Back

Host traits, phylogeny, and symbiont identity explain variation in colony-level symbiont switching across reef-building corals

Scott, C. B.

2026-07-23 ecology
10.64898/2026.07.20.739650 bioRxiv
Show abstract

Flexibility in the coral-algal symbiosis has been proposed as a mechanism by which coral colonies may acclimatize to shifting environmental conditions, yet the rules governing when and why switching occurs remain unresolved. Here, I compile longitudinal data spanning 59 coral species, 26 years, and 95 globally distributed reefs to estimate switching probability at the colony-level across directional environmental change and thermal variability. Switching was neither universal nor random: the identity of the resident symbiont was the strongest predictor of switching, with Cladocopium-dominated and mixed communities showing the greatest flexibility. Transition probabilities were asymmetrical, with shifts toward Durusdinium consistently more probable than shifts away, regardless of starting genus. Among host species- level effects, morphological traits best explained switching propensity along changing environmental conditions, while both traits and phylogenetic history explained sensitivity to thermal variability, suggesting that these two dimensions of abiotic environments impose distinct selective pressures on the symbiosis. Most switching was ultimately transient; colonies had a 78% chance of reverting back to their original symbiont in the following year. Together, these results challenge the assumption that increased environmental variability and rising temperatures universally correlate with higher switching rates, and suggest instead that host traits, phylogeny, and resident symbiont identity may help identify the subset of taxa for which flexibility represents a viable acclimatization strategy under climate stress.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.