Comparing seabed roughness result from QPS fledermaus software, benthic trrain modeler [BTM] and developed model derived FRM slope variability algorithm for hard coral reef detection / Nur Asikin Mohd Sayud

Benthic habitat complexity can be explained as habitat species that lives in the bottom
of seafloor which play an important role in marine biodiversity. Therefore, the
existence of benthic habitats can affect the surface complexity such as coral reefs
habitats. Rugosity measurement is one of the ways that can be used to understand the
complexity of the seafloor. For example, marine biologists using rugosity
measurement in order to understand the growth and pattern of the coral reefs and to
identify the existing of coral reefs. There are several methods that can be used in
determine the rugosity which are using virtual area based rugosity and virtual chain
tape rugosity. As for this study, virtual area based is been used which applied to
bathymetry data set that has been collected by using R2 Sonic 2020. In this study,
several models has been created which are from QPS Fledermaus model, BTM model
and Slope Variability model. Slope variability model is an algorithm that is being used
for detecting terrain roughness. Thus, focus of this study is to determine the best
model that can be used to detect coral reefs area and to know the capabilities of slope
variability model. In this study, rugosity is been developed by using QPS Fledermaus
software. In ArcGlS software, DEM data will be process by using terrain roughness
model that has been derived by using Slope Variability algorithm. Slope surface is
been created in BTM by using same of DEM data. Then, accuracy assessment has
been done by comparing the model from Fledermaus and BTM with slope variability
model as to know the capabilities of those models in detecting the corals reefs area.
Results shows in a percentage values which are 92% of similarity for QPS Fledermaus
and slope variability model result and lastly 90% of similarity for BTM and slope
variability model result.