Habitat capacity is decoupled from coral cover on turbid reef systems

Turbid coral reefs are increasingly considered potential climate refugia yet are often assumed to be functionally limited due to their structurally simple reef frameworks. The success of these systems is commonly inferred from coral cover, which may overestimate reef condition. Consequently, their capacity to provide habitat remains poorly constrained. Here, we combine colony-level underwater photogrammetry ( n = 119) with ex-situ light scanning of corals ( n = 138) across 24 genera to quantify the Absolute Spatial Refuge (ASR) on turbid reefs in Singapore. Bayesian Gaussian generalized linear models were used to derive morphology-specific scaling relationships between colony length and ASR. These relationships were then applied to phototransects to estimate mean ASR and evaluate reef-scale habitat capacity at two sites. Both reefs exhibited high coral cover (Pulau Satumu; 68.6 ± 3.2%, and Pulau Hantu; 51.6 ± 11.9%) but contrasting structural characteristics. Hantu, characterised by lower diversity (19.3 ± 1.3 taxonomic amalgamation units [TAUs] per transect) and colony density (5.3 ± 0.3 colonies m −2 ), supported a higher ASR value (3.05 ± 1.8 ASR m −2 ) driven by a few large colonies. In contrast, Satumu, despite higher diversity (33.5 ± 2.3 TAUs per transect) and colony density (12.2 ± 1.0 colonies m −2 ), exhibited substantially lower ASR (0.45 ± 0.1 ASR m −2 ), reflecting the dominance of small, low-complexity colonies. These results demonstrate that conventional metrics of reef condition do not reliably capture habitat capacity in turbid systems. High coral cover can coincide with low 3D habitat provision and therefore translating morphology-based estimates of spatial refuge provides a more robust framework for assessing reef function and resilience. • Quantifying habitat capacity in turbid reefs is vital as they are increasingly considered climate refugia. • A predictive ASR framework was developed from 258 colony models spanning 24 coral genera. • Habitat function is driven by morphology distribution and dominant coral structures rather than coral cover. • Habitat volume scales nonlinearly with colony diameter and reef demographic structure. • Historical phototransect datasets can reconstruct temporal changes in reef structure and habitat capacity