Page 2 - DiMaidaetal2013

G. Di Maida et al. / Marine Environmental Research 87-88 (2013) 96e102
rhizomes growing on matte and sand was greater than that on rocky
substratum. The opposite response was observed at meadow level,
with higher shoots density on rock than on sand and matte. Finally,
at the community level, epiphyte cover and composition did not
differ among shoots growing on the three substrata.
Less attention has been paid instead to the effect of different
substratum typologies on growth performance. The vertical elon-
gation rate of the rhizome is closely linked to sedimentation rate as
shoots adapt their growth to different sediment deposition rates
(Boudouresque and Jeudy de Grissac, 1983; Boudouresque et al.,
1984; Jeudy de Grissac and Boudouresque, 1985), through vertical
growth plasticity (Manzanera et al., 2011; Marbà and Duarte, 1994).
Marbà and Duarte (1997) observed that where horizontal rhizomes
were buried deeper than 20 cm below the sediment surface the
vertical growth rates were higher than those recorded over sub-
strates dominated by exposed P. oceanica matte. Furthermore, to
counteract sediment deposition rate, in sheltered areas and on
carbonate sediments rhizomes tend to grow vertically at a greater
rate, than those growing in exposed areas and on siliciclastic sed-
iments, which tend instead to grow with a lower rate and to expand
laterally because of the absence of sediment deposition (De Falco
et al., 2008).
Overall, these studies elucidated different aspects of P. oceanica
attributes developing on different substratum types, although to
date very few observations (if any) were simultaneously recorded on
meadows growing on the three prevalent types of substratum (i.e.
rock, sand, and matte) in the Mediterranean Sea. Here, we aim at
filling these gaps and at assessing the magnitude and variability of
the effect of substratum typologies on P. oceanica growth and
morphology attributes. More specifically, growth performance, leaf
biometry and shoot density are examined and compared in four
P. oceanica meadows growing on these prevalent substrata along the
north-western coasts of Sicily (Italy). We focused on these P. oceanica confounder (Tomasello et al., 2007).
attributes as these are now commonly used as seagrass quality ele-
ments to assess the ecological status of coastal water bodies under
the European Union Water Framework Directive (WFD, 2000/60/EC;
Lopez y Royo et al., 2010). Since substratum typologies are not
presently considered as a source of variation on seagrass character-
istics in this context, our findings e in addition to have interesting
methodological consequences on the causative variables potentially
Fig. 1. The four locations examined along the north-western coast of Sicily.
97
affecting meadows features and their health status e can also have
interesting policy and management implications.
2. Materials and methods
2.1. Study area, sampling and laboratory analyses
The study was carried out in four locations were P. oceanica
meadows grow on three types of substratum: sand, rock and matte
(Fig. 1). For substratum differentiation we choose plants growing on
sand, plants settled on visibly exposed rock and plants growing on
visibly exposed and not buried matte. At each location sampling
and measurements were carried out on meadows settled on each
kind of substratum. Shoot density was determined by using five
randomly placed quadrats (40 Â 40 cm), and fifteen short shoots
(Hemminga and Duarte, 2000) connected to the horizontal axis
were randomly collected for biometric analysis. Depth at each
location was more or less constant, varying between 8 and 10 m. All
samplings and measurements were carried out in July 2008, except
for the Favignana location, which was examined in October 2008.
Rhizomes were analyzed by lepidochronology (Pergent, 1990),
which provides reliable estimation of seagrass growth performance
(annual vertical growth and leaf production) (Pergent-Martini
et al., 2005). This retrodating technique allows us to isolate and
date rhizome segments corresponding to a 1-year period (lep-
idochronological year). Each lepidochronological year is dated
starting from the rhizome apex (sampling year) downward and
backdating the sequence of cycles with their corresponding
rhizome segment. This reiterative procedure is performed until the
rhizome segment connected to the horizontal axis is reached,
representing the year of shoot birth. Finally, the length of each
rhizome segment is measured and the shoot age determined so it
can be taken into account in the statistical analysis as a potential
For each rhizome only the three most recent lepidochrono-
logical years were considered in the analysis since the number of
preceding years per shoot varied considerably among the three
substrata.
On leaf bundles biometric analyses according to Giraud (1977)
and Pergent-Martini et al. (2005) were carried out. These