Does more coral necessarily equate to more fish?

The strong association between reef fishes and hard corals is seemingly apparent, yet often assumed. We review published research to assess this long standing paradigm.
Does more coral necessarily equate to more fish?

Not many of us have been able to experience first-hand the awe and splendour of tropical coral reefs. Yet amongst the lucky ones who have (or a quick google image search for those that haven’t), the vivid imagery conjured is one that is almost always identical ─ vast landscapes of beautiful hard coral, brimming with an abundance and diversity of life, notably fishes. This phenomenal depiction of coral reefs often evokes the simple question; why are there so many fishes on coral reefs?

Habitat associations have underpinned our understanding of ecological systems, both terrestrial and marine. Often touted as the rainforests of the ocean1, the sheer amount of life inhabiting coral reefs has traditionally been attributed to the very thing that makes these systems unique ─ the live hard coral2. Research trying to understand the associations between these two iconic groups (fishes and corals) began sometime in the 1970s and continues to garner significant interest today3. In fact, synchronous surveys of live hard coral cover and fish populations have not only been used to describe the relationship between these two animal groups but also routinely as proxies for ecosystem health. What this also implies is that there is a vast treasure trove of published research documenting these relationships, and across multiple geographic locations. For someone about to embark on an entire thesis on fish-coral associations, it only made sense to look at what was already done. It was there that things got interesting.

As it turns out, the relationship between fishes and hard corals was not always consistent. Some studies would report a strong relationship between the two groups, while others failed to document any sort of relationship (sometimes even a negative one). Personally, all I wanted to do was formulate a general consensus describing how reef fishes associate with live coral, just so I could write this (mandatory) literature review as part of my PhD programme. However, the conflicting findings provided a unique opportunity, and birthed the idea for the paper; to condense all of this empirical research using a quantitative, and statistically robust framework.

Making a rather recent splash into ecology4, meta-analyses proved to be the silver bullet; the ultimate tool to encompass all of this variation in an objective and transparent manner. The idea here was to provide quantitative estimates (with uncertainty errors) of fish-coral associations, using the correlation coefficient5 as a sensible proxy for measuring association. So my co-authors and I set out, scouring two databases for every single piece of peer-reviewed article that would have possibly surveyed fishes and corals. After meticulously sieving through 4653 articles, we narrowed our final dataset (using a stringent theoretical and statistical criteria) to 66 articles which provided an overall sample size of 723 correlation coefficients. These correlations ultimately captured the association between commonly measured fish metrics (abundance, species richness, and biomass) and percent live hard coral cover. From there, we formulated several statistical models to look at the overall correlation between fishes and hard corals while also exploring potential factors that regulate this association.

Our results proved to be very interesting. The overall correlation between fishes and hard corals proved to be consistently weak (Fig. 1). Of course, looking at an “overall” correlation may mask some of the more nuanced ecological relationships between fishes and corals. As we know, some reef fishes are considered to be more associated with live coral than others3. So we tested this by looking at specific fish families and trophic (dietary) groups. Again, these results were surprising, and served as a contrast to what has been routinely reported. We found that no specific fish family or trophic group had any evidence (0% probability) for recording strong correlations (Fig. 2) and in fact, the vast a majority of correlations were weak. This was even so among obligate corallivores (Fig. 2B), which undoubtedly rely on live coral for food, yet only recorded moderate associations. As we note in the paper, we tested across a range of parameters, and consistently find a signal of predominantly weak fish-coral correlations. What our results suggest is that when looking across multiple systems, this association between reef fishes and hard corals is a rather tenuous one and can vary between being strong in one system and potentially negligible in another. 

Fig. 1 Bayesian posterior distribution of correlation coefficients (r) between fish metrics (abundance, biomass, and species richness) and percent coral cover. Each plot displays the sample size n (number of r, number of papers from which r were extracted) for each model, and the credible interval (probability) of the estimates. Shaded regions provide interpretation of correlation coefficients (e.g. weak, moderate).


 Fig. 2 Bayesian posterior distribution of correlation coefficients (r) between fish metrics (abundance, biomass, and species richness) and percent coral cover for (A) fish families and (B) trophic groups.

 Does this then come as a huge shocker? Maybe at a first glance. But after some thought, it might not be as surprising after all. Rarely in any natural system is a single variable or resource the main driver of a larger community. As we highlight, this stands true on coral reefs, at least between fishes and corals. While coral cover may serve as a strong driver of fish communities in certain systems, generalising and suggesting that this relationship holds true across all other systems can limit our understanding of other more pertinent factors. The occurrence and persistence of both fishes and corals are undeniably shaped by multiple evolutionary, biogeographic and ecological processes2,3,6. Acknowledging that each and every coral reef system is unique and would therefore be regulated by its own set of processes would facilitate a holistic understanding of that respective system and can invite more novel research in this field.

And so to answer the question, does more coral equate to more fish? Not always.

The full study can be accessed here .


  1. Connell, J. H. Diversity in tropical rain forests and coral reefs. Science (80-. ). 199, 1302–1310 (1978).
  2. Coker, D. J., Wilson, S. K. & Pratchett, M. S. Importance of live coral habitat for reef fishes. Rev. Fish Biol. Fish. 24, 89–126 (2014).
  3. Siqueira, A. C., Muruga, P. & Bellwood, D. R. On the evolution of fish–coral interactions. Ecol. Lett. 1–11 (2023)
  4. Gurevitch, J., Koricheva, J., Nakagawa, S. & Stewart, G. Meta-analysis and the science of research synthesis. Nature 555, 175–182 (2018).
  5. Schober, P. & Schwarte, L. A. Correlation coefficients: Appropriate use and interpretation. Anesth. Analg. 126, 1763–1768 (2018).
  6. Bellwood, D. R., Goatley, C. H. R. & Bellwood, O. The evolution of fishes and corals on reefs: Form, function and interdependence. Biol. Rev. 92, 878–901 (2017).

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Subscribe to the Topic

Freshwater and Marine Ecology
Life Sciences > Biological Sciences > Ecology > Freshwater and Marine Ecology
Community and Population Ecology
Life Sciences > Biological Sciences > Ecology > Community and Population Ecology
Coral Reefs
Physical Sciences > Earth and Environmental Sciences > Earth Sciences > Ocean Sciences > Marine Biology > Coral Reefs
Physical Sciences > Earth and Environmental Sciences > Earth Sciences > Ocean Sciences > Marine Biology > Fisheries
Marine Biology
Life Sciences > Biological Sciences > Ecology > Ecosystems > Marine Biology