To quantify species-level changes in reef fish communities, the Institute for Marine Research employs the Diver-Operated Stereo Video System (SVS) to collect underwater video data across its long-term monitoring sites.
The resulting footage is analysed using EventMeasure, with the stereo measurement component enabling accurate three-dimensional reconstruction of fish size and spatial positioning. By transforming two-dimensional imagery into spatially explicit data through synchronised camera perspectives and rigorous calibration, this approach allows for precise and repeatable measurements.
By integrating these tools, observations of fish communities extend beyond qualitative assessment, becoming robust, quantitative data points with known and validated error margins.
Following analysis, the following metrics can be obtained:
metric tons
per km2
number of
individuals
per km2
number of
species
per km2
metric tons
per km2
per yr-1
On tropical coral reefs, structural complexity has been shown to be a major driver of productivity, biodiversity, and overall functionality of reef ecosystems. To understand more precisely the nature between structural complexity and reef health, we use 3-Dimensional (3D) Modelling as a tool to quantify topography, rugosity and other structural characteristics that play an important role in the ecology of coral reef communities. This 3D reconstruction of reef structure and complexity becomes integrated with other physiological and ecological parameters to improve monitoring of the health and function of coral reef ecosystems in Philippines, whilst providing a visual view of complexity change to IMRs partnering coastal managers.
Underwater benthic surveys are used to measure the composition and health of the reef “floor” — including corals, algae, rubble, sand, and other key habitat components. These surveys allow us to track changes over time, such as coral recovery after disturbance, shifts in algal dominance, or signs of reef degradation. By repeating surveys at fixed sites, we can detect both gradual trends and sudden ecological changes, providing a critical evidence base for reef management and conservation.
Benthic data for IMR’s long-term monitoring project is collected using underwater transects, with divers using cameras to capture high-resolution imagery of the seabed. These images are then analysed using ReefCloud, a machine-learning platform where IMR scientists train and validate models to rapidly identify and classify benthic organisms and substrate types. Once trained, the system automates much of the image analysis process, converting thousands of point observations into standardised estimates of percentage cover across key benthic categories.
This workflow ensures data is both efficient to process and consistent across time and sites, enabling robust comparisons and long-term tracking of reef condition.
The ability for coral populations to recover post-disturbance is highly dependent on patterns of reef connectivity, with the interchange of larvae among reefs supporting the recovery of disturbed reefs at varying distances from larval source reefs.
IMR monitors the recruitment, mortality and net turnover of juvenile corals (<5cm) arriving into Dauin’s reefs using 3D-modelling techniques. These techniques allow us to monitor volumetric growth, alongside determining the bottlenecks towards successful recruitment.
Trash (General / Fishing)
Corallivorous Invertebrates
(i.e. Acanthaster spp. & Drupella spp.)
Coral Disease
Coral Bleaching
© 2026 Institute for Marine Research. All Rights Reserved.
