3.3. Influence of Environment and Seagrass Structure on
Epifaunal Abundance
All seagrass variables including epiphytic algal biomass were
significant in explaining variation in epifaunal abundance (PERMANOVA:
DistLM, p < 0.01, Table 3.4). Initially, distance-based
linear models were run without including leaf width due to its strong
correlation with leaf length. However, a better fit was achieved when
leaf width was included. Among the seagrass variables, leaf width
contributed the highest proportion of variation in epifaunal abundance,
followed by leaf length, density, biomass, and epiphyte biomass (Table
3.4). All seagrass variables were selected in the first model that best
explained variability in epifaunal abundance determined by lowest AIC
and highest R2 values. The second best model which
included four variables (density, leaf width, epiphyte and seagrass
biomasses), and had an AIC value < 2 levels from the first
model, also adequately explained the observed variation (Table 3.4).
Environmental variables assessed independently in a DistLM marginal
test, found exposure to contribute the highest proportion of variation
followed by chl a and turbidity (Table 3.5). Temperature,
salinity, and oxygen accounted for smaller proportions of variation
(Table 3.5). All environmental variable contributions were significant
(p < 0.001). The top two models based on AIC and
R2 parameters, that best explained variation in
epifaunal abundance selected five (all) and four (excluding turbidity)
environmental variables respectively (Table 3.5).
The final structural equation model (SEM) employed to explore the
influence of environmental variables and seagrass metrics on epifaunal
abundance revealed several noteworthy findings. The direct effects of
temperature, salinity, pH, and exposure on abundance were found to be
non-significant (chi-square = 90.66, df = 45, p = 0.35, RMSEA =
0.083; Fig. 3.4B). However, turbidity exhibited a strong negative direct
effect (-0.85), while oxygen displayed a positive direct effect (1.09).
Indirectly, temperature, pH, exposure, and oxygen had negative effects
on epifaunal abundance, whereas turbidity had a positive indirect effect
(Table 3.6). Although all environmental variables directly influenced
leaf length, this did not translate into indirect effects on abundance.
In contrast, leaf width was negatively influenced by temperature and
turbidity, and positively influenced by pH and exposure. It exerted a
positive direct effect, explaining 65% of the variation in abundance
(Fig. 3.4B, Table 3.6). Similarly, shoot density had a positive direct
effect on abundance, accounting for 65% of variation (Fig 3.4B). Shoot
densities were negatively influenced by pH, oxygen and exposure and
positively affected by turbidity (Table 3.6). No other seagrass metrics
emerged to directly influence epifaunal abundances.