Unlocking fine-scale temporal dynamics of coral reef health: a multi-kingdom microbial fingerprinting approach across seawater and sediment.

Coral reefs are rapidly degrading worldwide, creating a need for sensitive biomonitoring tools to detect early ecological shifts before visible collapse. Traditional monitoring often misses these incipient changes. We hypothesized that multi-kingdom microbial fingerprinting could reveal fine-scale temporal dynamics and early bioindicators of reef environmental change. The study was conducted on the fringing reef of Réunion Island (south-west Indian Ocean), a system exposed to local anthropogenic pressures. Seawater and sediment samples were collected bimonthly from April to October 2022 at three reef sites. Bacterial (16S rRNA), fungal (ITS2), microalgal (TufA), and protistal (18S rRNA) communities were characterized using metabarcoding. Physicochemical parameters were integrated via multivariate analyses. Time emerged as a stronger structuring factor than space. Bayesian analyses identified 368 temporally associated ASVs (Amplicon Sequence Variant), revealing microbial succession patterns. Bacterial communities shifted from Pseudomonadota-dominated assemblages (April) to Bacteroidota-dominated communities later in the year, with recurring ASVs (e.g., Pseudohongiella, Neptuniibacter caesariensis, Mycobacterium). Fungal communities showed marked temporal fluctuations, including recurrent peaks of potentially pathogenic genera (Aspergillus, Candida, Ramichloridium). Microalgal assemblages were habitat-structured (Chlorophyta in seawater, Rhodophyta in sediments), while protists communities exhibited pronounced late-season shifts, including dinoflagellate enrichment in October. Measured physicochemical parameters explained only a limited fraction of community variability, suggesting a strong role for biological interactions and unmeasured drivers. These findings highlight the sensitivity of high-resolution, multi-kingdom microbial fingerprinting for detecting early-warning bioindicators, outperforming traditional low-frequency monitoring. This study provides a robust, transferable framework for proactive coral reef biomonitoring and ecosystem management under increasing anthropogenic pressure.