**2. Material and methods**

#### **2.1 Sampling sites**

248 Earthquake Research and Analysis – Statistical Studies, Observations and Planning

be in a small city (Gölcük) located on the southern coast of Izmit Bay. This seismic event caused the destruction of wastewater discharge systems and also dispersal of refined petroleum products onto the sea surface from the subsequent refinery fire. The surface waters of the Bay were partly covered by the thick petroleum layers and partly by a film (Güven et al., 2000, Ünlü et al., 2000). Petroleum layer covering the surface water reduced the transfer of oxygen from air/sea interface and also caused the subsequent death of living organisms. Increasing effluent discharges into the Bay produced an exceptional plankton bloom. Coupling of such factors leading to oxygen deficiency at the sea floor caused the formation of anoxic conditions. Okay et al., (2001) investigated ecological changes in Izmit

This paper presents the results of one-year monitoring program performed in Izmit Bay after the Earthquake, with the purposes of describing the abrupt changes in chemical oceanography and understanding the mechanism of H2S generation in the Bay which has not been occurred before. Furthermore, the factors controlling metal distributions in water

Izmit Bay is an elongated semi-enclosed water body with a length of 50 km, width varying between 2 to 10 km (Figure 1) and has an area of 310 km2. The bathymetry of the Bay constitutes three sub-basin separated by shallow sills from each other. The eastern basin is relatively shallow (at about 30 m) whereas the central basin has two small depressions with depths of 160 and 200 m. The western basin deepens in westward from 150 m to 300 m and connects the Bay to the Marmara Sea. Izmit Bay is oceanographically an extension of Marmara Sea, having a permanent two-layered water system. The upper layer is originated from less saline Black Sea waters (18.0-22.0 psu), whereas the lower layer originated from the Mediterranean Sea waters is more saline (37.5-38.5 psu) (Ünlüata et al., 1990). The permanent stratification occurs at about 25 m in the Marmara Sea (Beşiktepe et al., 1994), however it is highly variable in Izmit Bay (Oğuz and Sur, 1986) (Figure 2). The thickness of the upper layer changes seasonally from 9 to 18 m spring and autumn, respectively (Oğuz and Sur, 1986; Algan et al., 1999). The upper layer enters into the Bay in spring and summer, corresponding to the freshwater inflow changes in the Black Sea, while the lower layer flows to the Marmara Sea from the Bay. However, the upper layer flows towards the Marmara Sea in autumn and winter (Oğuz and Sur, 1986). Vertical mixing of the two layers is restricted and occurs at shallow depths. An intermediate layer develops throughout the year in the water column of the Bay with varying thickness (DAMOC, 1971; Baştürk et al., 1985; Tuğrul et al., 1989; Oğuz and Sur, 1986; Altok et al., 1996). The upper layer of Izmit Bay, in general, is saturated with DO (Tuğrul and Morkoç, 1990). DO concentrations in the lower layer of Izmit Bay has been found to be 2.5-3.0 mgl-1 in winter and spring periods and 0.7-1.5 mgl-1 in summer, in previous studies (Morkoç, et al., 1996). The minimum DO concentrations have been measured locally in the central basin (0.1-0.2 mgl-1) and in the eastern basin (0.5 mgl-1) during spring-summer period (Tuğrul and Morkoç, 1990). Izmit Bay and its surroundings is one of the most industrialized and populated area of Turkey, receiving more than 300 industrial and domestic effluents (Morkoç et al., 1996). Industrial effluents discharges a total of 163,000 m3/day wastewater, 24 tons/day BOD and 19,5 tons/day TSS to Izmit Bay (Morkoç et al., 2001). The eastern basin receives the highest inputs compare to other basins of the Bay. Based on the previous studies, no DHS has been measured in Izmit Bay (Morkoç

Bay, however their data is limited with the September 1999.

et al., 1988; Tuğrul et al., 1989; Morkoç et al., 1996).

**1.1 Study area** 

column and surface sediments of the Bay were discussed in this study.

The water samples were collected from 32 stations in İzmit Bay, including one station located off the western basin (R), on board the R/V Arar (Figure 1). Station (R) represents the characteristics of the Marmara Sea and hence, provides a comparison between the Bay and the Marmara Sea. The sampling stations in İzmit Bay represent the various depths of three basins, with a minimum of 17 m and maximum of 200 m. Sampling was carried out with a Rosette sampler assembled to the Sea Bird CTD System at about 10 m depth intervals through the upper and the lower layers. Sampling period includes August 1999, immediately after the Earthquake and performed monthly in 1999 and in February, May and August during 2000.

The Effect of Marmara (Izmit) Earthquake on the Chemical

titration method (Strickland and Parsons, 1972).

0.08; Cd, 0.05 0.03 mg/l; Pb, 0.20 0.30 mg/l.

HClO4 + HF acid mixture.

Reference

(Figure 1). The precision of method was estimated at 1.9 % .

equipped with a temperature-compensation adjustment on board.

Material Element Measured value (this study)

(SL-1 and SL-7), IAEA405 and BCR (CRM 142) reference materials. SL-1 and SL-7 are lake sediment and CRM 142 is a light sandy soil.

**2.2 Analytical methods** 

Oceanography and Mangan Enrichment in the Lower Layer Water of Izmit Bay, Turkey 251

Samples for DO determinations were drawn first from the Niskin bottles of Rosette to prevent any biological activity and gas exchanges with the atmosphere. DO determination was carried out by Winkler method (Greenberg et al., 1985) on board from all the stations

Dissolved hydrogen sulfide (DHS) was measured only at stations where DO concentrations are lower than the detection limit of the method (0.03 mgl-1) (Figure 1), in all the sampling periods, except August and September 1999. DHS contents were measured by an iodometric

pH values measured along the water column at all stations with a WTW 526 pH-meter

The water samples were filtered through 0.45 µm filters using metal clean tecniques (Bruland et al., 1979). The samples were stored in polyethylene bottles (LDPE) that were acid cleaned using methods described Patterson and Settle (1976). After collection, the samples were acidified to a pH between 1.5 and 2.0 using HNO3. Dissolved heavy metal concentrations (Fe, Mn, Pb, Cu and Cd) were measured by atomic absorption spectrophotometer (AAS) following preconcentration with ammonium 1 pyrrolidinedithiocarbamate (APDC) in an organic extraction (Bruland et al., 1985). The blanks for the metals analyzed were: Fe, 010 0.05 mg/l; Mn, 0.10 0.02 mg/l; Cu, 0.15

The surface sediments total carbonate contents were determined by a gasometric-volumetric method (Loring Rantala, 1992). Total organic carbon (Corg) was analyzed by the Walkey-Blake method, which involves titration after a wet combustion of the sample (Gaudette, 1974; Loring Rantala, 1992). Al, Fe, Mn, Cu, Zn, Co and Cr contents were determined by atomic absorption spectrophotometer (AAS) after a "total" digestion, involving HNO3 +

The sequential selective extraction analyses were carried out using 1M Na-acetate (pH=5 adjusted by acetic acid) for the dissolution of carbonate phase, 0.04M hydroxylamine hydrochloride (HAHC) in 25% acetic acid for dissolving Fe-Mn-oxyhydroxides, 0.02M nitric acid + 30% hydrogen peroxide (pH=2) for extracting organic matter, and HNO3 + HClO4 + HF mixture for the total extraction of the residual (lithogenous) fraction (Tessier et al., 1979).

ppm

SL-1 Fe 62 65-7-69.1 IAEA405 Al 63500 72700-83100 SL-1 Cr 98 95-113 SL-7 Mn 634 604-650 CRM-142 Cu 25 27.5 CRM-142 Zn 92 92.4 CRM-142 Co 13 7.9

Table 1. Accuracy of ASS analyses used in this study, as determined by Analysis of AQCS

Certified value or range ppm

Fig. 2. Depth profile of salinity and temperature distribution along the water column of some selected station from İzmit Bay (from Güven et al., 2000).
