Geology Reference
In-Depth Information
where
C
is CaCO
3
(g),
P
is total dry seagrass
(includes the carbonate) (g) and
P
i
is the weight
of individual seagrass groups used (weight
includes the carbonate) (g). As the quadrat size is
50
Most often log 10 transformation was found to
be appropriate. The probability level of
= 0.05
was assumed, unless transformation data failed
Levene's test for variance homogeneity in which
case a more conservative alpha value was adopted
(i.e
50 cm, then a multiplication of 4 is used to
bring the value up to 1 m
2
.
The value
C
in g m
2
is dependant on the weight
of the seagrass. As the weight of the seagrass
is measured by how much is present within a
certain area, this weight is also expressed as den-
sity. Density depends on the number of plants,
the type of plants, and the amount of carbonate
on those plants. An increase in plant density can
be attributed to either an increase in the number
of plants or an increase in the amount of carbon-
ate. Both an increase in plant weight and carbon-
ate value will increase the
C
g m
2
value. Hence
with increasing plant density it would also be
expected that there would be an increase in
C
g m
2
.
A disadvantage of the carbonate g m
2
val-
ues is that this is very subjective to the density
of the seagrass and not the amount of carbonate
that is actually found on each plant. To compare
carbonate values between localities it is necessary
to calculate the carbonate quantity independent
of the various seagrass densities from site to site.
To do this the amount of carbonate found upon a
known weight of seagrass is calculated, i.e. CaCO
3
per kg of seagrass. This is simply calculated by
= 0.01) (Underwood, 1981).
Post hoc
com-
parison of means determining where signifi cant
differences lie was carried out using Tukey's HSD
test. Non-parametric Kruskall-Wallace analysis
of variance was used where gross violations
of normal distribution and failure of homo-
geneity occurred (Underwood, 1981; Zar, 1996).
Signifi cant differences were identifi ed using the
post hoc
Tukey style, Nemenyi comparison of
means. Seagrass species distribution was analysed
by cluster analysis (JMP3.0.2) and multidimen-
sional scaling (MDS) using the Bray-Curtis
association measure for the similarity matrix
(Pcord4). Student's
t
-test was used to test differ-
ences in carbonate abundance between genera.
All averages given are means and standard errors
are provided where possible.
The method used to derive qualitative estimates
of the relative abundances from ranked data is
adapted from Saito and Atobe (1970) (cited in
English
et al
., 1997). The formula is
(3)
where
M
i
is the mid point value of each class and
f
i
is the frequency of each class (number of blades
with same class).
Estimation of the amount of carbonate derived
from each species was calculated using the abun-
dance ratio of each species, the standing stock val-
ues of calcareous epiphyte abundance from each
site on each seagrass species present.
(2)
The relationship between
C
kg
1
seagrass and sea-
grass sample size is a matter of ratio. The greater
the ratio, the lower the
C
kg
1
seagrass value and
vice versa.
Calcareous epiphyte productivity is achieved
by multiplying standing stock value by the num-
ber of crops per year. Biomass (standing stock) is
the dry weight of all biomass in each quadrant
(4)
where %
sp
i is the percentage of species '
i
' present
on seagrass species
a
and site
b
and
C
is the car-
bonate abundance (g kg
1
of seagrass) on the same
seagrass species at the same site. These calcula-
tions assume that the density (or specifi c gravity)
of each form of the coralline algae is the same.
4; seagrass biomass which
=
seagrass biomass
in g m
2
.
Factors affecting biomass and carbonate quan-
tity were analysed using Analysis of Variance
(ANOVA) with SPSS 10.0 software and JMP3.0.2.
Variance heterogeneity was checked using Leven's
test and normal distribution was tested using the
Kolmogorov-Smirnov test where the number of
samples was >20, otherwise the Shapiro-Wilk test
was used. Data were transformed where necessary,
the appropriate transformation estimated using
Taylor's power law and the ladder of powers.
RESULTS AND INTERPRETATION
Seagrass biomass
The seagrasses are variably distributed across
the region (Fig. 5).
Posidonia
and
Amphibolis
