. LT stations 50 54 35 37 34 35 74 82 126 70 117 134 143 78 61 78 82 162 164 70 74 37

No. PMC

No. POC

116 175

112

102

87

88 49 64

41 30 37

78 158 66

87 70

145 143 144 147 14 12 16 147 74 124 115 151 15 11 25 38 47 28

177 191 186 192 20 21 20

39 30 30 67

15 35 38 27 52

23 61 87 36 75

3193

1870

912

stations as ‘deep’ (water depth 4200 m) and ‘shallow’ (water depth o200 m). This classification derived from a generalized observation that turbid waters in the Eastern Mediterranean usually occupy the upper part of the water column, whilst waters deeper than 200 m are relatively particle-free down to the bottom (Karageorgis and Anagnostou, 2003; Karageorgis et al., 2003). For the deep stations, cw was set as the minimum recorded c of the profile.

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

For the shallow stations, cw was computed as the average of the deep-station c minima, c deriving from the same cruise. In cases where a cruise occupied only shallow stations, cw was set as the minimum c of the entire cruise. For shallow stations it is possible that cw values contained a small attenuation due to particles, so the reported cp values are slightly smaller than the true values. Subsequently, cp values were computed for each cruise and the entire data set was stored in a new ODV compilation. Data obtained from multiple casts at the same station, usually restricted to water sampling in intermediate or near-surface waters, were removed from this compilation, and after data reduction the total number of casts was 2463. 3.1.3. Handling missing data At most of the stations, either the 10-cm or both the 10- and 25-cm transmissometers were used. In order to produce more reliable maps and crosssections, we chose to use data from the 10-cm transmissometer, which were more abundant. However, on a number of cruises only the 25-cm instrument data were available. In these cases, beam cp values for the 10-cm transmissometer (SeaTech) were obtained by linear regression (Fig. 3; cp10 ¼ 0.992 cp250.0039, R2 ¼ 0.918, n ¼ 525,335).

185

3.2. PM concentration PMC was determined by onboard water filtration. Seawater from various depths (high, low optical turbidity and depths in between) was collected by 12 GoFlo or Niskin bottles attached to a General Oceanics rosette. Six to 10 l of water were filtered through pre-weighed polycarbonate isopore membrane filters (Millipore HTTP 04700: diameter 47 mm, pore size 0.4 mm) under low vacuum. Filters were carefully rinsed several times with MilliQ water to remove the salt and then airdried. At the laboratory, filters were stored in a desiccator for 24 h and then re-weighed to determine PMC. A total of 1870 measurements were collected from 7 research projects in the period 1997–2001 (Table 1). 3.3. Particulate organic carbon For the determination of POC concentration, seawater samples (2–8 l) collected at standard depths (usually 3, 20, 50, 100, 200, 500, 1000, 1500, 2000, 3000 m, and near-bottom) were filtered onboard through GF/F filters pre-combusted at 450 1C (Whatman: diameter 25 mm, nominal pore size 0.7 mm). The filters were stored at 20 1C in the dark. In the laboratory, the filters were dried at 60 1C. In order to remove the inorganic carbon, the filters were acidified with HCl (Verardo et al. 1990; Cutter and Radford-Knoery, 1991) and POC was subsequently determined with an EA 1108 CHN Fisons Instruments analyzer (instrument calibration by acetanilide or atropine). The POC values were corrected on the basis of blank filter measurements. Filter blanks were pre-combusted filters taken to the ship, but no filtered seawater was passed through them; adsorption of DOC and colloidal organic carbon onto filters (Moran et al., 1999; Gardner et al., 2003) could have introduced a positive bias in the measured values, particularly at low POC concentrations (o4 mmol l1). Over the period 1997–2001, 912 particulate samples from 6 research projects were collected and analyzed (Table 1). 3.4. Frequency distribution

Fig. 3. Linear regression of beam cp10 (SeaTech transmissometer) vs. beam cp25 (Chelsea transmissometer). cp10 ¼ 0.992 cp250.0039, R2 ¼ 0.918, n ¼ 525,335.

Monthly frequency distributions of collected data reveal that most stations (23%) were occupied during September (Fig. 4), which is a favorable month for measurements at sea in this area. There are no data available for January. August and

ARTICLE IN PRESS 186

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

Fig. 4. Left: monthly distribution of LT stations occupied, PMC and POC measurements. Right: sampling-depth frequency distribution of PMC and POC measurements.

November are also under-sampled (1%). Similar trends can be observed for PMC and POC measurements, which typically accompany CTD/ LT casts and water sampling. Samples for PMC and POC measurements were collected predominantly near the surface, followed by water depths in the range 50–100 m. Overall, 80% of the PMC samples and 83% of the POC samples were collected from depths p200 m; towards deeper waters the number of samples decreases considerably. This distribution pattern could be explained by research project requirements, which most often target studies of biogeochemical processes taking place in the upper part of the water column, i.e. in the euphotic zone. 4. Results and discussion 4.1. Distribution of PM in the Eastern Mediterranean Spatial distribution patterns of PM are discussed hereafter in terms of beam cp variations in selected depth layers, namely at 5 m (surface nepheloid layer, SNL), 20, 50, 100, 500 m, and 10 m above bottom (benthic nepheloid layer, BNL). In Figs. 5 and 6, the color scale of cp has been held constant in the range 0–1 m1 to allow a direct comparison between plots. However, higher cp values have been encountered (up to 2 m1), particularly in nearby river mouths. Temporal distribution is based on beam cp variations during ‘wet’ and ‘dry’ periods. 4.1.1. Winter/spring conditions (December– May) Spatial distribution of beam cp in the SNL during the ‘wet’ period clearly shows that most elevated values appear in the North Aegean Sea and

particularly in its NW sector, in the Thermaikos Gulf (Fig. 5; cp41 m1). Particle-rich waters are predominantly of riverine origin (Axios, Aliakmon, Pineios, Loudias, and Gallikos Rivers; Fig. 1) and the Axios River is the principal contributor of PM in Thermaikos Gulf (50%; Karageorgis and Anagnostou, 2003), as highest cp values appear just in front of the river mouth. Riverine particles are dispersed to the south/southwest following a narrow zone parallel to the coastline. This feature is attributed to the cyclonic circulation that prevails in Thermaikos Gulf (Durrieu de Madron et al., 1992; Karageorgis et al., 2000; Karageorgis and Anagnostou, 2001, 2003). However, the entire continental shelf of Thermaikos Gulf is dominated by relatively turbid waters; whereas towards the SE cp values decrease to open sea values, similar to those observed in the open North Aegean Sea. Likewise, the entire North Aegean continental shelf is characterized by relatively high cp especially near the Strymon, Nestos and Evros river mouths (Figs. 1 and 5; cp40.4 m1). The plume of Evros River appears to influence a greater area of the NE Aegean Sea. While turbid riverine waters tend to shift to the west, due to Ekman transport, the anticyclone of Samothraki shifts the plume to the east and then to the south, following the coastline, (Georgopoulos, 2002). This circulation pattern is also responsible for turbid SNL east of Limnos Island up to the Dardanelles Strait. Karageorgis et al. (2003) suggest that relatively high cp values are due to river runoff or local biogenic production, rather than input of particle-rich BSW. This is in agreement with Lykousis et al. (2002), who conclude that BSW is a contributor of dissolved organic carbon in the Aegean, but not of suspended particles.

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

187

Fig. 5. Spatial distribution of particulate matter (beam cp, m1) in the Eastern Mediterranean during winter/spring conditions (‘wet’ period, December to May) at various depths. SNL: surface nepheloid layer, BNL: benthic nepheloid layer.

The Central and South Aegean Sea are generally characterized by low (o0.1 m1) cp values, except for a sector ENE of Crete, where values are higher (Fig. 5; 0.2–0.3 m1). In the Levantine Sea, south of Rhodes Island, an impressive circular feature appears with beam cp ¼ 0.5 m1 at the center and fading to 0.3 m1 toward the edges, values that are

surprisingly high for open sea Mediterranean conditions. This pattern is related to the permanent cyclonic Rhodes Gyre, which will be discussed in more detail in the following section. The North (Otranto Strait) and South Ionian Sea exhibit low beam cp values (0.1–0.2 m1), particularly off the Ionian Islands (Fig. 5). However,

ARTICLE IN PRESS 188

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

Fig. 6. Spatial distribution of particulate matter (beam cp, m1) in the Eastern Mediterranean during summer/autumn conditions (‘dry’ period, June to November) at various depths. SNL: surface nepheloid layer, BNL: benthic nepheloid layer.

between the islands Lefkada, Kefallonia, Zakynthos and the mainland, local cp maxima are observed, most probably due to freshwater discharges from Acheloos and Alfeios rivers (Fig. 1). Likewise, relatively elevated cp values are observed in the gulfs of Patraikos and Korinthiakos, which receive terrigenous material from a number of small rivers discharging into the area. Finally, in the Saronikos

Gulf, which receives the effluents of Athens (population 5,000,000), the SNL is relatively turbid (cp40.3 m1), whilst near the Piraeus port area values exceed 1 m1. At 20 and 50 m depths, the aforementioned features can still be observed, but their intensity decreases gradually towards lower beam cp values all over the Eastern Mediterranean (Fig. 5). For

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202 Table 2 Mean beam cp (m1) during ‘wet’ and ‘dry’ periods in the Eastern Mediterranean Depth layer (m)

‘Wet’ period: mean beam cp (m1)71s

‘Dry’ period: mean beam cp (m1)71s

SNL-5 m 20 50 100 200 500 1000 BNL (1–10 mab)

0.39170.533 0.21870.207 0.12270.117 0.06670.073 0.03970.055 0.02470.031 0.00770.005 0.33570.550

0.18370.277 0.13370.157 0.11870.148 0.06970.108 0.03670.081 0.02870.041 0.01370.012 0.25470.418

example, the North Aegean shelf is characterized by relatively turbid waters at 20 m depth, but at 50 m most of the prominent features seen at 20 m are absent. The average cp value in the open sea is 0.1 m1, except for the Levantine, where, because of elevated biological productivity, cp values are 0.2–0.3 m1. At 100, 200, and 500 m depths, the PMC decreases and tends to be homogeneous in the area, with average cp values of 0.07, 0.04, and 0.02 m1, respectively (Table 2). The BNL exhibits high optical variability (Fig. 5). All over the North Aegean continental shelf and up to 200 m depth, the layers near the bottom were turbid, with a thickness ranging from 1–2 to 60 m above bottom, and showing locally even higher cp values than the SNL (on average the SNL exhibits cp ¼ 0.33 and the BNL cp ¼ 0.29). Once more, most turbid areas were found in the vicinity of river mouths. BNLs are generally attributed to resuspension of fine-grained surface sediment by waveinduced bottom currents (e.g. Gardner, 1989a, b; Poulos, 2001). This holds true for the shallow sector of the continental shelf (depth 30–40 m), whilst in deeper waters BNLs originate from lateral advection of particles resuspended nearshore (e.g. Biscaye and Eittreim, 1977; Richardson, 1987; Durrieu de Madron et al., 1992; Karageorgis and Anagnostou, 2003). Intense resuspension near river mouths is also related to the composition of prodeltaic surface sediments, being recently deposited, unconsolidated and probably deficient in organic matter that could bind the sediment and inhibit resuspension. Enhanced BNLs are also observed in coastal areas and enclosed gulfs, whereas in the open sea there is no evidence of sediment resuspension. However, the lack of BNLs in the deep sectors of the Eastern Mediterranean may be due to operational conditions. For example, in cases of stormy weather, the

189

casts were not conducted very near the bottom, for fear of a crash of the CTD/rosette unit. In addition, sometimes the available cable length was less than the actual station depth. 4.1.2. Summer/autumn conditions (June– November) The SNL during the ‘dry’ period exhibits lower cp values in the North Aegean Sea and particularly in Thermaikos Gulf and other coastal areas influenced by river discharges (Fig. 6). Although the distribution patterns are comparable to those of the ‘wet’ period, absolute cp values are lower (o1 m1). Moreover, turbid zones do not extend as far offshore as during the ‘wet’ period. Except in some coastal areas (e.g., Saronikos and Pagassitikos Gulfs), the SNL is fairly homogeneous throughout the study area, exhibiting average cp values of 0.18 m1. The situation is similar at 20 and 50 m depth, average cp values decreasing gradually to 0.13 and 0.12 m1, respectively (Fig. 6; Table 2). The major differences in the spatial distribution of cp between the ‘wet’ and the ‘dry’ period are mostly in the upper water layers, to 50 m depth. In deeper layers, waters become extremely transparent with average cp values ranging between 0.07 m1 at 100 m and o0.02 m1 at 1000 m depth (Fig. 6; Table 2). While mean BNL cp values are higher than SNL cp values during the ‘dry’ period (0.25470.418 and 0.18370.277, respectively), this relationship is reversed during the ‘wet’ period; i.e. SNL cp values are higher than BNL cp values (0.39170.533 and 0.33570.550, respectively), though perhaps the differences are not statistically significant. ‘Wet’ period cp values are greater than ‘dry’ period values in both layers. This pattern suggests clearly that the formation of BNLs is independent of season, although higher terrigenous inputs during the rainfall periods may increase the intensity of the BNL in some coastal locations. 4.2. Relationships between cp and mesoscale dynamic features Apart from beam cp spatial variations during ‘wet’ and ‘dry’ periods, a number of interesting features were identified during data analysis dealing largely with elevated cp values either in the open sea or coastal areas and were attributed to the influence of cyclonic/anticyclonic gyres on the distribution of cp. Some characteristic cases are discussed hereafter.

ARTICLE IN PRESS 190

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

4.2.1. The Rhodes Gyre The Rhodes Gyre is described by Anati (1984) as follows: ‘‘The horizontal distribution of heat content of the upper layers in the Levantine Basin reveals a curious phenomenon: in a narrow region southeast of Crete and Rhodes, this heat content is considerably lower than the adjacent areas. This phenomenon seems to be a permanent feature of this region, not restricted to any particular season’’. He estimates the area occupied by this feature to be less than 50,000 km2, with a diameter of 100 km. The Rhodes Gyre is a permanent multi-centered cyclonic gyre situated east of Rhodes Island and is considered to be the most prominent dynamic feature in the Eastern Mediterranean. It is surrounded by the AMC in its northern and western parts and by branches of the MMJ in its southern part (The POEM Group, 1992). This region is considered to be the main source area for the LIW formation under the cold and dry north winds blowing from the continent (Ovchinnikov and Plakhin, 1984), and recently it has been reported as a zone of deep-water formation in the Eastern Mediterranean Sea during exceptionally cold winters (Ozsoy et al., 1993; Gertman et al., 1994). We present data gathered during the LIWexperiment, which was specifically designed to monitor the dome evolution of the convective chimney of the Rhodes Gyre during March–April 1995 and covered the spreading phase of the phenomenon (The LIWEX Group, 2003). By this time, the Rhodes Gyre showed considerable recapping, as stratification was gradually restored by the surface warming. The spatial distribution of beam cp was characterized by relatively elevated values (up to 0.5 m1) at the main dome of Rhodes Gyre (Fig. 7a). The recapping was accompanied by an oxygen-rich upper layer, which exhibited higher dissolved oxygen concentrations in the same sector (5.5 ml l1; Fig. 7a, c). The impressive dome of Rhodes Gyre was reflected clearly in the vertical distribution of salinity (Fig. 7b), which appeared to influence the water column down to 750 m. This feature was accompanied by a 50–60 m layer rich in PM (Fig. 7d). The upwelling of nutrient-rich deep waters (Siokou-Frangou et al., 1999; Souvermezo-

191

glou and Krasakopoulou, 1999) favors an increase in primary productivity and thus an accumulation of living and non-living particulate organic matter in the euphotic zone of the Rhodes cyclone and its peripheries. Evidently, elevated cp values in this sector of the Levantine Sea are related to higher primary productivity and abundance of biogenic particles in the upper water column. According to Ediger et al. (2005) the upper layer of the Rhodes Gyre and its peripheries reveal great temporal variations in both abundance and composition of bulk POM due to large fluctuations in nutrient input from intermediate depths, being higher during late winter–early spring when it can reach values comparable to those of the more productive NW Mediterranean basin. Similar results (not shown here) were also obtained during the PELAGOS experiment in March 1994 (Theocharis et al., 1999). On the other hand, during the POEM-BC experiment in November 1991 (which appears to be an early stage of dome development) two cyclonic domes were identified. The domes were characterized by higher PMCs compared to the corresponding surrounding anticyclones. 4.2.2. The PELAGOS cyclonic gyres The extreme hydrological complexity of the Cretan Sea and Straits has been demonstrated by Balopoulos et al. (1999) during the Mediterranean Targeted PELAGOS Project, as well as by previous investigations in the area (e.g. Georgopoulos et al., 1989; The POEM Group, 1992; Theocharis et al., 1993). In general, a succession of cyclones, anticyclones, and smaller scale eddies dominate throughout the year. In addition, biomass concentrations, primary production rates, the biogeochemistry of major elements in suspended PM, and fluxes of settling particles, confirm that the Cretan Sea is one of the most oligotrophic areas in the world (Balopoulos et al., 1999; Price et al., 1999). In addition, based on productivity and light attenuation data, Ignatiades (1998) found that the waters of the Cretan Sea belong to Jerlov’s Optical Water Type I (the most transparent), having a deep blue color and an average value of spectral attenuation coefficient at 480 nm of 0.040 m1.

Fig. 7. (a) Beam cp (m1) spatial distribution at 5-m depth in the Levantine Sea during April 1995. Superimposed are dissolved oxygen concentration contours (white color, in ml l1) and sampling station locations (blue dots); black line indicates the west–east transect of (b) salinity, (c) oxygen (ml l1), and (d) beam cp (m1) along 351N in the NW Levantine Basin; (e) Winter composite of MODIS (4 km resolution) chlorophyll-a (mg m3) illustrating the productivity of the Rhodes Gyre for the period 2002–2007. In the study area, chlorophyll-a usually reaches maximum abundance at the end of February and beginning of March, thus the winter composite is representative of the normal size and intensity of the bloom.

ARTICLE IN PRESS 192

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

Fig. 8. Beam cp (m1) spatial distribution at 5-m depth in the Cretan Sea during March 1994. Superimposed are sigma–theta contours (white color, in kg m3) and sampling station locations (dark-blue dots).

During March 1994, relatively elevated beam cp values (0.3 m1) were observed in the eastern Cretan margin, extending northerly into the Cretan Sea (Fig. 8). Price et al. (1999) have reported elevated values of particulate Al and Ca in the same area; they suggest that these elements originate primarily from atmospheric fallout, and secondly by direct input from Crete Island or by transport of resuspended sediment from the shelf. However, in surface waters the Ca/Al ratio was found to be about five times higher than in deeper waters, and thus it was assumed to be newly formed CaCO3 from primary production; similarly, biogenic Si was found to be elevated in surface waters. When sigma–theta contours are superimposed on the cp spatial distribution plot, it becomes evident that high cp values correspond to fields of higher sigma–theta, further implying the presence of cyclonic gyres in the sector NE of Crete Island (Fig. 8). As in the Rhodes cyclonic gyre, upwelling nutrient-rich waters seem to create favorable conditions for primary production, and subsequently higher cp values. Given that the Cretan Sea is characterized as a highly oligotrophic environment (e.g. Dugdale and Wilkerson, 1988; Danovaro et al., 1999; Psarra et al., 2000; Tselepides et al., 2000; Lykousis et al., 2002), the formation of cyclonic gyres and eddies plays a paramount role in the biological productivity of the region.

During December 1994–January 1995, representing the EMT period, the newly formed Cretan Deep Waters (CDW) are characterized by high-salinity values (439) and high-density (sy429.2 kg m3) (Fig. 9a–c). The corresponding beam cp values are very low (o0.01 m1, from 750 m depth and deeper). In the western basin of the Cretan Sea, an intense BNL is observed, with a thickness of 60 m, and beam cp value 0.040 m1 (Fig. 9d). In the western and eastern straits of the Cretan Arc, highdensity waters outflow (Fig. 9c). The outflowing CDW water is characterized by very low cp values. During the EMT, the old deep waters occupying the deep basins of the Cretan Sea, part of the Ionian Sea and the western Levantine Basin, depleted in oxygen and rich in nutrients and PM were replaced by the newly formed denser and with very low beam cp waters. The old waters were subsequently uplifted into shallower layers. 4.2.3. The Saronikos Gulf case The Saronikos Gulf is situated in the west-central region of the Aegean Sea and is separated into two sectors by a shallow zone; the western basins (Megara and Epidavros Basins) have depths exceeding 400 m, whilst in the eastern basin depths are around 100 and 200 m (Fig. 10). Furthermore, the semi-enclosed Elefsis Bay, situated to the north, is separated from the Gulf by two shallow sills. The

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

193

Fig. 9. (a) Location map of a west–east transect in the PELAGOS Project study site, during December 1994–January 1995 (station locations in white dots). Vertical (0–2500 m) distributions of (b) salinity, (c) sigma–theta (kg m3), and (d) beam cp (m1). Station profiles are denoted by blue lines.

predominant circulation is cyclonic, with open Aegean Sea waters entering the Gulf from the south, following the eastern coastline and then moving towards the western deep basins (Hopkins, 1974). However, the circulation in the different areas of the Gulf is mainly wind-driven. Several circulation patterns are described by Hopkins and Coachman (1975) and Kontoyiannis and Papadopoulos (1999). For example, during a study period between May 1998 and February 1999, cyclonic dynamic features are observed under the influence of northerly winds during both winter and summer, whilst an anticyclone appears in the western basins sector (Kontoyiannis and Papadopoulos, 1999). Apart from atmospheric fallout and minor streams of temporary flow, PM introduced into the Saronikos Gulf derives from the wastewater treatment plant (WWTP) constructed on the small

Psyttaleia Island. Prior to the WWTP operation starting in 1994, effluents were released untreated into the Gulf, while presently it daily treats 720,000 m3 of 1st stage domestic and industrial effluents (http://www.eydap.gr). The effluents are discharged into the Gulf by a V-shaped pipeline situated close to the bottom at 63 m depth, south of Psyttaleia Island. Krasakopoulou and Karageorgis (2005) have reported POC concentrations in the vicinity of the outfall often higher than 400 mg l1. South of the outfall, POC values decline substantially towards the open sea. Field measurements conducted during August 1998 in the Saronikos Gulf (Fig. 10a) have revealed that the effluents discharged from the WWTP pipeline move to the south–southwest, then pass through the Salamina-Aegina strait and then follow an anticlockwise transport into the Megara Basin

ARTICLE IN PRESS 194

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

Fig. 10. (a) Location map of a west–east transect in the Saronikos Gulf, during August 1998 (station locations in white dots). Vertical (0–450 m) distributions of (b) beam cp (m1), and (c) dissolved oxygen (ml l1). Station profiles are indicated in blue lines.

(Fig. 10b). Beam cp proved to be a valuable tracer of the effluents, which are clearly more turbid than ambient sea water (Fig. 10a, b). The effluents are dispersed almost horizontally at depths between 48 and 53 m, where beam cp values vary from 0.854 m1 at the discharge point, to 0.113 m1, where the signal is lost. The same pattern was generally observed during the June and December 1998 and February 1999 cruises, with a difference in the effluent dispersal

depth (45–80 m). This distribution field is attributed to the prevailing cyclonic circulation in the Saronikos Gulf and may have significant implications for the dissolved oxygen concentrations in the Megara and Epidavros basins. According to Kontoyiannis and Papadopoulos (1999) the renewal time of the aforementioned basins is 8 years: thus re-oxygenation rates of the deep layers are very low. A continuous input of organic-rich material originating from the

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

195

WWTP would further reduce the dissolved oxygen concentrations, which during August 1998 were dysoxic (1.43 ml l1) in the deepest sector of the Epidavros Basin (Fig. 10c). 4.3. PMC vs. cp Measured PMC varies by three orders of magnitude, from 6 to 6365 mg l1. The lower values were found in intermediate waters (750–1000 m) of the South Ionian Sea, 200 km southwest of Peloponnisos and west of Crete (ADIOS Project). The highest PMCs were observed in the North Aegean Sea, at the SNL of three shallow (10 m depth) stations neighboring Evros River (INTERREG-NA Project). A frequency distribution plot shows that low PMCs (o200 mg l1) constitute 55% of the measurements, whilst PMCs 41000 mg l1 represent only 8% of the data (Fig. 11). Due to variations of PM composition, particle size and refractive index, the correlation of beam cp against PMC sometimes shows almost linear relationships, but slopes and intercepts exhibit both spatial and temporal variation (Baker and Lavelle, 1984). Bishop (1986) has explored effectively the calibration of SeaTech transmissometer vs. PMC but concludes that a calibration equation should be cruise-specific. The regression of PMC vs. beam cp is illustrated in Fig. 12. A Model II fit (used when both variables in the regression equation are independent; Legendre, 2001) yields the equation PMC ¼ 1072cp+ 33 (R2 ¼ 0.732; n ¼ 1764). This good correlation appears to be controlled partly by the data from the MetroMed, (NW Aegean Sea) and INTERREGNA (North Aegean Sea) projects, which exhibit the highest PMCs and the largest concentration ranges.

Fig. 11. Frequency distribution of particulate matter concentration measurements in the Eastern Mediterranean Sea.

Fig. 12. Regression between PMC and beam cp.

Some data sets, e.g. the ADIOS, are characterized solely by low PMCs (o100 mg l1), therefore introducing a lot of scatter at low concentrations. Similar behavior is observed for all data sets consisting of only low PMC concentrations. Therefore a single composite equation has been used for the entire region to obtain an average relationship. Separate equations for each area, especially for areas with small concentration ranges, might lead to discontinuities at arbitrary boundaries. The above equation is used as an empirical function relating optical measurements and PMC. This is useful for estimating PM budgets or in models. Moreover, historical LT measurements in other areas could be used to give a rough estimate of PMC, provided that beam cp are processed according to the procedure described previously. 4.4. POC vs. cp The regression of POC vs. beam cp for each one of the research projects or regions of the Eastern Mediterranean Sea is illustrated in Fig. 13, and Model II linear regression parameters are summarized in Table 3. Near-bottom data pairs were excluded from the correlation because, in the presence of BNLs due to resuspension, beam cp appears to be elevated, but is not related to an increase of POC concentration. Four samples (not shown) from Thessaloniki Bay (Project MetroMed) were considered outliers and were not included in the regression; these samples are related to highly eutrophic conditions and represent very high POC concentrations, which are unlikely to appear in other regions of the Eastern Mediterranean. In addition, six samples from the NE Aegean

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

196

Fig. 13. Regressions between POC and beam cp acquired in different research Projects and regions of the Eastern Mediterranean Sea: (a) ADIOS, open Ionian Sea; (b) INTERREG-ION, open and coastal areas of the Ionian Sea, including Patraikos and Korinthiakos Gulfs; (c) INTERREG-NA, North Aegean Sea; (d) KEYCOP, northeastern Aegean Sea; (e) METROMED, NW Aegean Sea including the bay and gulf of Thessaloniki and Thermaikos Gulf; and (f) SARONIKOS, Saronikos Gulf (Krasakopoulou and Karageorgis, 2005). Model II fit parameters are presented in Table 3.

Table 3 Model II linear regression parameters (POC vs. cP) for different research projects implemented in the Eastern Mediterranean Sea Model II parameters

Slope Intercept SD slope SD intercept n R2

Research project ADIOS

INTERREG-ION

INTERREG-NA

KEYCOP

METROMED

SARONIKOS

All data

30.4 1.33 2.37 0.102 101 0.386

10.8 1.61 0.937 0.302 47 0.649

17.5 1.41 1.10 0.458 68 0.731

16.6 2.09 1.04 0.198 118 0.542

16.7 2.54 0.870 0.418 104 0.716

22.9 2.35 0.912 0.324 202 0.679

18.5 2.06 0.420 0.138 636 0.674

Sea (Project Keycop), and in particular from the BSW-LIW frontal area exhibit substantial deviation from the other available data. The samples are characterized by high-beam cp, but very low POC concentrations, possibly representing input of detrital inorganic particles or an abundance of colored dissolved organic matter. Due to their high varia-

tion and limited geographical extent the samples were excluded from the regression. Likewise, seven surface samples (3-m depth) from nearshore regions (Project INTERREG-ION) were also excluded. Excluded data do not appear in Figs. 13 and 14. Based on Model II statistics, we may infer that at very low cp values (o0.04 m1) a high scatter is

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

Fig. 14. Regression between POC and beam cp.

observed in POC concentrations, as for example for the ADIOS data set, which comprises exclusively open sea stations. Correlation coefficients R2 are markedly higher when the data set includes high cp values and corresponding high POC concentrations (Table 2). A Model II fit of all available data pairs yields the equation POC ¼ 18.5cp+2.06 (POC in mM, R2 ¼ 0.674; n ¼ 636) or POC ¼ 222cp+25 (POC concentration in mg l1). The monthly frequency distribution of POC data (Fig. 4) reveals absence of data for October, November, and January that could potentially affect the representation of winter conditions. Separate regressions were conducted for the winter–spring and the summer–autumn seasons, which resulted in similar linear equations (not shown here). The overall variation between the two fits was in the range of 10%; therefore, one single equation can be used to convert beam cp readings to POC for both the ‘wet’ and the ‘dry’ seasons. The slope of the cp vs. POC regression is about 60% of those observed in open-ocean regions by Gardner et al. (2006); this means that for a given cp value, predicted POC is much lower with the Eastern Mediterranean equation than with the global POC-beam cp regression. Since beam cp is a function of particle size, shape and refractive index, the lower slope of the Eastern Mediterranean might be due to a different composition and size spectra of the planktonic communities produced in this regional ecosystem. In fact, in the oligotrophic environment of the Eastern Mediterranean the dominance of small-size (o2.0 mm) phytoplankton is well documented (Ignatiades et al., 2002; Li et al.,

197

1993; Yacobi et al., 1995) even during the annual phytoplankton bloom (Vidussi et al., 2001). As in the previous case for the regression of PMC vs. beam cp, the general equation can be used for the estimation of POC budgets. Obviously the use of a general combined regression is a compromise: from Table 3 it is clear that in the South Ionian Sea (ADIOS data) this regression will underestimate the POC values, while in the North Ionian and the Korinthiakos Gulf, POC values will be slightly overestimated. For the North and South Aegean Sea this empirical algorithm seems to be a good fit. However, the above POC/cp relationship is considered representative of the relatively low POC concentrations of the Eastern Mediterranean environment and provides a useful tool to predict POC levels from cp profiles, quantify the POC standing stocks in different regions over different seasons, and further improve biogeochemical and ecosystem models by assimilating these data. Although POC is a small constituent of the total carbon budget, it is a dominant component of primary production, which creates particles that can settle through the water column, across isopycnals, scavenging or aggregating other particles and transporting carbon and associated elements to deep waters, where they enter the sediments or, more likely, are remineralized. 4.5. PM and POC standing stocks Using the equations described above, the standing stock of PM and POC was estimated for the upper 150 m of the water column based on beam cp profiles in a total area of 595,000 km2. PM standing stock estimates were 10.7  106 and 8.9  106 t during the ‘wet’ and ‘dry’ periods, respectively. The annual average PM stock estimate from all available profiles was 9.8  106 t. Similarly, the POC standing stock estimates were 3.5  106 and 3.2  106 t for the ‘wet’ and ‘dry’ periods (annual average from all profiles 3.3  106 t). Using the common assumption that twice the POC concentration roughly equals total organic matter (Gordon, 1970; Gundersen et al., 1998), it appears that 65–72% of the suspended PM in the upper 150 m of the Eastern Mediterranean is organic, the remaining 30–35% being carbonate and opal skeletons plus terrigenous-lithogenic fine particles originating in atmospheric fallout and locally in riverine inputs. Additional calculations of standing stocks (both for PM and POC) were made for the

ARTICLE IN PRESS 198

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

periods March–August (high primary production) and September–February (low primary production), but differences were not statistically significant. However, since there is a much larger number of profiles per square km in coastal areas, where concentrations are higher, the standing stocks are probably skewed to high concentrations. Profiles in the open ocean are further apart, but they represent larger areas of low-concentration water. The standing stock of POC (integrated over 0–150 m) in the South Ionian Sea (ADIOS site: latitude 341300 –351400 N, longitude 201090 –201470 E) in an area of 17.9  109 m2, is about 90  103 t POC, which corresponds to a mean concentration of 33.6 mg C l1. By definition, the POC flux (mg C m2 d1) is the product of the POC concentration (mg C l1) times the corresponding sinking speed (m d1) for each assemblage of particles having different settling velocities. Settling particles have sizes ranging from sub-micron colloids to picoplankton (ffi1 mm) to large microplankton (X300 mm) and particle aggregates. Size (or diameter) and density are the primary properties of particles that affect their settling speeds in the water column, being slower for the small picoplankton (o1 m d1) and faster for large particles and aggregates (0.1–100 m d1). If we assume the presence of an almost uniform particle population with an average settling velocity of 5 m d1, the estimated POC flux through 150 m is 168 mg C m2 d1. Comparing this value with the mean annual POC flux (3.97 mg C m2 d1) measured at a sediment trap deployed at 186 m in the same area for the period May 2001–April 2002 (S. Stavrakakis, pers. comm.), then only 2.5% of the carbon standing crop at 0–150 m needs to reach the trap at 186 m to produce the observed flux. Consequently, about 97% of POC would be recycled in the upper 150 m or degraded into dissolved organic and inorganic forms. The estimated carbon export from the upper layer is lower than the values of 5–10% observed in open ocean (Buesseler, 1998) and suggests that about 97% of the primary production is sustained by internal recycling of organic matter. Integrated primary production rates from two stations in the South Ionian Sea during summer 1996 are somewhat higher than the previously estimated POC flux through 150 m (255 and 313 mg C m2 d1), but the export of carbon was estimated to represent 3.3% of the integrated primary production (Moutin and Raimbault, 2002). Conversely, we could roughly estimate a range of mean sinking speeds of POC by

assuming that 5–10% of the carbon is exported from the upper layer. Such a calculation suggests that the mean settling velocity is only about 1.2–2.4 m d1 in the upper water column, which is much slower than the 100 m d1 rate estimated from traps near the seafloor (Deuser et al., 1981). 5. Conclusions The analysis of historical light transmission measurements from the NE Mediterranean Sea provided useful algorithms and estimates of PM and POC. Data reduction methods were standardized, which yielded more accurate and reliable beam cp data from a great number of different research projects and cruises; thus cp data can now be directly interrelated. Plots of beam cp over the entire study area at various depths exhibit distribution patterns typical of other studies (high surface values, low subsurface values with little variation with depth, and occasional increases near the seafloor) and good fit between different data sets. We were able to differentiate distribution patterns between ‘wet’ and ‘dry’ seasons and observed seasonal variability of PM in the NE Mediterranean. Regions with abnormally high and low values of beam cp were identified and directly related to physical factors affecting PM distribution in the water column (e.g. river input or eutrophic waters). The correlation of PMC and POC concentration vs. optical measurements is significant and will contribute to generalized equations for the NE Mediterranean. Although there were differences in the regressions of POC/cp or PM/cp data within this area, there was no significant seasonal difference found in the combined data. The slope of the POC concentration vs. cp regressions is about 60% of those observed in open-ocean regions by Gardner et al. (2006) and could reflect different ecosystems in low-nutrient waters. These equations will be useful in future research projects for estimates of PM/POC budgets and in modeling exercises. Using these equations, the average PM and POC stocks at the upper part of the water column (0–150 m) were estimated at 9.8  106 and 3.3  106 t, respectively. A comparison of the POC standing stock in the South Ionian Sea with available sediment trap data suggested that about 97% of the POC may be recycled in the upper 150 m, and only the remaining minor proportion escapes towards deeper depths. A closer inspection of the data set in different sectors of the Eastern Mediterranean, influenced by

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

permanent or semi-permanent cyclonic mesoscale features, revealed the close relationships between beam cp and primary productivity; in the Rhodes Gyre and the eastern Cretan margin, the upwelling of nutrient-rich waters substantially favored primary productivity, a process very important for the extremely oligotrophic Eastern Mediterranean Sea. In the case of Athens WWTP in the Saronikos Gulf, beam cp proved to be a useful tracer of effluent dispersal route into the sea, appearing as an intermediate nepheloid layer distinguishable over several kilometers from the underwater discharge point. Acknowledgments We wish to thank the coordinators and chief scientists of large-scale EU-funded and Hellenic projects who provided the raw LT, PMC and POC data, as well as the accompanying CTD data: G. Chronis, E. Balopoulos, A. Theocharis, A. Sioulas, V. Lykousis, I. Siokou, and E. Christou. The help of E. Kaberi, A. Papageorgiou, and E. Kabouri during field and laboratory work is gratefully acknowledged. We thank D. Raitsos for preparing the chlorophyll-a composite image and S. Kioroglou for his valuable help in data mining. This work was significantly improved by the insights and comments from two anonymous reviewers. Finally, the continuous support and assistance of the officers and crew of R/V Aegeao is highly appreciated. Part of this work has been supported by a Fulbright fellowship granted to A.P. Karageorgis at Texas A&M University, USA. References Abdel-Moati, A.R., 1990. Particulate organic matter in the subsurface chlorophyll maximum layer in the Southeastern Mediterranean. Oceanologica Acta 13, 307–315. Anati, D.A., 1984. A dome of cold water in the Levantine Basin. Deep-Sea Research 31 (10), 1251–1257. Baker, E.T., Hickey, B.M., 1986. Contemporary sedimentation processes in and around an active west coast submarine canyon. Marine Geology 71, 15–34. Baker, E.T., Lavelle, J.W., 1984. The effects of particle size on the light attenuation coefficient of natural suspensions. Journal of Geophysical Research 89, 8197–8203. Balopoulos, E., Theocharis, A., Kontoyiannis, H., Varnavas, S., Voutsinou-Taliadouri, F., Iona, A., Souvermezoglou, A., Ignadiades, L., Gotsis-Scretas, O., Pavlidou, A., 1999. Major advances in the oceanography of the southern Aegean Sea–Cretan Straits system (eastern Mediterranean). Progress in Oceanography 44, 109–130.

199

Bartz, R.J., Zaneveld, R.V., Pak, H., 1978. A transmissometer for profiling and moored observations in water. SPIE, Ocean Optics V, 102–108. Biscaye, P.E., Eittreim, S.L., 1977. Suspended particulate loads and transports in the nepheloid layer of the abyssal Atlantic Ocean. Marine Geology 23, 155–172. Biscaye, P.E., Flagg, C.N., Falkowski, P.G., 1994. The shelf edge exchange processes experiment, SEEP-II: an introduction to hypotheses, results and conclusions. Deep-Sea Research II 41, 231–252. Bishop, J.K.B., 1986. The correction and suspended particulate matter calibration of SeaTech transmissometer data. DeepSea Research 33, 121–134. Bricaud, A., Morel, A., Prieur, L., 1981. Absorption by dissolved organic matter of the sea (yellow substance) in the UV and visible domains. Limnology and Oceanography 26 (1), 43–53. Buesseler, K.O., 1998. The decoupling of production and particulate export in the surface ocean. Global Biogeochemical Cycles 12 (2), 297–310. Cacchione, D.A., Drake, D.E., 1986. Nepheloid layers and internal waves over continental shelves and slopes. GeoMarine Letters 6, 147–152. Castaing, P., Froidefond, J.M., Lazure, P., Weber, O., Prud’homme, R., Jouanneau, J.M., 1999. Relationship between hydrology and seasonal distribution of suspended sediments on the continental shelf of the Bay of Biscay. Deep-Sea Research II 46, 1979–2001. Cutter, G.A., Radford-Knoery, J., 1991. Determination of carbon, nitrogen, sulfur and inorganic sulfur species in marine particles. In: Hurd, D.C., Derek, D.W. (Eds.), Marine Particles: Analysis and Characterization, Geophysical Monograph, vol. 63, pp. 57–63. Danovaro, R., Dinet, A., Duineveld, G., Tselepides, A., 1999. Benthic response to particulate fluxes in different trophic environments: a comparison between the Gulf of Lions– Catalan Sea western Mediterranean) and the Cretan Sea (eastern-Mediterranean). Progress in Oceanography 44, 287–312. De Lazzari, A., Boldrin, A., Rabitti, S., Turchetto, M.M., 1999. Variability and downward fluxes of particulate matter in the Otranto Strait area. Journal of Marine Systems 20, 399–413. Deuser, W.G., Ross, E.H., Anderson, R.F., 1981. Seasonality in the supply of sediment to the deep Sargasso Sea and implications for the rapid transfer of matter to the deep ocean. Deep-Sea Research 28, 495–505. Drake, D.E., 1971. Suspended sediment and thermal stratification in Santa Barbara Channel, California. Deep-Sea Research 18, 763–769. Dugdale, R.C., Wilkerson, F.R., 1988. Nutrient sources and primary production in the Eastern Mediterranean. Oceanologica Acta 9, 178–184. Durand, N., Fiandrino, A., Fraunie´, P., Ouillon, S., Forget, P., Naudin, J.J., 2002. Suspended matter dispersion in the Ebro ROFI: an integrated approach. Continental Shelf Research 22, 267–284. Durrieu de Madron, X., 1994. Hydrography and nepheloid structures in the Grand-Rhoˆne canyon. Continental Shelf Research 14, 457–477. Durrieu de Madron, X., Nyffeler, F., Godet, C.H., 1990. Hydrographic structure and nepheloid spatial distribution in the Gulf of Lions continental margin. Continental Shelf Research 10, 915–929.

ARTICLE IN PRESS 200

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

Durrieu de Madron, X., Nyffeler, F., Balopoulos, E.T., Chronis, G., 1992. Circulation and distribution of suspended matter in the Sporades Basin (northwestern Aegean Sea). Journal of Marine Systems 3, 237–248. Ediger, D., Tug˘rul, S., Yilmaz, A., 2005. Vertical profiles of particulate organic matter and its relationship with chlorophyll-a in the upper layer of the NE Mediterranean Sea. Journal of Marine Systems 55, 311–326. Eittreim, S., Thorndike, E.M., Sullivan, L., 1976. Turbidity distribution in the Atlantic Ocean. Deep-Sea Research 23, 1115–1127. Frignani, M., Courp, T., Cochran, J.K., Hirschberg, D., Vitoria i Codina, L., 2002. Scavenging rates and particle characteristics in and near the Lacaze-Duthiers submarine canyon, northwest Mediterranean. Continental Shelf Research 22, 2175–2190. Gardner, W.D., 1989a. Baltimore Canyon as a modern conduit of sediment to the deep sea. Deep-Sea Research 36, 323–358. Gardner, W.D., 1989b. Periodic resuspension in Baltimore Canyon by focussing of internal waves. Journal of Geophysical Research 94 (C12), 18,185–18,194. Gardner, W.D., Biscaye, P.E., Zaneveld, J.R.V., Richardson, M.J., 1985. Calibration and comparison of the LDGO nephelometer and the OSU transmissometer on the Nova Scotian Rise. Marine Geology 66, 323–344. Gardner, W.D., Walsh, I.D., Richardson, M.J., 1993. Biophysical forcing of particle production and distribution during a spring bloom in the North Atlantic. Deep-Sea Research II 40 (1–2), 171–195. Gardner, W.D., Chung, S.P., Richardson, M.J., Walsh, I.D., 1995. The oceanic mixed-layer pump. Deep-Sea Research II 42 (2–3), 757–775. Gardner, W.D., Richardson, M.J., Carlson, C.A., Hansell, D., Mishonov, A.V., 2003. Determining true particulate organic carbon: bottles, pumps and methodologies. Deep-Sea Research II 50 (3–4), 655–674. Gardner, W.D., Mishonov, A.V., Richardson, M.J., 2006. Global POC concentrations from in-situ and satellite data. Deep-Sea Research II 53, 718–740. Georgopoulos, D., 2002. Water masses, dynamic structure and circulation in the north Aegean Sea. Ph.D. Thesis, Univ. Patras, 324pp. Georgopoulos, D., Theocharis, A., Zodiatis, G., 1989. Intermediate water formation in the Cretan Sea (South Aegean Sea). Oceanologica Acta 12 (4), 353–359. Gertman, I.F., Ovchinnikov, I.M., Popov, Y.I., 1994. Deep convection in the eastern basin of the Mediterranean Sea. Oceanology 34, 19–25. Gordon, D.C., 1970. Some studies of the distribution and composition of particulate organic carbon in the North Atlantic Ocean. Deep-Sea Research 17, 233–244. Guerzoni, S., Chester, R., Dulac, F., Herut, B., Loye-Pilot, M.D., Measures, C., Migon, C., Molinaroli, E., Moulin, C., Rossini, P., Saydam, C., Soudine, A., Ziveri, P., 1999. The role of atmospheric deposition in the biogeochemistry of the Mediterranean Sea. Progress in Oceanography 44, 147–190. Gundersen, J.S., Gardner, W.D., Richardson, M.J., Walsh, I.D., 1998. Effects of monsoons on the seasonal and spatial distributions of POC and chlorophyll in the Arabian Sea. Deep-Sea Research II 45, 2103–2132.

Hickey, B., Baker, E., Kachel, N., 1986. Suspended particle movement in and around Quinault submarine canyon. Marine Geology 71, 35–83. Hopkins, T.S., 1974. A discussion of the marine pollution problems in the Saronikos Gulf as disclosed by current research. Unpublished Technical Report IOKAE, 17pp. Hopkins, T.S., 1978. Physical processes in the Mediterranean basin. In: Kjerfve, B. (Ed.), Estuarine Transport Processes. University of South Carolina Press, Columbia, pp. 269–310. Hopkins, T.S., Coachman, L.K., 1975. Circulation patterns in the Saronikos Gulf in relation to the winds. Unpublished Interim Technical Report 1, Environmental Pollution Control Project-Athens, Ministry of Public Works, Athens, 66pp. Ignatiades, L., 1998. The productive and optical status of the oligotrophic waters of the Southern Aegean Sea (Cretan Sea), Eastern Mediterranean. Journal of Plankton Research 20 (5), 985–995. Ignatiades, L., Psarra, S., Zervakis, V., Pagou, K., Souvermezoglou, E., Assimakopoulou, G., Gotsis-Skretas, O., 2002. Phytoplankton size-based dynamics in the Aegean Sea (Eastern Mediterranean). Journal of Marine Systems 36, 11–28. IOC, IHO, BODC, 2003. Centenary Edition of the GEBCO Digital Atlas. Published on CD-ROM on behalf of the Intergovernmental Oceanographic Commission and the International Hydrographic Organization as part of the General Bathymetric Chart of the Oceans. British Oceanographic Data Centre, Liverpool. Jerlov, H.G., 1968. Optical Oceanography. Elsevier Science, NewYork, 194 p. Karageorgis, A.P., Anagnostou, Ch.L., 2001. Particulate matter spatial-temporal distribution and associated surface sediment properties: Thermaikos Gulf and Sporades Basin, NW Aegean Sea. Continental Shelf Research 21, 2141–2153. Karageorgis, A.P., Anagnostou, Ch., 2003. Seasonal variation in the distribution of suspended particulate matter in the NW Aegean Sea. Journal of Geophysical Research 108 (C8), 3274. Karageorgis, A.P., Stavrakakis, S., 2005. Particulate matter dynamics and fluxes. In: Papathanassiou, E., Zenetos, A. (Eds.), State of the Hellenic Marine Environment. HCMR publ., pp. 78–87. Karageorgis, A.P., Anagnostou, Ch., Georgopoulos, D., Albuisson, M., 2000. Distribution of suspended particulate matter determined by in-situ observations and satellite images in the NW Aegean Sea (Greece). Geo-Marine Letters 20 (2), 93–100. Karageorgis, A.P., Kaberi, H.G., Tengberg, A., Zervakis, V., Anagnostou, Ch.L., Hall, P., 2003. Comparison of the particulate matter distribution between a mesotrophic and an oligotrophic marine area: the Skagerrak Sea and the northeastern Aegean Sea. Continental Shelf Research 23 (17–19), 1787–1809. Kawahata, H., 2002. Suspended and settling particles in the Pacific. Deep-Sea Research II 49, 5647–5664. Kontoyiannis, H., Papadopoulos, V., 1999. Hydrography and circulation. In: Siokou-Frangou, I. (Ed.), Monitoring of the Saronikos Gulf Ecosystem affected by the Psittalia Sea Outfall. NCMR, October 1999, Final Report, pp. 21–64. Krasakopoulou, E., Karageorgis, A.P., 2005. Spatial and temporal distribution patterns of suspended particulate matter and particulate organic carbon in the Saronikos Gulf (eastern Mediterranean, Greece). Geo-Marine Letters 25 (6), 343–359.

ARTICLE IN PRESS A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202 Kubilay, N., Saydam, A.C., 1995. Trace elements in atmospheric particulates over the Eastern Mediterranean; concentrations, sources and temporal variability. Atmospheric Environment 29 (17), 2289–2300. Lacombe, H., Tchernia, P., Benoist, G., 1958. Contribution a l’ e´tude hydrologique de la Mer Ege´e en pe´riode d’ e´te´. Bulletin d’ Information Cahiers Oce´anographique du C.O.E.C. 8, 454–468. Lapouyade, A., Durrieu De Madron, X., 2001. Seasonal variability of the advective transport of particulate matter and organic carbon in the Gulf of Lion (NW Mediterranean). Oceanologica Acta 24, 295–312. Legendre, P., 2001. Model II Regression-User’s Guide. De´partement de sciences biologiques, Universite´ de Montre´al, 23p. /http://www.bio.umontreal.ca/legendre/index.htmlS. Li, K.W., Zohary, T., Yacobi, Y.Z., Wood, A.M., 1993. Ultraphytoplankton in the Eastern Mediterranean Sea: towards deriving phytoplankton biomass from flow cytometric measurements of abundance, fluorescence and light scatter. Marine Ecology Progress Series 102, 79–87. Lykousis, V., Chronis, G., Tselepides, A., Price, B., Theocharis, A., Siokou-Frangou, I., Van Wambeke, F., Danovaro, R., Stavrakakis, S., Duineveld, G., Georgopoulos, G., Ignatiades, L., Souvermezoglou, A., Voutsinou-Taliadouri, F., 2002. Major outputs of the recent multidisciplinary biogeochemical researches undertaken in the Aegean Sea. Journal of Marine Systems 33–34, 313–334. Lykousis, V., Karageorgis, A.P., Chronis, G.T., 2005. Delta progradation and sediment fluxes since the last glacial in the Thermaikos Gulf and the Sporades Basin, NWAegean Sea, Greece. Marine Geology 222–223, 381–397. Maillard, C., Balopoulos, E., Giorgetti, A., Fichaut, M., Iona, A., Larour, M., Latrouite, A., Manca, B., Maudire, G., Nicolas, P., Sanchez-Cabeza, J.A., 2002. An integrated system for managing multidisciplinary oceanographic data collected in the Mediterranean Sea during the basin-scale research project EU/MAST-MATER (1996–2000). Journal of Marine Systems 33-34, 523–538. Malanotte-Rizzoli, P., Manca, B.B., d’ Alcala, M.R., Theocharis, A., Bergamasco, A., Bregant, D., Butillon, G., Civitarese, G., Georgopoulos, D., Michelato, A., Sansone, E., Scarazzato, P., Souvermezoglou, E., 1997. A synthesis of the Ionian Sea hydrography, circulation and water mass pathways during POEM-Phase I. Progress in Oceanography 39, 153–204. Malanotte-Rizzoli, P., Manca, B.B., d’Alcala, M.R., Theocharis, A., Brenner, S., Budillon, G., Ozsoy, E., 1999. The Eastern Mediterranean in the 80 s and in the 90s: the big transition in the intermediate and deep circulations. Dynamics of Atmospheres and Oceans 29, 365–395. Martin, J.M., Elbaz-Poulichet, F., Guieu, C., Loy¨e-Pilot, D., Han, G., 1989. River versus atmospheric input of material to the Mediterranean Sea: an overview. Marine Chemistry 28, 159–182. McCave, I.N., Hall, I.R., 2002. Turbidity of waters over the Northwest continental margin. Progress in Oceanography 52, 299–313. McCave, I.N., Hall, I.R., Antia, A.N., Chou, L., Dehairs, F., Lampitt, R.S., Thomsen, L., Van Weering, T.C.E., Wollast, R., 2001. Distribution, composition and flux of particulate material over the European margin at 47–501N. Deep-Sea Research II 48, 3107–3139. Monaco, A., Biscaye, P.E., Soyer, J., Pocklington, R., Heussner, S., 1990. Particle fluxes and ecosystem response on a

201

continental margin: the 1985–1988 Mediterranean ECOMARGE experiment. Continental Shelf Research 10, 809–839. Moody, J.A., Butman, B., Bothner, M.H., 1987. Near-bottom suspended matter concentration on the Continental Shelf during storms: estimates based on in situ observations of light transmission and a particle size dependent transmissometer calibration. Continental Shelf Research 7 (6), 609–628. Moran, S.B., Charette, M.A., Pike, S.M., Wickland, C.A., 1999. Differences in seawater particulate organic carbon concentration in samples collected using small-volume and largevolume methods: the importance of DOC absorption to the filter blank. Marine Chemistry 67, 33–42. Moutin, T., Raimbault, P., 2002. Primary production, carbon export and nutrients availability in western and Eastern Mediterranean Sea in early summer 1996 (MINOS cruise). Journal of Marine Systems 33-34, 273–288. Nihle´n, T., Mattsson, J.O., Rapp, A., Gagaoudaki, C., Kornaros, G., Papageorgiou, J., 1995. Monitoring of Saharan dust fallout on Crete and its contribution on soil formation. Tellus 47B, 365–374. Ovchinnikov, I.M., 1966. Circulation in the surface and intermediate layers of the Mediterranean Sea. Oceanology 18 (1), 165–168. Ovchinnikov, I.M., Plakhin, E.A., 1984. Formation of intermediate waters of the Mediterranean Sea in the Rhodes cyclonic Gyre. Oceanology 23, 960–962. Ozsoy, E., Hecht, A., U¨nlu¨ata, U., 1989. Circulation and hydrography of the Levantine Basin. Results of POEM coordinated experiments 1985–1986. Progress in Oceanography 22, 125–170. Ozsoy, E., Hecht, A., U¨nlu¨ata, U., Brenner, S., Sur, H.I., Bishop, J., Latif, M.A., Rozentraub, Z., Oguz, T., 1993. A synthesis of Levantine basin circulation and hydrography 1985–1990. Deep-Sea Research II 40, 1075–1120. Palanques, A., Biscaye, P.E., 1992. Patterns and controls of the suspended matter distribution over the shelf and upper slope south of New England. Continental Shelf Research 12 (5/6), 577–600. Park, Y.A., Choi, J.Y., Gao, S., 2001. Spatial variation of suspended particulate matter in the Yellow Sea. Geo-Marine Letters 20, 196–200. Poulos, S.E., 2001. The contribution of near-bed currents to modern sedimentation processes in the deep waters of the Hellenic Arc–Trench system, Eastern Mediterranean. GeoMarine Letters 20, 201–208. Poulos, S.E., Chronis, G., 1997. The importance of the river systems in the evolution of the Greek coastline. In: Briand, F., Maldonado, A. (Eds.), Transformations and Evolution of the Mediterranean Coastline. Bulletin de l’Institut Oce´anographique Monaco, CIESM Science Series No. 3, vol. 18, pp. 75–96. Poulos, S.E., Collins, M.B., 2002. Fluviatile sediment fluxes to the Mediterranean Sea: a quantitative approach and the influence of dams. In: Jones, S.J., Frostick, L.E. (Eds.), Sediment Flux to Basins: Causes, Controls and Consequences, vol. 191. Geological Society, Special Publications, London, pp. 227–245. Poulos, S.E., Drakopoulos, P.G., Collins, M.B., 1997. Seasonal variability in sea surface oceanographic conditions in the Aegean Sea (Eastern Mediterranean): an overview. Journal of Marine Systems 13, 225–244. Price, N.B., Lindsay, F.S., Pates, J.M., 1999. The biogeochemistry of major elements of suspended particulate matter of the Cretan Sea. Progress in Oceanography 44, 677–699.

ARTICLE IN PRESS 202

A.P. Karageorgis et al. / Deep-Sea Research I 55 (2008) 177–202

Psarra, S., Tselepides, A., Ingnatiades, L., 2000. Primary productivity in the oligotrophic Aegean Sea (NE Mediterranean): seasonal and interannual variability. Progress in Oceanography 46 (2–4), 187–204. Puig, P., Palanques, A., 1998. Nepheloid structure and hydrographic control on the Barcelona continental margin, northwestern Mediterranean. Marine Geology 149, 39–54. Puig, P., Palanques, A., Guille´n, J., El Khatab, M., 2004. Role of internal waves in the generation of nepheloid layers on the northwestern Alboran slope: implications for continental margin shaping. Journal of Geophysical Research 109, C09011. Rabitti, S., Bianchi, F., Boldrin, A., Da Ros, L., Socal, G., Totti, C., 1994. Particulate matter and phytoplankton in the Ionian Sea. Oceanologica Acta 17, 297–303. Richardson, M.J., 1987. Particle size, light scattering and composition of suspended particulate matter in the North Atlantic. Deep-Sea Research 34, 1301–1329. Robinson, A.R., Golnaraghi, M., 1994. The physical and dynamical oceanography of the Mediterranean Sea. In: Malanotte-Rizzoli, P., Robinson, A.R. (Eds.), Ocean Processes in Climate Dynamics: Global and Mediterranean Examples, NATO ASI Series C, vol. 419. Kluwer Academic Publishers, Dordrecht, Boston, London, pp. 255–306. Roether, W., Manca, B.B., Klein, B., Bregant, D., Georgopoulos, D., Beitzel, V., Kovacevic, V., Luchetta, A., 1996. Recent changes in the Eastern Mediterranean deep water. Science 271, 333–335. Sakellariou, D., Lykousis, V., Karageorgis, A.P., Anagnostou, C., 2005. Geomorphology and tectonic structure. In: Papathanassiou, E., Zenetos, A. (Eds.), State of the Hellenic Marine Environment. HCMR publ., pp. 16–20. Schlitzer, R., 2003. Ocean Data View /http://www.awi-bremerhaven. de/GEO/ODVS. Schlitzer, R., Roether, W., Oster, H., Junghans, H.-G., Hausmann, M., Johannsen, H., Michelato, A., 1991. Chlorofluoromethane and oxygen in the Eastern Mediterranean. DeepSea Research 38, 1531–1551. Siokou-Frangou, I., Gotsis-Skretas, O., Christou, E.D., Pagou, K., 1999. Plankton characteristics in the Aegean, Ionian and NW Levantine Seas. In: Malanotte-Rizzoli, P., Eremeev, V.N. (Eds.), The Eastern Mediterranean as a Laboratory Basin for the Assessment of Contrasting Ecosystems, vol. 51. NATO Science Series, vol. 2. Kluwer Academic Publishers, Dordrecht, Boston, London, pp. 465–473. Socal, G., Boldrin, A., Bianchi, F., Civitarese, G., De Lazzari, A., Rabitti, S., Totti, C., Turchetto, M.M., 1999. Nutrient, particulate matter and phytoplankton variability in the photic layer of the Otranto strait. Journal of Marine Systems 20, 381–398. Souvermezoglou, E., Krasakopoulou, E., 1999. The effect of physical processes on the distribution of nutrients and oxygen in the NW Levantine Sea. In: Malanotte-Rizzoli, P., Eremeev, V.N. (Eds.), The Eastern Mediterranean as a Laboratory Basin for the Assessment of Contrasting Ecosystems, vol. 51. NATO Science Series, vol. 2. Kluwer Academic Publishers, Dordrecht, Boston, London, pp. 225–240. Spinrad, R.W., 1986. An optical study of the water masses in the gulf of Maine. Journal of Geophysical Research 91, 1007–1018.

Spinrad, R.W., Zaneveld, J.R.V., Kitchen, J.C., 1983. A study of the optical characteristics of the suspended particles in the benthic boundary layer of the Scotian Rise. Journal of Geophysical Research 88, 7641–7645. The LIWEX Group, 2003. The Levantine intermediate water experiment (LIWEX) group: Levantine basin—a laboratory for multiple water mass formation processes. Journal of Geophysical Research 108 (C9), 8101. The POEM Group, 1992. General circulation of the Eastern Mediterranean. Earth Science Reviews 32, 285–309. Theocharis, A., Kontoyiannis, H., 1999. Interannual variability of the circulation and hydrography in the Eastern Mediterranean (1986–1995). In: Malanotte-Rizzoli, P. (Ed.), The Eastern Mediterranean as a Laboratory Basin for the Assessment of Contrasting Ecosystems. Kluwer Academic Publishers, Norwell, MA, pp. 453–464. Theocharis, A., Georgopoulos, D., Lascaratos, A., Nittis, K., 1993. Water masses and circulation in the central region of the Eastern Mediterranean (E. Ionian, S. Aegean and NW Levantine). Deep-Sea Research II 40 (6), 1121–1142. Theocharis, A., Gacic, M., Kontoyiannis, H., 1998. Physical and dynamical processes in the coastal and shelf areas of the Mediterranean. In: Robinson, A.R., Brink, K.H. (Eds.), The Sea, vol. 11. Wiley, New York, pp. 863–887. Theocharis, A., Balopoulos, E., Kioroglou, S., Kontoyiannis, H., Iona, A., 1999. A synthesis of the circulation and hydrography of the South Aegean Sea and the Straits of the Cretan Arc (March 1994–January 1995). Progress in Oceanography 44 (4), 469–509. Tselepides, A., Zervakis, V., Polychronaki, T., Danovaro, R., Chronis, G., 2000. Distribution of nutrients and POM in relation to the prevailing hydrographic features of the Cretan Sea (NE Mediterranean). Progress in Oceanography 46 (2–4), 113–142. U¨nlu¨ata, U., Oguz, T., Latif, M.A., Ozsoy, E., 1990. On the physical oceanography of the Turkich straits. In: Pratt, L.J. (Ed.), The Physical Oceanography of Sea Straits. Kluwer Academics Publishers, Norwell, MA, pp. 25–60. Verardo, D.J., Froelich, P.N., McIntyre, A., 1990. Determination of organic carbon and nitrogen in marine sediments using the Carlo Erba NA-1500 analyzer. Deep-Sea Research 37 (1), 157–165. Vidussi, F., Claustre, H., Manca, B.B., Luchetta, A., Marty, J.C., 2001. Phytoplankton pigment distribution in relation to upper thermocline circulation in the eastern Mediterranean. Journal of Geophysical Research 106 (C9), 19,939–19,956. Yacobi, Y.Z., Zohary, T., Kress, N., Hecht, A., Robarts, R.D., Waiser, M., Wood, A.M., 1995. Chlorophyll distribution throughout the southeastern Mediterranean in relation to the physical structure of the water mass. Journal of Marine Systems 6 (3), 179–190. Zervakis, V., Georgopoulos, D., 2002. Hydrology and circulation in the North Aegean (Eastern Mediterranean) throughout 1997 and 1998. Mediterranean Marine Science 3 (1), 5–19. Zervakis, V., Georgopoulos, D., Drakopoulos, P.G., 2000. The role of the North Aegean in triggering the recent Eastern Mediterranean climatic changes. Journal of Geophysical Research 105 (C11), 26,103–26,116. Zodiatis, G., 1994. Advection of the Black Sea water in the North Aegean Sea. The Global Atmosphere and Ocean System 2, 41–60.