See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/227871984

A case study of sewage discharge in the shallow

coastal area of the Northern Adriatic Sea (Gulf
of Trieste)

Article in Marine Ecology · December 2008

DOI: 10.1111/j.1439-0485.2008.00257.x

CITATIONS READS

20 79

3 authors:

Patricija Mozetič Vlado Malacic

National Institute of Biology - Nacionalni inšt… National Institute of Biology - Nacionalni inšt…
56 PUBLICATIONS 728 CITATIONS 70 PUBLICATIONS 723 CITATIONS

SEE PROFILE SEE PROFILE

Valentina Turk
National Institute of Biology - Nacionalni inšt…
45 PUBLICATIONS 768 CITATIONS

SEE PROFILE

All content following this page was uploaded by Valentina Turk on 18 May 2015.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.
 Marine Ecology. ISSN 0173-9565

ORIGINAL ARTICLE

A case study of sewage discharge in the shallow coastal

area of the Northern Adriatic Sea (Gulf of Trieste)
Patricija Mozetič, Vlado Malačič & Valentina Turk
Marine Biology Station, National Institute of Biology, Piran, Slovenia

Keywords Abstract
Adriatic Sea; eutrophication; faecal coliforms;
Gulf of Trieste; nutrients; phytoplankton; A case study was carried out in 2000 in the shallow coastal area of the Northern
sewage. Adriatic Sea (Gulf of Trieste) where untreated domestic sewage and industrial
wastes are discharged at rate of 5500 m3Æday)1. The sewage plume above the
Correspondence outfall was followed using faecal coliforms (FC) and overturning length scale
Patricija Mozetič, Marine Biology Station,
(lT). The latter was rejected as a marker as the discharge conditions prohibit fol-
National Institute of Biology, Fornače 41,
6330 Piran, Slovenia.
lowing the turbulence of sewage water. Intermittent sewage discharge is reflected
E-mail: mozetic@mbss.org in the minimal effect of eutrophication. Increase of phytoplankton biomass is
thus only minor compared with the unpolluted area regardless of elevated con-
Accepted: 22 May 2008 centrations of sewage-derived nutrients (confirmed by correlation coefficients
between FC and NH4+, TP, PO43): 0.78, 0.71 and 0.67, respectively). Deterio-
doi:10.1111/j.1439-0485.2008.00257.x rated trophic status, determined by the TRIX index, was observed only in the
surface layer (average TRIX: 5.67). High FC content well above the regulation
limit (up to 2.6 · 105 FCÆ100 ml)1) represents, therefore, the major negative
impact of the improperly treated waste for the risk to human health.

Braga et al. 2000; Aslan-Yilmaz et al. 2004). Besides faecal

Problem
pollution and the deteriorative appearance of bathing
Coastal waters are highly variable ecosystems due to their waters, which are the immediate consequences of waste-
natural characteristics, and may become even more vul- water discharge, the other important aspect is eutrophica-
nerable when exposed to negative consequences of tion, which can be viewed as decreased biomass and
human activities. In the shallow and semi-enclosed Gulf number of species, proliferation of persistent species and
of Trieste (Northern Adriatic Sea) the variability of community changes, mostly in benthic communities
hydrological, physico-chemical and biological characteris- (Chisholm et al. 1997; Soltan et al. 2001; Savage et al.
tics has been attributed mostly to its geo-morphological 2002), whereas the impact of sewage on pelagic commu-
properties and the freshwater regimes of rivers that enter nities is more difficult to detect. The dilute nature of the
the Gulf (Olivotti et al. 1986; Mozetič et al. 1998; Malačič effluent obscures the impact of sewage discharge on the
& Petelin 2001). The short southern coastline is also den- already complex coastal environment and thus makes dis-
sely populated (232 individualsÆkm)2) (Turk & Potočnik tinguishing between the ‘background’ of receiving waters
2001) and concentrates several activities that exploit the and wastewater an almost impossible task (Faganeli 1982;
economic value of the sea. It is therefore important that Mozetič et al. 1999), even in areas with high effluent
pollution pressure, which derives either from land-based flows and non-efficient wastewater treatments (Smith-
sources or from different activities at sea, is kept under Evans & Dawes 1996; da Silva et al. 2002). The greatest
control. effects are usually observed in the immediate vicinity of
Largely untreated domestic and industrial wastes are the sewage discharge and rapidly decrease towards the
considered one of the main causes of the deterioration of open sea, where polluted and brackish waters are no
the state of the Mediterranean Sea (UNEP ⁄ WHO 2007) longer detectable (Theodorou 1992; Torres & del Rio
as well as seawaters worldwide (e.g. Adingra & Arfi 1998; 1995). It is therefore crucial to describe the sewage plume

Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 483
Case study of sewage discharge Mozetič, Malačič & Turk

Fig. 1. Location of the sampling station IZ06 above the endpoint of the underwater pipe, which releases wastewaters from the municipality of
Izola into the coastal waters. The rectangle denotes the grid of 11 stations, with station IZ06 being the central one. The position of the non-
impacted, reference station 000F is also shown.

as precisely as possible and to measure the variations in tic sewage and 10,000 PE for industrial waste (Turk &
parameters that most likely reflect the impact of sewage. Potočnik 2001). Waste is collected in a treatment basin
Such a study was performed in 2000 in the coastal area where only rough mechanical purification (screening by
of the Gulf of Trieste, which receives largely untreated 1-cm rake) is carried out. The effluent is pumped from
sewage from a small town on the Slovenian coast (Izola; the sedimentation basin and discharged into the 12-m-
Fig. 1). The sampling design followed that applied in the deep seawater column through an approximately 200-m-
similar study of the outfalls of the neighbouring town long pipe. In 2000 the average daily flow rate of effluent
Piran (Malačič 2005; Malačič & Mozetič 2005), where the was estimated to be around 5300 m3Æday)1 considering
overturning length scale was adopted. This is the measure the total yearly amount of working hours (7739 h),
of higher turbulence intensity and can serve as an indica- capacity (70 lÆs)1) and average daily operation (21.2 h) of
tion of a formed patch of pollutant in the surrounding, the pump (E. Požar, personal communication). It must
i.e. marine environment. be stressed that the pump is switched on when the basin
Areas in the Slovenian coastal sea influenced by waste- is filled. Just before the pipe sweeps into the sea, a dis-
water discharges have been surveyed for years within the charge from the fish-cannery pre-treatment plant (flow
framework of monitoring activities of the Barcelona con- rate 225 m3Æday)1) joins the last ground collector of the
vention. This study represents, therefore, an upgrade of sewage system, thus influencing the nature of the effluent.
these activities and it also meets some objectives of the
initiative INTERREG III CBC Phare (Italy-Slovenia), pro-
Fieldwork
ject OBAS (Biological Oceanography of the Northern
Adriatic Sea), carried out during 2000–2006. Four samplings were performed in the year 2000 to define
the impact of Izola’s sewage discharge on seawater prop-
erties (ecological survey) and to detect its spread across
Material and Methods the sea surface and along the vertical (CTD survey). To
accomplish the latter a detailed oceanographic survey,
Wastewater disposal and sampling location
including distribution of the overturning length scale, was
The pollution load of the Izola municipality is estimated carried out on 10 May 2000 on a grid of 11 stations dis-
to be up to 23,000 population equivalent (PE) for domes- tributed above the outlet (Fig. 1). Within the grid, the

484 Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd
Mozetič, Malačič & Turk Case study of sewage discharge

stations were separated by 0.1¢ (185 m) in S–N direction drop speed of about 1 mÆs)1 (instrumental details in
and 0.1¢ (130 m) in E–W direction. Malačič 2005). The overturning length scale (lT) was cal-
Surveys of the chemical and biological properties of the culated as the root-mean square (rms) of vertical dis-
sea water were carried out at the central station of the placement over layers of thickness of 0.5 m.
grid – IZ06 (4532.64¢ N, 1339.75¢ E; Fig. 1), at a depth Information about the input load was gained from
of about 12 m on 6 March (A), 10 May (B), 29 August chemical and microbiological analysis of sewage samples
(C), and 23 October 2000 (D). The chosen dates reflect that were collected at the outflow of the treatment basin
mixed and stratified water columns and different growing and at the fish-cannery pre-treatment plant in the morn-
seasons of the phytoplankton. ing hours before each survey.
Samples were taken at six depths (0.3, 1, 3, 5, 7 m and
above the bottom) during three surveys, whereas on 10
Analyses
May a different sampling scheme was applied. On that
date the sea water was sampled at 10 depths where small- Concentrations of inorganic nutrients (nitrite, nitrate,
scale wiggles were detected on salinity profiles (see ammonia, phosphate, silicate) from unfiltered seawater
Fig. 2), and where turbulent effluent was expected. Sea- and filtered sewage samples were measured using standard
water samples were collected using 5-l Niskin bottles and colorimetric procedures (Grasshoff et al. 1999). Total
were kept in a dark and cold place prior to analysis in nitrogen (TN) and total phosphorus (TP) were analysed
the laboratory. in unfiltered samples (Grasshoff et al. 1999). Sewage sam-
Vertical profiles of temperature, salinity, density and ples were filtered through glass-fibre filters (Whatman
oxygen were obtained using a fine-scale multiparameter GF ⁄ F).
CTD (conductivity, temperature, depth) probe with Dissolved oxygen (O2) was determined following the
vertical resolution of about 2.5 cm for a conventional modified standard Winkler method (Grasshoff et al.

March S σT (kg m–3) O2 (ml l–1)

May S σT (kg m–3) O2 (ml l–1)
37.4 37.8 38.2 29.0 29.4 29.8 6.0 6.5 7.0 33.0 35.0 37.0 23.0 25.0 27.0 5.0 6.0 7.0
0 0
A B
2 2
Depth (m)

4 4

6 6

8 8

10 10

12 12

8.3 8.5 8.7 14.0 16.0 18.0 20.0

T (oC) T (oC)

August S σT (kg m–3) O2 (ml l–1) October S σT (kg m–3) O2 (ml l–1)
37.4 37.8 38.2 26.0 26.5 27.0 4.0 4.5 5.0 37.4 37.8 38.2 26.0 27.0 28.0 4.0 4.5 5.0
0 0
C D
2 2

4 4
Depth (m)

6 6
T
S
8 8

10 10

12 12

21.4 21.6 21.8 22.0 19.0 19.2 19.4

T (oC) T (oC)

Fig. 2. Vertical profiles of temperature, salinity, density (rT) and dissolved oxygen recorded at station IZ06 during four surveys in the year 2000.

Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 485
Case study of sewage discharge Mozetič, Malačič & Turk

1999). The same method was applied for determining the grams of pollution hot spots along the coast and of the
biochemical oxygen demand (BOD5) in water samples quality of coastal sea at stations IZ06 and 000F, respec-
after they were incubated for 5 days in the dark at room tively. Overlapping sampling dates in the 3-year period
temperature. were picked out for the comparison.
Concentrations of total suspended matter (TSM), par-
ticulate organic carbon (POC) and particulate nitrogen
(PN) in sewage and seawater samples were analysed on Results
Whatman GF ⁄ F glass-fibre filters, which were pre-combu-
Sewage composition
sted for 3 h at 480 C to eliminate organic contaminants.
Samples of sewage liquid (up to 100 ml) and seawater During the first survey in March, samples from the fish-
(1 l) were filtered and the material collected on the filters cannery treatment basin were not taken. Concentrations
was rinsed several times with distilled water to remove of all parameters were higher in the industrial wastes of
salts. All samples were freeze-dried for 24 h. TSM was the fish-cannery, except for nitrate and nitrite (Table 1).
determined gravimetrically. The POC and PN content of Ammonia filled up almost the entire pool of DIN. More
the freeze-dried and acid-washed sample was determined than 50% of total nitrogen was organic in both types of
using a Carlo Erba model 1108 elemental analyzer wastes. The predominant form of phosphorus was inor-
(Hedges & Stern 1984). ganic phosphate (PO43)) in domestic wastes and organic
Chlorophyll a concentrations (Chl a), corrected for P compounds in the fish-cannery wastes. Higher C ⁄ N
phaeopigments, were determined fluorometrically on a ratios in the effluent of the fish-cannery (on average 37)
Turner 112 fluorimeter (Holm-Hansen et al. 1965). Sub- reflect a composition richer in fats compared with the
samples of 20 ml were filtered onto 0.22-lm Millipore domestic sewage (mean C ⁄ N = 14). FC were counted
filters, extracted in 90% acetone and measured for only in the domestic sewage and on average they were
fluorescence. 4.3 · 106 FCÆ100 ml)1.
Samples for phytoplankton analysis were preserved with Daily loads of pollutants, calculated from their concen-
neutralized formaldehyde (2% final concentration). trations and daily flow rates of both effluents (in Turk &
Phytoplankton cells were identified and counted in sub- Potočnik 2001), were roughly estimated to vary from 15.8
samples of 50 ml on an inverted microscope where 100 to 166 kg DIN, 0.7 to 24 kg PO43), 2.2 to 32 kg SiO44), 26
fields of the bottom sedimentation chamber were exam- to 1340 kg TN, 4.5 to 33 kg TP, and 256 to 1853 kg TSM.
ined at 400· magnification (Utermöhl 1958).
The number of faecal coliform bacteria (FC) was deter-
Physical properties of the water column and the overturning
mined following the recommendations of UNEP ⁄ WHO
length scale
(1994). Water samples were filtered through the 0.45-lm
pore-size sterile Millipore filters and incubated on FC- Results of four vertical CTD casts above station IZ06
agar medium at 44.5 ± 0.2 C for 24 h. Results are show the presence of a surface lens of less-saline water at
expressed as number of FCÆ100 ml)1. each survey (Fig. 2). These surface lenses were 1–1.5-m-
The trophic status of seawater was estimated by applying thick during the surveys A, C and D and the salinity
the trophic index TRIX (Vollenweider et al. 1998), taking oscillations within the lenses were in the range 0.05–0.21.
into account concentrations of Chl a, dissolved inorganic These small vertical gradients of salinity induced a weak
nitrogen (DIN) and TP, and the absolute deviation (%) pycnocline that was always present near the surface
from oxygen saturation. The index is scaled from 0 to 10. (depth of approximately 1 m), whereas the rest of the
Relationships between different parameters, measured water column was generally homogenized. Vertical pro-
at station IZ06 during four surveys, were tested using lin- files of dissolved oxygen (Fig. 2) showed no peculiarities;
ear correlation analysis and the significance was calculated the highest concentrations (up to 6.9 mlÆl)1) were mea-
with the Student’s t-test. Data, except for salinity, temper- sured during the spring period.
ature, density, lT and TRIX index, were log transformed Stratified conditions were observed only on 10 May.
to achieve normal distribution. Statistical analysis During this survey not only we measured the lowest
MANOVA was used to compare the influenced station salinity of all campaigns (33.73) but the surface layer
IZ06 and the reference station 000F (4532.42¢ N, was more perturbed and much thicker. Salinity oscilla-
1333.05¢ E; Fig. 1) using nutrients, chlorophyll a and tions were especially pronounced in the layer between 3
trophic index. To obtain a more reliable comparison with and 6 m (maximal difference 1.32). The upper mixed
a larger number for statistical analysis (n = 68 for both surface layer of fresher water with an average tempera-
stations), data from 1999 to 2001 were also considered; ture of 18.47 C was separated from deeper layers by a
these data were acquired during regular monitoring pro- sharp pycnocline at a depth of 2 m. Another, but

486 Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd
Mozetič, Malačič & Turk Case study of sewage discharge

Table 1. Composition of the effluent of the

Izola sewage system and of the fish cannery, date 6 March 10 May 29 August 23 October Mean SD
collected before each survey in the year 2000. Izola sewage system
NO2) (mgÆl)1) 0.13 0.01 0.01 0.01 0.04 0.06
NO3) (mgÆl)1) 1.28 0.04 0.01 0.07 0.35 0.62
NH4+ (mgÆl)1) 20.14 19.50 30.10 31.22 25.24 6.28
DIN (mgÆl)1) 21.55 19.54 30.10 31.29 25.62 5.94
PO43) (mgÆl)1) 4.33 4.53 2.21 3.38 3.61 1.06
SiO44) (mgÆl)1) 4.53 5.49 6.01 4.19 5.06 0.84
TN (mgÆl)1) 252.61 38.42 73.23 80.66 111.23 96.03
TP (mgÆl)1) 6.19 5.00 5.14 5.37 5.43 0.53
BOD5 (mgÆl)1) 267.21 226.34 168.50 234.55 224.15 41.08
TSM (mgÆl)1) 200.56 92.27 162.08 294.44 187.34 84.30
POC (mgÆl)1) 71.22 56.10 69.33 130.52 81.79 33.17
PN (mgÆl)1) 4.19 4.43 6.01 10.07 6.18 2.72
C⁄N 17 13 12 13 14 2
FCÆ100 ml)1 2.2E+06 3.9E+06 9.9E+06 5.3E+06 4.0E+06
Fish cannery
NO2) (mgÆl)1) 0.01 0.01 0.01 0.01 0.00
NO3) (mgÆl)1) 0.03 0.01 0.02 0.02 0.01
NH4+ (mgÆl)1) 69.99 120.70 70.75 87.15 29.06
DIN (mgÆl)1) 70.03 120.71 70.77 87.17 29.05
PO43) (mgÆl)1) 18.30 2.87 27.83 16.33 12.60
SiO44) (mgÆl)1) 9.90 15.42 11.08 12.13 2.91
TN (mgÆl)1) 114.12 334.92 254.86 234.63 111.78
TP (mgÆl)1) 20.23 63.05 66.51 49.93 25.78
BOD5 (mgÆl)1) 1460.13 3658.54 2884.56 2667.74 1115.13
TSM (mgÆl)1) 1138.00 8235.00 6677.14 5350.05 3729.98
POC (mgÆl)1) 402.84 4325.76 3805.77 2844.79 2130.71
PN (mgÆl)1) 17.92 80.37 107.44 68.58 45.91
C⁄N 22 54 35 37 16

On the morning of 6 March, effluent from the fish cannery was not taken. Mean values of the
parameters and standard deviations (SD) are shown.

200 –1 1.0
NS distance (m)

100 –3 0.8
Depth (m)

–5 0.6
0
–7 0.4
–100 –9 0.2
–200 –11
–200 –100 0 100 200 0.0
–250 –150 –50 50 150 250
E-W distance (m) E-W distance (m)
Fig. 3. Left: horizontal distribution of lT at a depth of 0.2 m on 10 May 2000 above the outfall off Izola. Right: vertical distributions of lT in E–W
direction that cuts through the central station of the near-field for the respective cruises. Small bars mark the positions where CTD casts were
made; dark straight line marks the pipe in a simplified way. [From Malačič & Mozetič (2005), their Fig. 3 (bottom part); with the permission of
ACS journals.]

weaker, pycnocline was detected at a depth of 4 m. On turning length scale it is clear that the sewage was rising
May 10 we also performed a CTD survey across the grid to the surface, where, in a layer of thickness of about
of 11 stations to estimate the extent of the near-field of 2.5 m, it was spreading horizontally, reaching an extent
the sewage plume. From the distribution of the over- of several 100 m (Fig. 3).

Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 487
Case study of sewage discharge Mozetič, Malačič & Turk

NH4+ (µmol l–1)

0 6 12 18 24 30 0 1 2 3 0 6 12 18 24 30 0 6 12 18 24 30
A B C D
lT (m)
0 2 4 6 8 10 0 1 2 3 0 2 4 6 8 10 0 2 4 6 8 10
0 0 0 0

2 2 2 2

4 4 4 4
Depth (m)

6 6 6 6 lT
FC
8 8 8 8 NH4+

10 10 10 10

12 12 12 12

0x100 1x104 2x104 0x100 1x102 2x102 0x100 1x105 2x105 3x105 0x100 3x104 6x104 9x104
–1
FC (No. 100 ml )

Fig. 4. Vertical distribution of ammonia, faecal coliform bacteria (FC) and overturning length scale (lT) at station IZ06 during four surveys in the
year 2000. A, March; B, May; C, August; D, October.

From the vertical profiles of lT which are plotted expanded down to 3 m depth (29 August). On 10 May
together with FC and ammonia (Fig. 4), commonly con- vertical oscillations of FC were more pronounced; how-
sidered indicative parameters of sewage pollution, it is ever, their highest concentration (170 FCÆ100 ml)1) was
difficult to distinguish a general pattern of lT distribution. much lower than the highest ever measured
Profiles of surveys A and D were quite similar, both hav- (2.6 · 105 FCÆ100 ml)1).
ing a distinctive maximum (8.7 and 6.8 m) at the bottom The highest concentrations of phosphate (Fig. 5) were
and the lowest value (around 0.2 m) in the upper layer. always measured in the surface layer (0.91–1.44 lmolÆl)1)
Vertical distribution of lT during survey B was just the except on 10 May, when the maximum was observed at
opposite: a maximum (1.3 m) was found at 2 m depth the bottom (1.19 lmolÆl)1). An increase of bottom con-
and was the lowest of all maxima recorded. The most centration was also observed on 6 March. Peaks of TP
variable distribution of lT was that of survey C, with sev- (data not shown) coincided with peaks of phosphate,
eral local maxima along the vertical and the minimum at which indicates that at the highest concentrations the
the bottom. majority of phosphorus compounds were in the inorganic
form. The vast majority of total nitrogen was, on the con-
trary, in the organic form (on average 80%; data not
Faecal pollution and nutrients
shown), whereas nitrate was on average the main inor-
The highest concentrations of ammonia (Fig. 4) were ganic form (4.05 ± 1.77 lmolÆl)1), though it never
generally measured in the surface layer (15.04– reached instantaneous concentrations as high as ammonia
27.03 lmolÆl)1) and decreased sharply towards the bot- (Fig. 5). Concentrations of nitrate varied from 1.25 to
tom to reach values (<0.1–1.55 lmolÆl)1) typical for the 7.85 lmolÆl)1.
undisturbed marine environment. The exception was the Concentrations of silicate (data not shown) were in the
sampling on 10 May, when the peak concentration of range of those regularly measured in the adjacent coastal
ammonia (2.03 lmolÆl)1) was measured in the layer waters (0.86–11.12 lmolÆl)1).
above the bottom; this sampling, however, differed from The cross-correlation analysis revealed several correla-
the others by being the one with the lowest values of both tions at a level of statistical significance of 5% (Table 2).
ammonia and FC. The distribution of FC along the water One of the highest and most statistically significant corre-
column (Fig. 4) almost followed that of ammonia. Peaks lation was found between FC and ammonia (r = 0.78),
of FC were recorded in the surface layer and, similarly to followed by correlation coefficients between FC and TP
ammonia, bacteria rapidly decreased in the layers below (r = 0.71), PO43) (r = 0.67), DIN (r = 0.66) and NO2)
the surface. Occasionally the surface sewage plume (r = 0.52). In 17 pairs of quantities with r ‡ 0.65, TP and

488 Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd
Mozetič, Malačič & Turk Case study of sewage discharge

PO43– (µmol l–1)

A 0.0 1.0 2.0 B 0.0 1.0 2.0 C 0.0 1.0 2.0 D 0.0 1.0 2.0
0 0 0 0

2 2 2 2

4 4 4 4
Depth (m)

6 6 6 6

8 8 8 8

10 10 10 10
PO43–

12 12 12 NO3– 12

0.0 3.0 6.0 9.0 0.0 3.0 6.0 9.0 0.0 3.0 6.0 9.0 0.0 3.0 6.0 9.0
NO3– (µmol l–1)

Fig. 5. Vertical distribution of phosphate and nitrate at station IZ06 during four surveys in the year 2000. A, March; B, May; C, August; D, October.

Table 2. Top part: Coefficients of cross-correlation analysis performed on a dataset of physical, chemical, microbiological and biological parame-
ters (n = 530), which were measured in the seawater at station IZ06 during four surveys in the year 2000. Bottom part: pairs of quantities with
the correlation coefficient ‡ 0.65, in decreasing order.

temperature salinity rT lT FC PO43) TP NO2) NO3) NH4+ DIN TN SiO44) Chl a TSM Phyto TRIX

temperature
salinity )0.13
rT )0.73 0.76
lT 0.10 0.61 0.32
FC 0.35 0.52 0.10 0.08
PO43) 0.03 0.68 0.41 0.30 0.67
TP 0.14 0.44 0.19 0.07 0.71 0.87
NO2) 0.48 0.53 0.04 0.53 0.52 0.65 0.65
NO3) )0.06 )0.34 )0.20 )0.09 )0.13 )0.25 )0.08 0.09
NH4+ 0.51 0.18 )0.21 )0.19 0.78 0.51 0.66 0.45 0.04
DIN 0.26 0.08 )0.12 )0.12 0.66 0.43 0.65 0.49 0.52 0.79
TN )0.57 0.61 0.76 0.20 0.40 0.55 0.47 0.23 0.25 0.16 0.46
SiO44) 0.85 )0.32 )0.76 0.04 0.15 )0.02 0.20 0.48 0.13 0.35 0.25 )0.57
Chl a 0.05 )0.14 )0.15 0.06 )0.17 0.17 0.15 0.37 0.53 0.12 0.27 0.13 0.27
TSM 0.19 0.43 0.14 0.35 0.38 0.66 0.54 0.49 )0.58 0.17 )0.05 )0.01 0.12 )0.06
Phyto )0.18 )0.41 )0.14 )0.45 )0.40 )0.07 )0.05 )0.21 0.17 0.06 0.00 )0.10 0.00 0.49 )0.18
TRIX 0.41 )0.17 )0.38 )0.21 0.34 0.41 0.60 0.55 0.39 0.65 0.74 0.09 0.57 0.64 0.09 0.37

order 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
4) ) )
parameter 1 TP SiO4 DIN NH4+ rT TN TRIX TP PO43) PO43) DIN TSM NH4+ NO2 NO2 DIN TRIX
parameter 2 PO43) Temp. NH4+ FC Sal. rT DIN FC Sal. FC FC PO43) TP PO43) TP TP NH4+
r 0.87 0.85 0.79 0.78 0.76 0.76 0.74 0.71 0.68 0.67 0.66 0.66 0.66 0.65 0.65 0.65 0.65

All data, except for temperature, salinity, rT, lT and the TRIX index were log-transformed. Values in bold are significant at P < 0.05 level.

PO43) are five times involved and FC, DIN and NH4+
Phytoplankton assemblage
four times. Conversely, silicate and nitrate did not corre-
late with FC. Nitrate was rather highly correlated with Phytoplankton biomass varied from 0.62 to 2.96 lg Chl
Chl a (r = 0.53) and inversely correlated with suspended aÆl)1 and it was more or less uniformly distributed
matter (r = )0.58). throughout the water column (Fig. 6). Nevertheless, the

Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 489
Case study of sewage discharge Mozetič, Malačič & Turk

Chl a (µg l–1)

A 0.0 1.0 2.0 3.0 B 0.0 1.0 2.0 3.0 C 0.0 1.0 2.0 3.0 D 0.0 1.0 2.0 3.0
0 0 0 0

2 2 2 Chl a
2
Phyto.
4 4 4 4
Depth (m)

6 6 6 6

8 8 8 8

10 10 10 10

12 12 12 12

0x100 1x106 2x106 0x100 1x106 2x106 0x100 1x106 2x106 0x100 1x106 2x106
Phytoplankton (cells l–1)

Fig. 6. Vertical distribution of phytoplankton biomass using chlorophyll a, and of phytoplankton abundance at station IZ06 during four surveys in
the year 2000. A, March; B, May; C, August; D, October.

highest concentrations of Chl a were always measured in heterogeneous group of nanoflagellates we were able to
the bottom layer (1.00–2.92 lgÆl)1) except on 10 May. identify organisms belonging to cryptophytes, chryso-
During the sampling the pronounced vertical oscillations phytes, euglenophytes and green algae (chlorophytes and
of Chl a concentrations and a surface maximum prasinophytes), which at times became important. On 10
(2.96 lgÆl)1) were observed. May, cryptophytes were more abundant than diatoms (up
Phytoplankton abundance displayed a similar pattern to 5 · 105 cellsÆl)1), representing on average 24% of total
when compared to phytoplankton biomass (Fig. 6), which abundance. During other surveys their contribution was
is also indicated by the statistically significant correlation much lower. Another important group during the May
between the two parameters (r = 0.49 between ‘Phyto’ survey was that of green algae.
and ‘Chl a’ in Table 2). Total abundance varied from
4.6 · 105 cellsÆl)1 (survey C) to 2.5 · 106 cellsÆl)1 (survey
Trophic status
B) and the largest variations along the vertical were again
observed during the May survey (B), when high values The trophic status of the coastal area was determined by
> 106 cellsÆl)1 were regularly counted. applying the trophic index TRIX (Table 3). Its values var-
The most abundant groups were autotrophic nanofla- ied from 3.48 (survey B, 5 m depth) to 6.16 (survey D,
gellates and diatoms, which on average represented 52% surface layer), thus encompassing very different trophic
and 45% of the total abundance, respectively. In the conditions. High values of TRIX, which indicate less
March and October surveys, diatoms dominated the com- favourable trophic conditions, were generally found in the
munity (55–80% of total abundance) reaching densities surface layer and only once at the bottom (survey B).
up to 1.3 · 106 cellsÆl)1 on 23 October. The most abun- TRIX values were also correlated to the measured param-
dant diatoms in the March survey were Skeletonema costa- eters and, as expected, significant correlations were found
tum s. l., Pseudo-nitzschia calliantha and Chaetoceros spp., with those parameters (e.g. TP, DIN, Chl a) which define
whereas in October various species of Chaetoceros domi- this index. However, the correlation cannot be very high
nated the community, followed by Leptocylindrus danicus, due to the non-linear dependence of TRIX on these
Cerataulina pelagica and Cylindrotheca closterium. In May quantities.
there was clear evidence of a nanoflagellate bloom that
developed throughout the water column with maximal
abundances in the bottom layer (2.2 · 106 cellsÆl)1). On Discussion
29 August, nanoflagellates and diatoms (dominated by
Detecting the sewage plume above the discharge
C. closterium, P. calliantha and Lauderia annulata)
reached similar densities. Dinoflagellates, coccolithophor- In our study we combined both faecal bacterial counts
ids and silicoflagellates were poorly represented (on aver- and in situ measurements of the physical properties of the
age around 1% of total abundance). In the taxonomically water (temperature, salinity, density and its derived

490 Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd
Mozetič, Malačič & Turk Case study of sewage discharge

Table 3. Trophic status of seawater at station IZ06 during four sur- It is likely that in such complex near-coast areas other
veys in the year 2000, described by values of the TRIX index. methods for tracking the sewage would be more appro-
TRIX priate (e.g. fluorescent dyes, radioactive isotopes or sew-
age-derived water-soluble markers) (Smith-Evans &
depth (m) 6 March 10 May 29 August 23 October Dawes 1996; Managaki et al. 2006).
0.3 4.82 5.13 4.90 6.16
1.0 4.07 5.36
Water quality and eutrophication effect
1.5 4.97
3.0 3.62 4.81 4.83 5.15 Variable hydrological conditions, mostly an unstable
3.5 4.63 water column and irregular emptying of the sewage-col-
4.0 4.28
lection basin, also blur another effect of sewage discharge,
4.5 4.24
5.0 3.48 4.35 4.12 5.05
i.e. the extent of eutrophication, in the surrounding sea-
5.5 4.75 water.
6.5 4.42 As the concentrations of BOD5, TSM, TP and NH4+ in
7.0 4.00 4.63 4.76 the effluent (Table 1) were always above national regula-
8.0 4.00 tion limits (25 mg BOD5Æl)1, 80 mg TSMÆl)1, 2.0 mg
10.5 4.38 5.25 3.92 TPÆl)1 and 10 mg NH4+Æl)1) (Official Gazette of Republic
12.0 4.82
of Slovenia 1996) we expected that the poor quality of
(±SD) 4.06 (0.49) 4.68 (0.36) 4.40 (0.44) 5.22 (0.51)
the effluent would also be reflected in the deteriorated
Mean values of the parameter and standard deviations (SD) are water quality of the coastal sea. Abundance of FC was,
shown. indeed, elevated especially in the surface layer (up to
2.6 · 105 FCÆ100 ml)1) and 60% of all measurements of
FC have exceeded the national regulation limit for bath-
parameter lT). Previous studies of the spatial distribution ing waters (100 FCÆ100 ml)1) (Official Gazette of Repub-
of lT in the Gulf of Trieste (Malačič 2005; Malačič & lic of Slovenia 1988). Although this regulation cannot be
Mozetič 2005) showed that lT could serve as a quick indi- applied to water bodies which are not designed for recre-
cator of turbulent alien water, although several constraints ational activity, it nevertheless gives an accurate estimate
to this approach have been pointed out. The turbulence of water quality. However, the nearest beach is less than
of alien sewage water is far from being the only one 300 m southwest of the outfall; therefore there is a risk
detected by lT as the ambient fluid already has a layered that the sewage plume could occasionally reach it (see
structure of turbulence generated by physical phenomena Fig. 3).
(e.g. internal waves, their breaking, wind-induced turbu- A comparison made between station IZ06 and the ref-
lence). erence station 000F (see Fig. 1, Table 4) showed that both
It is likely that these natural ‘disturbances’ masked the stations are statistically significantly different over the 3-
turbulence of the sewage water during our surveys as the year period at the significance level of 5% (Table 5). All
lT approach did not produce satisfactory results. There parameters were on average higher at station IZ06 but
was no meaningful visual correlation between lT and FC those that contributed the most to this difference were
(Table 2) which could confirm that the presence of tur- TP, PO43), NH4+ and DIN. Nitrate was on average 1.2
bulent alien water was due to sewage discharge. Varia- times higher at station IZ06, but the maximal value was
tions of lT, even in the range of a few metres, were not rather lower compared with the reference station
detected on the profiles of FC. Yet in the situation of the (Table 4). This indicates that augmented nitrate concen-
most pronounced stratification, during survey B, when it trations often occur at station IZ06 but are not related to
is less likely that the ambient turbulence could mask the sewage discharge, as suggested by the correlation coeffi-
turbulence of the sewage water, vertical profiles of lT and cient in Table 2. The average values of the TRIX index
FC only roughly matched. show a slight difference in trophic status, being high
Other constraints in using lT above the Izola submarine (3.59) at the reference station and good (4.71) at station
outfall are related to the steadiness of the discharge and IZ06 (Table 4). The persistently poorest conditions indi-
to the sloping bed. The discharge at Izola emerges from a cated by the average moderate trophic status
single orifice and is driven by pumps, intermittently putt- (5.67 ± 0.89) are found only in the surface layer of sta-
ing bursts of sewage into the sea and, secondly, the outfall tion IZ06 where positively buoyant sewage plume usually
is on a seabed with a slope (around 10 ⁄ 200 = 0.05), emerges. Below this layer, nutrient concentrations are
which can create trapped and reflected internal waves in similar to ‘unpolluted’ values and follow the seasonal dis-
stratified fluids. tribution. Only on one occasion, 10 May, did the back-

Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 491
Case study of sewage discharge Mozetič, Malačič & Turk

Table 4. Comparison of the mean concentrations of nutrients and Chl a measured at station IZ06 and at reference station 000F in the period
1999–2001.

NO3) NO2) NH4+ PO43) SiO44) DIN TN TP Chl a

(lmolÆl)1) (lmolÆl)1) (lmolÆl)1) (lmolÆl)1) (lmolÆl)1) (lmolÆl)1) (lmolÆl)1) (lmolÆl)1) (lgÆl)1) TRIX

station IZ06
Mean 3.85 0.14 2.06 0.15 7.24 7.72 45.18 0.45 0.98 4.71
Mean+1*SD 9.83 0.40 7.80 0.62 12.57 16.27 88.47 1.07 1.81 5.45
Mean)1*SD 1.51 0.05 0.54 0.04 4.17 3.67 23.07 0.19 0.54 3.97
Min 0.01 0.02 0.07 0.01 0.86 2.02 3.06 0.03 0.22 2.97
Max 12.77 0.84 70.37 11.45 17.96 70.58 287.28 12.68 2.96 7.02
station 000F
Mean 3.18 0.13 0.93 0.04 7.10 5.07 41.22 0.18 0.75 3.59
Mean+1*SD 8.69 0.46 2.00 0.09 11.44 10.19 62.22 0.38 1.40 4.48
Mean)1*SD 1.16 0.04 0.43 0.02 4.40 2.52 27.30 0.09 0.40 2.71
Min 0.35 0.01 0.09 0.01 2.13 1.70 12.35 0.01 0.11 1.01
Max 20.18 1.21 3.70 0.27 20.01 21.10 79.48 0.48 2.83 5.38

Mean values of nutrients and Chl a were calculated as geometric means, whereas those of TRIX as arithmetic means. Minimal (Min) and maximal
(Max) values and 1 standard deviation (1*SD) are shown.

Table 5. Results of the multivariate analysis of variance (MANOVA) sewage discharge in an Australian estuary. No occurrences
of nutrients, Chl a and TRIX measured in the seawater at station IZ06 of this sort were ever observed in our case.
and reference station 000F in the period 1999–2001.
In addition to nitrogen, whatever its form, phosphorus
parameter F P is also one of the major nutrients for phytoplankton
growth. In many coastal waters more than half of the
NO3) 1.236 0.2685
phosphate uptake can be due to bacteria (Turk et al.
NO2) 0.309 0.5792
NH4+ 16.599 0.0001
1992; Turk & Hagström 1997). In our study the majority
PO43) 39.362 0.0000 of phosphorus was in the inorganic form, especially at
SiO44) 0.044 0.8335 high concentrations of TP, and due to the P-depleted
DIN 10.823 0.0013 organic substrate, bacteria might have prevailed over phy-
TN 0.841 0.3608 toplankton for phosphate. This was previously observed
TP 40.636 0.0000 in several enrichment experiments carried out in the Gulf
Chl a 6.052 0.0152
of Trieste (Malej et al. 2003) when the maximal rate of
TRIX 52.672 0.0000
bacterial production was measured on the second or third
Values in bold are significant at P < 0.05 level (Wilks’ k = 0.422, day of experiments in enclosures with P addition. The
df1 = 10, df2 = 117). same was also recorded in an enclosure enriched with
sewage from the Izola treatment basin (Authors’ unpub-
ground conditions prevail throughout the whole water lished data).
column (thermo-haline stratification; see Fig. 2). Little biological response was observed in other con-
The observed discrepancy between high nutrient con- fined areas such as estuaries which receive high concen-
centrations and the minor biological response is even trations of sewage-originated phosphate and ammonia
more accentuated if we consider that ammonia is, on (Aslan-Yilmaz et al. 2004; Pavlidou et al. 2004; Mallin
average, the preferred nitrogen form for phytoplankton et al. 2005; O’Higgins & Wilson 2005; Garcia-Barcina
(Dortch 1990; Dortch et al. 1991; Watson & McCauley et al. 2006). O’Higgins & Wilson (2005) suggested that
2005) and that nitrate uptake is often inhibited in the due to physical and chemical processes (accumulation
presence of ammonia at concentrations > 1 lmolÆl)1 and remineralization) organic matter remains entrapped
(Dortch et al. 1991). This preference for ammonia was in the estuary. In the case of short water retention time,
recently linked to some species of the genus Alexandrium phytoplankton growth is not sustained in spite of eutro-
(Cannon 1990; Sorokin et al. 1996; Takeuchi & Yoshida phic levels of nutrients (Demir & Kirkagac 2005).
1999; Maguer et al. 2007), which might have important
ecological and human health-related implications, these
Conclusions
taxa being Harmful Algal Bloom (HAB) forming organ-
isms. Cannon (1990) also observed Alexandrium minutum The intermittent discharge of effluent and the discharge
blooms in a strongly nutrient-enriched area close to of sewage onto a sloping bed near the shoreline do not

492 Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd
Mozetič, Malačič & Turk Case study of sewage discharge

allow an accurate tracking of the sewage plume with the B., Edler L., Anderson D.M. (Eds), Toxic Marine Phytoplank-
otherwise successfully applied approach of the overturn- ton. Elsevier, New York: 110–115.
ing length scale. The effect of intermittency is also Chisholm J.R.M., Fernex F.E., Mathieu D., Jaubert J.M. (1997)
reflected in the minimal effect of eutrophication in the Wastewater discharge, seagrass decline and algal prolifera-
case of phytoplankton. High amounts of nutrients proba- tion on the Cote d’Azur. Marine Pollution Bulletin, 34, 78–
bly dilute rapidly into the surrounding media under 84.
unsteady conditions, not allowing a sustainable growth of Demir N., Kirkagac M.U. (2005) Plankton composition and
phytoplankton. Deteriorated average conditions as indi- water quality in a pond of spring origin in Turkey. Limnol-
ogy, 6, 189–194.
cated by moderate trophic status are observed only in the
Dortch Q. (1990) The interaction between ammonium and
surface layer.
nitrate uptake in phytoplankton. Journal of Experimental
The major negative influence of improperly treated
Marine Biology and Ecology, 61, 243–264.
wastewaters is demonstrated mostly by the microbiologi-
Dortch Q., Thompson P.A., Harrison P.J. (1991) Short-term
cal quality of the coastal sea. High concentrations of pol-
interactions between nitrate and ammonium uptake in
lution-related parameters in the effluent, in many cases Thalassiosira pseudonana: effect of preconditioning nitro-
well above national regulation limits, are evidence of that. gen source and growth rate. Marine Biology, 110, 183–
There is a constant threat to human health during the 193.
bathing season and to the integrity of seafood (i.e. Faganeli J. (1982) Nutrient dynamics in seawater column in
angling) due to the spreading of the sewage plume the vicinity of Piran submarine sewage outfall (North Adri-
towards the beach and its presence in the surface layer. atic). Marine Pollution Bulletin, 13, 61–66.
To solve the problem of poor sewage treatment, the Garcia-Barcina J.M., Gonzalez-Oreja J.A., De la Sota A. (2006)
municipality of Izola will, in the near future, send their Assessing the improvement of the Bilbao estuary water qual-
sewage through a recently built pipeline system to a new ity in response to pollution abatement measures. Water
sewage treatment plant, which is designed for tertiary Research, 40, 951–960.
wastewater treatment. Grasshoff K., Ehrhardt M., Kremling K. (1999) Methods of Sea-
water Analysis, 3rd completely revised and extended edition.
Wiley-VCH, Weinheim: 600 pp.
Acknowledgements Hedges J.I., Stern J.H. (1984) Carbon and nitrogen determina-
The project was co-financed by the Ministry of Higher tions in carbonate-containing solids. Limnology and Ocean-
Education, Science and Technology (L1-1570-0105-99) ography, 29, 657–663.
and the Municipality of Izola. The authors wish to thank Holm-Hansen O., Lorenzen C.J., Holmes R.W., Strickland
Ernest Požar of the public enterprise Komunala Izola for J.D.H. (1965) Fluorometric determination of chlorophyll.
Journal de Conseil pour l’Exploration de la Mer, 30, 3–15.
valuable information regarding the construction and
Maguer J.F., L’Helguen S., Madec C., Labry C., Le Corre P.
operation of the sewage system of the community of Izo-
(2007) Nitrogen uptake and assimilation kinetics in Alex-
la. The authors are also grateful to Dr Nives Kovač for
andrium minutum (Dinophyceae): effect of N-limited
some chemical analyses and to Milijan Šiško for statistical
growth rate on nitrogen and ammonium interactions. Jour-
analysis.
nal of Phycology, 43, 295–303.
Malačič V. (2005) Initial spread of an effluent and the over-
References turning length scale near an underwater source in the north-
ern Adriatic. Journal of Marine Systems, 55, 47–66.
Adingra A.A., Arfi R. (1998) Organic and bacterial pollution Malačič V., Mozetič P. (2005) Modeling of the spread of an
in the Ebrié Lagoon, Côte d’Ivoire. Marine Pollution Bulle- effluent and the overturning lenghth scale near an underwa-
tin, 36, 689–695. ter source in the northern Adriatic. Journal of Chemical
Aslan-Yilmaz A., Okus E., Ovez S. (2004) Bacteriological indi- Information and Modeling, 45, 1491–1501.
cators of anthropogenic impact prior to and during the Malačič V., Petelin B. (2001) Gulf of Trieste. In: Cushman-
recovery of water quality in an extremely polluted estuary, Roisin B., Gačić M., Poulain P.-M., Artegiani A. (Eds), Phys-
Golden Horn, Turkey. Marine Pollution Bulletin, 49, 951– ical Oceanography of the Adriatic Sea: Past, Present and
958. Future. Kluwer Academic Press, Dordrecht: 167–181.
Braga E.S., Bonetti C.V.D.H., Burone L., Bonetti Filho J. Malej A., Mozetič P., Turk V., Terzić S., Ahel M., Cauwet G.
(2000) Eutrophication and bacterial pollution caused by (2003) Changes in particulate and dissolved organic matter
industrial and domestic wastes at the Baixada Santista estua- in nutrient-enriched enclosures from an area influenced by
rine system – Brazil. Marine Pollution Bulletin, 40, 165–173. mucilage: the Northern Adriatic Sea. Journal of Plankton
Cannon J.A. (1990) Development and dispersal of red tides in Research, 25, 949–966.
the Port River, South Australia. In: Graneli E., Sundstrom

Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd 493
Case study of sewage discharge Mozetič, Malačič & Turk

Mallin M.A., McIver M.R., Wells H.A., Parsons D.C., Johnson of a Mediterranean sewage outfall after setting up a treat-
V.L. (2005) Reversal of eutrophication following sewage ment plant. Marine Pollution Bulletin, 42, 59–70.
treatment upgrades in the New River Estuary, North Caro- Sorokin Y.I., Sorokin P.Y., Ravagnan G. (1996) On an
lina. Estuaries, 28, 750–760. extremely dense bloom of the dinoflagellate Alexandrium
Managaki S., Takada H., Kim D.-M., Horiguchi T., Shiraishi tamarense in lagoons of the Po River Delta: impact on the
H. (2006) Three-dimensional distributions of sewage mark- environment. Journal of Sea Research, 35, 251–255.
ers in Tokyo Bay water – fluorescent whitening agents Takeuchi T., Yoshida Y. (1999) Relationship between the
(FWAs). Marine Pollution Bulletin, 52, 281–292. blooms of Alexandrium catenella and the quality of
Mozetič P., Fonda Umani S., Cataletto B., Malej A. (1998) Sea- meteorological factors. Nippon Suisan Gakkaishi, 65, 826–
sonal and inter-annual plankton variability in the Gulf of 830.
Trieste (Northern Adriatic). ICES Journal of Marine Sciences, Theodorou A.J. (1992) Ecological consequences of untreated
55, 711–722. waste-water discharge in the Saronikos Gulf, Greece. Water
Mozetič P., Malačič V., Turk V. (1999) Ecological characteris- Science and Technology, 25, 115–124.
tics of seawater influenced by sewage outfall. Annales Series Torres E.S., del Rio J.G. (1995) Spatial variations of phyto-
Historia Naturalis, 9, 177–190. plankton community structure in a highly eutrophicated
O’Higgins T.G., Wilson J.G. (2005) Impact of the river Liffey coast of the Western Mediterranean Sea. Water Science and
discharge on nutrient and chlorophyll concentrations in the Technology, 32, 313–322.
Liffey estuary and Dublin Bay (Irish Sea). Estuarine, Coastal Turk V., Hagström Å. (1997) Seasonal distribution of nanofla-
and Shelf Science, 64, 323–334. gellates and bacterivory in the Gulf of Trieste (Northern
Official Gazette of Republic of Slovenia (1988) Regulations on Adriatic). Periodicum Biologorum, 99, 205–208.
hygienic requirements for bathing waters. No. 9 (11.III.1988). Turk V., Potočnik B. (2001) Pollution hot spots and sensitive
Ljubljana: 660–661. areas along the Slovenian coast. Annales Series Historia Nat-
Official Gazette of Republic of Slovenia (1996) Decree on emis- uralis, 11, 239–252.
sion of substances and heat during drainage of waste waters Turk V., Rehnstam A.S., Lundberg E., Hagström Å. (1992)
from pollution sources. No. 35 (5.VII.1996). Ljubljana: 2953– Release of bacterial DNA by marine nanoflagellates, an
3000. intermediate step in phosphorus regeneration. Applied and
Olivotti R., Faganeli J., Malej A. (1986) Eutrophication of Environmental Microbiology, 58, 3744–3750.
coastal waters, Gulf of Trieste. Water Science and Technology, UNEP ⁄ WHO (1994) Guidelines for Health-Related Monitoring
18, 303–316. of Coastal Recreational and Shellfish Areas. Bacterial Indicator
Pavlidou A., Kontoyiannis H., Psyllidou-Giouranovits R. Organisms. UNEP, Copenhagen.
(2004) Trophic conditions and stoichiometric nutrient bal- UNEP ⁄ WHO (2007) Assessment of the State of Microbiological
ance in the inner Saronikos Gulf (Central Aegean Sea) Pollution of the Mediterranean Sea. UNEP(DEPI) ⁄ MED
affected by the Psittalia sewage outfall. Fresenius Environ- WG.316 ⁄ Inf. 5, Athens.
mental Bulletin, 13, 1509–1514. Utermöhl H. (1958) Zur Vervollkommung der quantitativen
Savage C., Elmgren R., Larsson U. (2002) Effects of sewage- Phytoplankton-Methodik. Mitteilungen. Internationale Verei-
derived nutrients on an estuarine macrobenthic community. ningung fur Theoretische und Angewandte Limnologie, 9,
Marine Ecology Progress Series, 243, 67–82. 1–38.
da Silva J.F., Duck R.W., Hopkins T.S., Anderson J.M. (2002) Vollenweider R.A., Giovanardi F., Montanari G., Rinaldi A.
Nearshore circulation revealed by wastewater discharge from (1998) Characterization of the trophic conditions of
a submarine outfall, Aveiro Coast, Portugal. Hydrology and marine coastal waters, with special references to the
Earth System Sciences, 6, 938–988. NW Adriatic Sea: proposal for a trophic scale turbidity
Smith-Evans M., Dawes A. (1996) Early experiences in moni- and generalized water quality index. Environmetrics, 9, 329–
toring the effects of Hong Kong’s new generation of sewage 357.
outfalls on the marine environment. Marine Pollution Bulle- Watson S.B., McCauley E. (2005) Ammonia and nitrate:
tin, 33, 317–327. growth or inhibition? Large- and small-scale patterns in
Soltan D., Verlaque M., Boudouresque C.F., Francour P. chrysophytes and the potential role of inorganic nitrogen.
(2001) Changes in macroalgal communities in the vicinity Nova Hedwigia, Suppl. 128, 231–247.

494 Marine Ecology 29 (2008) 483–494 ª 2008 The Authors. Journal compilation ª 2008 Blackwell Publishing Ltd