International Journal of Marine Science 2014, Vol.4, No.61, 1-13
http://ijms.biopublisher.ca
3
The species availability map was created by
converting the presence of 3.366 data of scallop into a
grid size of 100 ×100 m
2
with ArcGIS software. The
grid conversion has resulted 721 species presence grids.
1.3 Scallop Habitat Modeling with ENFA
Habitat suitability map generated by Ecological Niche
Factor Analysis (ENFA). ENFA could produce habitat
suitability maps with the data linking to the species
availability with environmental variables (Table 1) in
order to determine the ecological niche of a species
(Hirzel et al., 2002). This technique integrated in the
BioMapper software (Hirzel, 2001). The program also
combined descriptive statistics with Geographic
Information Systems (GIS).
Table 1 Eco-geographical variable applied in the ENFA
Eco-geographical
Variables
Data Source
Data Synthesis Method
Seabed Sediments
Distribution of seabed sediment type from P3GL
The sediment type data was digitized and
converted to a raster format then transformed
to φ Wentworth (Williams et al., 2006)
Bathymetry
Points data and depth contours of bathymetry map from
Dishidros
Data interpolation with IDW method in
geostatistic-ArcGIS
Distance from Estuary
Indonesia topographic map from Bakosurtanal updated
with 2008 ASTER satellite image
Synthesis data by
buffering
method in
geostatistics-ArcGIS
Sea Surface TSS
Sea water sampling analyzed in laboratory
Extracted from Landsat 7 ETM+ satellite
image by generating the TSS Algorithm
Seabed Current Velocity Hourly wind data from BMG; Bathymetry and tidal data
from Dishidros; Coastline of 2008 Aster satellite image
SMS 8.1 Modeling
Seabed Temperature
Primary data obtained by
in-situ
analysis using CTD+
Data interpolation with IDW method in
geostatistics-ArcGIS
Seabed Salinity
Seabed Acidity (pH)
Primary data obtained by
in-situ
analysis using
pH-meter
.
Water was sampled using Nansen Bottle
Data interpolation with IDW method in
geostatistics-ArcGIS
Seabed Plankton Density Seabed planktons maintained in Nansen Bottle was
analyzed in the laboratory
Data interpolation with IDW method in
geostatistics-ArcGIS
Note: All data were converted into raster data with 100 ×100 m
2
grid size
In BioMapper, each variables value will be
transformed to Box-Cox format, thereby it normally
distributed and could be overlaid. For each thematic
map, the value of each location where scallop species
calculated produced a score appearing as several
classes in the frequency histogram. By assuming the
data distributed normally, the maximum score was
around the median and decreased on both sides (Hirzel
et al., 2002). Hereafter, class of each grid within the
study area would be determined and the value of
"partial suitability" of each thematic map was
generated based on scores from classes in the
histogram. The farther the grid from the median, the
lower the habitat suitability was.
Furthermore, global suitability maps would be
generated by calculating the weighted averages of
several partial suitability value of each thematic map,
producing a rescaled habitat suitability index value in
the isopleths methods ranging from 0-100, where a
value of 0 indicated that there were no suitable habitat
and vice versa (Hirzel, 2001). ENFA summarized all
eco-geographical variables (thematic maps) into a
number of unrelated components with each other, such
as Principal Component Analysis (PCA) (Manly, 1986;
Reutter et al., 2003). These components represented
the combined factors explaining variability. One thing
that distinguished ENFA from PCA was the formed
components had a direct ecological significance.
The first component referred to as species ecological
niche marginality (marginality) which described the
distribution of species in relation to the mean of global
distribution (study area). The higher the marginality
coefficient, the different the habitats of species from
the average condition of environmental variable in the
study area showed. Marginality coefficient was
determined as the absolute difference between the