The study, published open access in the journal Coral Reefs by Megan H. Williams of James Cook University and colleagues, set out to address a gap in reef ecology. While the direct harms macroalgae inflict on corals, including shading, abrasion, allelochemicals, and microbe-altering dissolved organic carbon release, are well documented, far less is known about how macroalgal biomass reshapes the broader early-successional benthic community, and how those shifts in turn influence coral settlement and persistence. Because recruitment is essential for reef recovery, and because early life stages are a well-recognized bottleneck in coral population dynamics, the question has direct implications for how restoration on macroalgae-dominated reefs should be managed. Across the tropics, inshore reefs exposed to nutrient enrichment and reduced grazing pressure are increasingly dominated by fleshy seaweeds, and managers need to know whether removing them can realistically tip communities back toward coral dominance.

The experiment took advantage of an ongoing local management trial known informally as "sea-weeding." At two inshore fringing reef sites, Arthur Bay and Florence Bay, roughly eight kilometres offshore from Townsville, twelve 25-square-metre plots had been established at three to five metres depth, with six randomly assigned to regular manual macroalgal removal and six left as untreated controls. Removal, which targets canopy-forming Sargassum species, began in October 2018 and continued two to three times per year. During the study period, macroalgae were cleared in July 2021, October 2021, and July 2022, and biomass was estimated from holdfast density and thallus height using an established allometric relationship. Across the study, average algal biomass in control plots was 560.8 grams per square metre, roughly 3.5 times the 162.1 grams per square metre recorded in removal plots. The residual biomass in removal plots reflects the practical reality of manual clearance: eradication is nearly impossible in a system where Sargassum recruits readily, and the goal is suppression below the level at which the canopy exerts ecosystem-scale effects.

To track community development and coral settlement, the team deployed 240 unglazed terracotta tiles, ten per plot, each measuring 11 by 11 by 1 centimetre. Tiles were mounted horizontally about five centimetres above the substrate on stainless-steel rods, allowing distinct communities to form on sunlit upper surfaces and shaded undersides. Installed in late August 2021, roughly two months before the annual mass spawning event of around 22 October 2021, the tiles received only naturally produced coral larvae. They were retrieved at approximately three, six, and thirteen months after deployment, photographed for community analysis, examined under a microscope for corals smaller than one centimetre, and then returned to their exact original positions to preserve microhabitat conditions. Percent cover of 27 biotic and abiotic categories was quantified from photographs using CoralNet, with categories distinguishing live from dead crustose coralline algae, long sediment-laden algal turfs from short productive turfs, biofilms, microbial mats, macroalgae, and bare tile. The design deliberately mimicked a key feature of real reef surfaces, where the undersides of rubble and overhangs serve as preferred settlement habitat because they combine low light with reduced sedimentation.

Multivariate analyses revealed that time was the strongest driver of community composition on both tile surfaces, reflecting clear successional progression from early colonisation stages in November 2021 to more developed assemblages by September 2022. Treatment effects were smaller but statistically significant, and their character depended on tile orientation. On the shaded bottom surfaces, macroalgal removal changed the trajectory of succession itself: removal plots were colonised faster, with crustose coralline algae covering 24.5 percent of tile bottoms in removal plots versus 4 percent in controls at the first census, and bryozoans doubling in removal plots by February 2022. By the final census, these differences had largely converged. On top surfaces, the pattern was reversed early on, with control plots, shaded by dense canopies, actually supporting more coralline algae than removal plots, a result the authors attribute to the photoinhibition that some coralline species suffer under high light. This counterintuitive outcome underscores how strongly microhabitat and canopy effects interact: what benefits calcifiers on a shaded surface can harm them on an exposed one.

Overall Shannon diversity responded surprisingly little to the intervention. The only significant difference was higher diversity on top surfaces of control plots, while bottom-surface diversity was unaffected by treatment throughout. The authors conclude that macroalgal removal shaped the successional trajectories of particular taxa, especially calcifiers, without substantially altering community-wide diversity, and note that seasonal senescence of Sargassum, including a marked natural biomass decline in July 2022, may have blurred treatment differences as the study progressed. Sargassum on the Great Barrier Reef typically dies back in the austral winter and re-establishes from perennial holdfasts in spring, meaning control and removal plots can temporarily converge in appearance even without management action.

The consequences for coral settlement were stark. Of 1,134 coral observations recorded across the study, 77 percent occurred at the first census, shortly after spawning, and roughly 80 percent of all corals were found on bottom tile surfaces. In November 2021, bottom surfaces in removal plots averaged 1.77 settlers per tile compared with 0.17 in controls, a greater than tenfold difference. Densities had fallen by February 2022 but remained significantly higher in removal plots at 0.37 versus 0.03 settlers per tile. By September 2022, corals were nearly absent from all tiles, at or below 0.002 recruits per tile, with no detectable treatment effect. Top surfaces showed no treatment differences at any time point. The trajectory follows the steep early-mortality curve familiar to coral ecologists: most larvae that settle never survive their first year, succumbing to competition with turf algae, sediment smothering, and predation by small grazers and invertebrate predators.

Statistical modelling linked these patterns to specific benthic features. Using the Boruta feature-selection algorithm to identify candidate predictors, then fitting generalised linear mixed models, the researchers found that live crustose coralline algae cover was a significant positive predictor of settlement one month after spawning: each 10 percent increase in coralline cover corresponded to roughly 1.4 times more settlers. This aligns with a long body of evidence that coralline algae emit chemical cues that induce coral larvae to settle. By the post-settlement phase, coralline cover no longer predicted coral abundance, while bare tile emerged as a strong negative predictor, with each 10 percent increase in bare surface associated with about 46 percent fewer surviving corals. Other confirmed predictors, including dead coralline algae, bryozoans, and turf categories, showed no independent effects, suggesting they merely co-occurred with favourable settlement conditions. In other words, coralline algae appear to help larvae choose a home, but they do little to keep that home habitable once the young coral begins to grow.

The authors propose several mechanisms for how macroalgal canopies suppress the calcifiers that facilitate settlement. On bottom surfaces, where the tile itself provides shade and canopy removal cannot change light, they suggest hydrodynamics: dense canopies suppress flow velocities and thicken boundary layers, and removing them can enhance mixing and nutrient delivery, conditions that favour calcifying organisms. Consumer dynamics likely contributed as well, since reduced canopy cover may have increased grazer access to tiles and suppressed turf, although grazing activity was not directly quantified. Fish and invertebrate herbivores often forage more freely where dense seaweed structure no longer offers shelter from predators or physical obstruction. The researchers also caution that recent work at the same sites found no measurable sediment differences following macroalgal removal, so sedimentation was probably not the driver, and they deliberately avoid attributing patterns to sediment dynamics that they did not measure.

Recruitment levels overall were far lower than in earlier experiments at the same site, where a 2022 study reported about 46 recruits per tile in removal plots compared with fewer than one per tile surface here. The authors point to the 2020 mass bleaching event, which likely reduced larval supply by impairing gamete production, as a probable cause. Thermal stress can cause corals to divert energy from reproduction toward survival, and inshore central Great Barrier Reef reefs were heavily affected during that event, leaving the local adult population depleted and reproductively compromised. Methodological differences also matter: the earlier study bleached tiles with sodium hypochlorite before counting, making corallites much easier to find, whereas this study used non-destructive live counts to preserve the developing community, likely underestimating total settlement but arguably giving a more accurate picture of survivors. The repeated handling required by the non-destructive design may itself have caused some mortality, though because all tiles were handled identically, the authors argue this is unlikely to bias treatment comparisons. Tile sides, which earlier work identified as prime settlement habitat, could not be photographed and were excluded from analysis, adding a further caveat to absolute counts.

The wider lesson is one of transience. Macroalgal biomass appears to suppress coral settlement indirectly by limiting the development of calcifier-rich communities, and clearing it can create a short-lived "settlement window" timed to the post-spawning period. But that window closed as communities converged, seasonal Sargassum dynamics erased biomass differences, and competition and predation drove the sharp mortality typical of Type III survivorship, the pattern in which most individuals die young and only a tiny fraction reach adulthood. Notably, persistence itself was not enhanced by removal, and on top surfaces removal actually reduced persistence probability, from 18 percent in controls to 2 percent in removal plots, possibly because surfaces cleared of canopy experienced harsher light or grazing exposure during the vulnerable post-settlement phase.

The authors conclude that managing macroalgae can meaningfully enhance early settlement opportunities, but sustained coral recovery on macroalgae-dominated inshore reefs will require interventions that also address post-settlement mortality and the broader environmental stressors that shape reef resilience. For practitioners, the timing message may be the most actionable element: clearance that is coordinated with the annual spawning season, and paired with measures such as herbivore protection, sediment and nutrient control, or assisted recruitment, stands a better chance of converting a brief settlement pulse into lasting population gains.