Predicting Coral Cover Trends From Local to Broad Spatial Scales

Area of Science:

• Marine biology
• Ecological modeling
• Conservation science

Background:

• Biodiversity monitoring requires rapid assessments of species abundance and distribution.
• Quantifying uncertainty is crucial for detecting changes in habitats and regions.
• Sparse coral reef monitoring data hinders accurate assessments of reef habitat changes.

Purpose of the Study:

• To introduce a comprehensive prediction framework for estimating hard coral cover trends and uncertainty at a local scale.
• To account for spatial and temporal dependencies in coral cover data.
• To incorporate environmental variables like heat stress and tropical cyclones.

Main Methods:

• A latent process formulation models spatial and temporal dependencies in coral cover.
• Weighted spatial aggregation predicts trends at subregional and regional scales.
• Outputs from multiple models are combined via the ReefCloud platform for broader scale predictions.

Main Results:

• The framework quantifies the effects of heat stress and tropical cyclones on coral cover.
• Predictions were demonstrated using case studies in the Great Barrier Reef and American Samoa.
• The approach successfully propagated uncertainty across different spatial scales.

Conclusions:

• The integrated framework offers a widely applicable method for predicting coral cover trends at various scales.
• The study highlights the potential for improved coral reef monitoring and decision-making.
• Addressing data limitations is essential for enhancing trend detection and attribution to environmental disturbances.