Open Access Te Herenga Waka-Victoria University of Wellington
thesis_access.pdf (11.03 MB)

Modelling regime shifts of coral reefs to sponge reefs

Download (11.03 MB)
posted on 2021-12-08, 14:36 authored by Alberto Rovellini

Coral reef ecosystems have been degrading globally for decades due to global climate change and anthropogenic pressure, and corals are expected to continue declining in the future. Some coral reef sponges, although also vulnerable to warmer seawater, are more tolerant than corals to the combined effects of projected ocean warming and acidification. As a result, sponges have been proposed as potential winners on future coral reefs. Shifts towards sponge dominance on coral reefs may result into altered reef trophodynamics. However, there is limited information on how ecological processes on coral reefs will be affected by an increase in sponge dominance. Therefore, the potential ecological functioning of sponge-dominated reefs is unclear. To understand how reef ecosystems could function if sponges will become dominant, it is important to clarify: how perturbations that have a negative effect on corals propagate through the reef ecosystem; how sponge assemblages can develop over time; and how increased sponge dominance will affect other coral reef organisms. Focusing on Indo-Pacific coral reefs, this thesis investigates the temporal dynamics and the ecological processes that characterise reefs where sponges form abundant and highly diverse assemblages. Aims of this thesis were: 1) to explore the ecosystem-level effects of climate change-induced shifts towards sponge dominance on coral reef ecosystems; 2) to quantify the temporal variability of sponge abundance and biodiversity on coral reefs; 3) to understand how benthic sessile organisms benefit from free substrate that becomes available when corals die; 4) and to determine whether sponges can provide structural complexity similar to that of corals.  In my first data chapter, I use qualitative mathematical modelling to identify the key ecological processes that determine the functioning of a sponge-dominated reef. With a set of qualitative models, I obtain qualitative predictions of the ecosystem-level effects of increased temperature, acidity, and turbidity on a coral reef ecosystem, and of a shift towards sponge dominance. Model predictions showed that simulated environmental change caused coral decline, subsequent loss of reef structural complexity, and increased sponge and macroalgal abundance, while enhancing secondary productivity and detrital pathways. By analysing the uncertainty around model predictions, I highlight the need to elucidate some aspects of sponge ecology to better understand how sponge-dominated reef may function. These processes include temporal changes in sponge assemblages, competitive interaction between sponges and algae, and the contribution of sponges to reef structural complexity. I address these aspects in the following chapters.  In my second data chapter, I characterise the temporal variability of an abundant and diverse sponge assemblage on an Indonesian coral reef (Hoga Island, Wakatobi Marine Park, SE Sulawesi), studying temporal changes in sponge abundance and biodiversity over 13 years (2005-2017). Mean sponge abundance showed large temporal variability at this reef over the study period. In total, 141 sponge taxa were recorded at this reef over the study period, and species composition of the assemblage showed strong temporal patterns that were determined by the population dynamics of a few abundant taxa (e.g. Protosuberites sp., Sycon sp., and Pericharax sp., on average ~ 25%, 20% and 5% of total sponge abundance, respectively). In this chapter, I highlight that Indo-Pacific sponge assemblages can develop rapidly, undergoing large temporal changes that are driven by species-specific population variability.  In my third data chapter, I describe the temporal dynamics of all benthic organisms on the Hoga reef over the period 2006-2017, and investigate the effects of free space on benthic cover at subsequent sampling events. In contrast with other reefs in the Wakatobi where corals have declined, corals were dominant and stable over time at this site, and no other benthic group increased in abundance over time. However, percentage cover of turf algae and sponges showed larger interannual variability than corals and crustose coralline algae (CCA). Bare substrate was a good predictor of turf percentage cover in the following year across sites. Moreover, percentage cover of algal groups combined with bare space was a good predictor of CCA percentage cover the following year, and of sponge cover at one site. In this chapter, I show that turf algae in the Indo-Pacific rapidly colonise newly available substrate, and that other benthic groups like CCA and sponges may overgrow substrates already fouled by turf.  In my fourth data chapter, I use 3D photogrammetry to compare the structural complexity of coral-dominated and sponge-dominated reef quadrats on Hoga Island, and to identify which coral and sponge growth forms contribute to reef structure. Computer-generated 3D models were used to measure reef rugosity (R, denoting complexity of reef contours), vector dispersion (1/k, denoting heterogeneity of the angles between reef surfaces) and fractal dimension (D₁₋₅, denoting complexity of refuge holes at five different spatial scales between 120-1 cm). Structurally complex coral-dominated and sponge-dominated quadrats were similar in terms of coarse scale structural complexity (R and 1/k). However, complexity at the scale of small refuge holes (D₅) was higher for coral-dominated quadrats, indicating that corals provide smaller refugia compared to sponges, mainly due to branching corals. Branching and massive corals determined the overall 3D structure of coral-dominated quadrats, and barrel sponges of sponge-dominated quadrats. In this chapter I show that, while sponge reefs can be three-dimensionally complex if upright sponge growth forms are abundant, sponges do not provide small-scale structural complexity (5-1 cm refuge holes) comparably to corals.  In my fifth data chapter, I present how I captured sponges and their ecological roles in the ecosystem model Atlantis, by drawing on the previous data chapters and on the existing literature. I then present my contribution to the development of an Atlantis application for the Australian Great Barrier Reef (GBR), focusing on how sponges were parameterised in this model. Important sponge functional roles were captured in Atlantis either by developing novel components of the Atlantis code, or by modifying existing code specific to other benthic organisms (like corals), in such a way that it would be applicable to sponges. The sponge-specific ecological processes added to Atlantis in the scope of this chapter include: competition of sponges for free substrate with other benthic organisms; sponge-provided structural complexity; sponge-mediated bioerosion of carbonates; and Silicon-limitation to sponge growth. This sponge framework was tested with Atlantis GBR. I captured sponges in Atlantis GBR as three groups (heterotrophic, phototrophic, and bioeroding), and I also added to the model one group of spongivorous fishes. I participated to the development and parametrisation of the coral reef benthos of the model, with a focus on sponges. I present the development and calibration (ongoing) of Atlantis GBR.  In summary, in this thesis I found that sponge-dominated reefs could be highly dynamic ecosystems, with different structural properties compared to coral reefs. I discuss how reef trophodynamics are likely to be altered as a result of increased sponge dominance, with potential increases in detrital carbon pathways that will depend on sponge population dynamics. I also highlight that some aspects of sponge ecology still need to be elucidated to better understand how shifts towards sponge dominance may happen, and how the resulting reef states will function. These aspects include the role of sponges as net producers of detritus, the competitive interactions between sponges and algae, and the use of sponge-provided structural complexity by reef fauna. Understanding how sponge-dominated reefs may originate and how they will function will be crucial to maintain the ecological, economic, and social value of coral reefs in the near future.


Copyright Date


Date of Award



Te Herenga Waka—Victoria University of Wellington

Rights License

Author Retains Copyright

Degree Discipline

Marine Biology

Degree Grantor

Te Herenga Waka—Victoria University of Wellington

Degree Level


Degree Name

Doctor of Philosophy

ANZSRC Type Of Activity code


Victoria University of Wellington Item Type

Awarded Doctoral Thesis



Victoria University of Wellington School

School of Biological Sciences


Bell, James; Dunn, Metthew; Fulton, Elizabeth