Page 4 - m313p261

264 Mar Ecol Prog Ser 313: 261–270, 2006

Table 2. Measurements of δ13C and δ15N in particulate and sedimentary organic matter (POM and SOM, respectively) at each loca-
tion in each distance category. CAST: Gulf of Castellammare (isotopic data from this site partially derived from Sarà et al.

2004); MEN: Menfi Bay; CORL: Capo D’Orlando Bay; EI: Egadi Islands (control sites)

Hydrodynamic Distance POM δ13C SOM δ13C POM δ15N SOM δ15N
level
(m) Mean (±SE) Min. Max. Mean (±SE) Min. Max. Mean (±SE) Min. Max. Mean (±SE) Min. Max.

HYDRO 1 (CAST) 0 –22.3 (0.5) –22.9 –21.9 –22.0 (0.1) –22.1 –21.9 5.2 (0.2) 5.0 5.4 4.5 (0.1) 4.4 4.6
–24.0 (0.4) –24.5 –23.6 –22.1 (0.1) –22.2 –22.0 2.9 (0.2) 2.6 3.0 3.1 (0.1) 3.0 3.2
500 –23.0 (0.1) –23.1 –22.9 –22.2 (0.3) –22.4 –21.8 3.5 (0.1) 3.4 3.6 2.3 (0.1) 2.2 2.4

1000

HYDRO 2 (MEN) 0 –22.1 (0.2) –22.3 –21.8 –20.2 (0.3) –20.5 –19.9 2.6 (0.1) 2.5 2.6 4.2 (0.1) 4.1 4.2
–22.0 (0.2) –22.2 –21.8 –21.3 (0.4) –21.9 –21.0 3.9 (0.1) 3.7 4.0 4.5 (0.1) 4.3 4.6
500 –22.4 (0.5) –22.9 –21.9 –23.6 (0.2) –23.8 –23.4 3.3 (0.1) 3.2 3.4 4.1 (0.1) 3.9 4.2

1000

HYDRO 3 (CORL) 0 –22.0 (0.1) –22.1 –21.8 –21.3 (0.2) –21.4 –21.0 1.8 (0.1) 1.6 1.9 2.1 (0.2) 1.8 2.3
500 –21.5 (0.1) –21.6 –21.3 –23.7 (0.2) –23.8 –23.4 4.2 (0.1) 4.1 4.4 3.4 (0.2) 3.2 3.6
–21.9 (0.2) –22.1 –21.8 –22.5 (0.1) –22.6 –22.4 4.6 (0.1) 4.5 4.6 4.4 (0.2) 4.1 4.6
1000
–22.7 (0.2) –23.1 –22.2 –18.7 (0.2) –19.3 –18.4 5.9 (0.3) 5.1 6.6 4.2 (0.2) 3.9 4.7
Control (EI) 0 –21.7 (0.6) –23.2 –20.5 –18.3 (0.0) –18.4 –18.2 5.0 (0.2) 4.6 5.4 4.4 (0.1) 4.1 4.6
500 –22.5 (0.1) –22.5 –22.3 –17.9 (0.5) –18.7 –16.8 5.3 (0.1) 5.1 5.6 4.6 (0.0) 4.5 4.6
1000

C and N in POM and SOM at variable distances enrichment gradient with increasing distance from the
from fish farms cages to far sites. With regard to POM, the decreasing
contribution of total fish farm waste with increasing
HYDRO 1 — Gulf of Castellammare distance from the cages (pellet + ejection = 16%) to
farthest sites (pellet + ejection = 1.5%) was clearly
Partial comparisons in the Gulf of Castellammare evident. In contrast, information from partial compar-
(i.e. at current velocity of 10 to 12 cm s–1) showed sig- isons of SOM between distance categories indicated
nificant differences (ANOVA, p < 0.05) in δ13CPOM and that δ13CSOM values showed a significant depletion gra-
δ15NPOM values between the following sites: 0 and dient with increasing distance from cages (ANOVA,
500 m, 500 and 1000 m, and 0 and 1000 m (Fig. 2a). p < 0.05), whereas δ15NSOM appeared almost constant
When moving from 0 m sites (pellet + ejection = 37%) between the cages and farthest sites (Fig. 3b). The
to 1000 m sites (pellet + ejection = 13%), mixing model trend of the percent contribution of total fish farm
results showed a decreasing contribution of fish farm waste to sediments showed an inverse gradient with
waste signals to POM. SOM results were similar to respect to the water column: this contribution was
those of POM regarding nitrogen (Fig. 3a). There were lower at cages (pellet + ejection = 7.5%) and higher at
significant differences among all distance categories 1000 m sites (pellet + ejection = 38%).
(ANOVA, p < 0.05), which described a well-defined
nitrogen-depletion gradient moving away from the HYDRO 3 — Capo D’Orlando Bay
focus of emission to farther sites. In contrast, when all
distances were compared, δ13CSOM values were similar In Capo D’Orlando Bay, δ13CPOM values were similar
(ANOVA, p > 0.05). However, the contribution of total among all distance categories (ANOVA, p > 0.05),
fish farm waste to carbon decreased with increasing whereas δ15NPOM showed a clear and significant en-
distance from the centre of the fish farm (pellet + ejec- richment gradient (ANOVA, p < 0.05) with increasing
tion = 31 and 3.3% at 0 and 1000 m sites, respectively). distance from the cages (δ15NPOM = 1.8 ‰) toward more
distant sites (δ15NPOM = 4.6 ‰) (Fig. 2c). Thus, the total
HYDRO 2 — Menfi Bay contribution of fish farm waste to POM increased sig-
nificantly with increasing distance from cages (pellet +
In Menfi Bay, δ13CPOM values did not show significant ejection < 1%) to 1000 m sites, where highest values
differences among the 3 distance categories (ANOVA, were detected (pellet + ejection = 19%). For SOM, all
p > 0.05) (Fig. 2b) but remained notably constant up to distance categories were significantly different in both
1000 m from the cages. In contrast, δ15NPOM values δ13CSOM and δ15NSOM composition (ANOVA, p < 0.05)
were significantly different among all distance cate- (Fig. 3c); values were greater at more distant sites com-
gories (ANOVA, p < 0.05), and described a nitrogen- pared to close sites, and thus depicted a well-defined