Estimation of coral reef area volume and weight using a three dimensions (3D) photogrammetry

Current methods for measuring coral biomass remain limited, as they require direct data on coral weight and volume. Traditional techniques for estimating coral volume and weight are often invasive and inefficient. This study integrates underwater 3D photogrammetry based on the Structure-from-Motion (SfM) technique to develop predictive models for estimating coral volume and weight at Gili Labak Island, Sumenep. The analysis revealed that linear regression provided the highest accuracy for volume estimation (y = 0.532x + 208.6; R² = 0.887; RMSE = 148.7 ml), while the power regression performed best for weight estimation (y = 1.0047x^1.005; R² = 0.906; RMSE = 205.3 g). Application of these models to field plots produced realistic biomass estimates—8.6 tons in Plot A and 0.8 tons in Plot B—with corresponding carbon stocks of 97.5 kg C (4.9 kg C m⁻²) and 9.1 kg C (0.5 kg C m⁻²), respectively. These findings demonstrate that regression-calibrated 3D photogrammetry provides a non-destructive, accurate, and scalable approach for estimating coral biomass, offering significant relevance for coral reef conservation and management in data-limited regions.