Elsevier

Ecological Modelling

Volume 423, 1 May 2020, 108999
Ecological Modelling

Agent-based model of Eastern Pacific damselfish and sea urchin interactions shows increased coral reef erosion under post-ENSO conditions

https://doi.org/10.1016/j.ecolmodel.2020.108999Get rights and content

Highlights

  • Agent based modeling provides new insight into the effect damselfish behavior has on coral reefs.

  • Damselfish aggression affects the spatial distribution of echinoids on a Pacific coral reef.

  • Spatial distribution of echinoids affected by damselfish aggression increases coral bioerosion.

Abstract

Significant increases in the population of the echinoid bioeroder Diadema mexicanum that may soon follow severe El Niño events, such as in 1982–1983, can have a significant negative impact on the carbonate budget of coral reefs. We developed a spatially explicit model that uses agent-based modeling techniques to simulate the interactions between damselfish and sea urchins on an eastern Pacific coral reef following an El Niño-Southern Oscillation (ENSO) event where high echinoid abundances and low coral cover were prevalent. Our modeling study suggests that the agonistic behavior of damselfish towards echinoids invading their defended algal lawn territories has a magnifying effect on the degree of bioerosion attributed to echinoids. This is due to the increased likelihood that sea urchins ejected from damselfish territories will form concentrated aggregations and remain grazing and eroding coral for longer periods of time. This is a novel insight that contrasts with the previous understanding of the positive effect that damselfish have on reef carbonate budgets by protecting carbonate substrates that lie within their defended algal lawn territories. The increased degradation of coral stands attributed to the indirect damselfish effect, if it results in sea urchins eroding subsurface coral framework structures, causing instability and collapse (especially during periods of high-water motion), may contribute to the fracturing of large portions of coral framework blocks, affecting the recovery trajectory of reefs following a severe El Niño disturbance event.

Introduction

Trophic and non-trophic interactions can have significant influences on ecosystems (Schmitz, 2009). For example, aggression or predation can lead to shifts in habitat use in many organisms, marine as well as terrestrial (Pringle et al., 2019). On eastern Pacific coral reefs, population and movement dynamics of a sea urchin (Diadema mexicanum) can be positively or negatively influenced by damselfish (Stegastes acapulcoensis) aggression (Glynn, 1990; Hixon, 1997, 2015; Jones, 1991; Karlson, 1999; Kaufman, 1977; Pennings, 1997; Potts, 1977; Vine, 1974). In defending their algal lawn territories, damselfishes aid (Sammarco, 1980; Wellington, 1982) or interfere (Lobel, 1980; Potts, 1977) with sea urchins, by providing trophic resources or in limiting grazing time respectively. But damselfish also influence coral recruitment, competitive outcomes among corals (Kaufman, 1977), reduce coral mortality from predators and competitors (Glynn and Colgan, 1988; Sammarco and Carleton, 1982; Sammarco and Williams, 1982), and affect bioerosive processes (Risk and Sammarco, 1982; Sammarco et al., 1987; Strömberg and Kvarnemo, 2005). Thus, a tight coupling of movement and population dynamics of sea urchins and damselfish behaviors and abundance can be expected that in turn will influence corals and reef-building accretion. But the net influence of this multi-faceted interaction on the ecosystem and on reef-building are still poorly understood. It furthermore remains unquantified whether the spatial distribution of damselfish on a Panamanian coral reef (Uva Island) might exacerbate the erosion and degradation of the reef framework, especially following severe El Niño Southern Oscillation (ENSO) events, such as in 1982–1983.

The fundamental ecological interactions between damselfish, which cultivate and protect algal lawn territories, and sea urchins that graze on algae and in the process erode coral substrates, are herein explored in a modeling study based on empirical observations. ENSO events (Cortés et al., 1984; Glynn, 1983, 1984; Robinson, 1985; Von Prahl, 1983) trigger these interactions by precipitating increases in sea urchin population abundances (Eakin, 1996; Glynn, 1988). Sea urchins attempting to invade and graze on algal lawns are aggressively ejected by damselfish and tend to form aggregations outside of defended lawn territories. The sea urchin aggregations persist and result in severe coral erosion (bioerosion) (Eakin, 1996, 2001; Glynn, 1988). Later sea urchin declines coincided with loss of topographic complexity, suggesting echinoid population collapse was precipitated by their increased bioerosional activity and loss of sheltering habitat (Eakin, 2001). After their population collapse, Uva reef experienced robust recovery and its present-day morphology consists of patches with high live coral cover interspersed with areas of little to no coral cover strewn with dead coral rubble (Fong et al., 2017).

From this sequence of events we derive the testable null-hypothesis that damselfish ejection of sea urchins from their territories has no effect on coral bioerosion. Our empirical observations support an alternative hypothesis, namely that the defense of algal lawns by damselfish can indeed promote sea urchin aggregations that exacerbate coral bioerosion. Thus, a spatially heterogeneous pattern of high and low densities of sea-urchins will lead to areas of increased erosion versus areas of favored coral growth. By protecting their algal lawns from sea-urchins, the damselfish at the same time may create a refuge for corals. High densities of damselfish territories could therefore protect corals, which is in contrast to the Caribbean where some damselfish species kill corals to establish their territories and can therefore negatively impact coral health (Precht et al., 2010; Vermeij et al, 2015).

To evaluate the suspected role of damselfish-sea urchin interaction, we developed a spatially explicit model of a study site located on Uva reef (Gulf of Chiriquí, Panama) following an El Niño-Southern Oscillation (ENSO) event that resulted in high echinoid abundances and low coral cover using agent-based modeling techniques to simulate the daily interactions between damselfish and sea urchins across both space and time to explore whether this alternative hypothesis could indeed be supported.

Section snippets

Study site

Studies were conducted at the Uva Island coral reef (7°48′46″N, 81°45′35″W) in the Gulf of Chiriquí, Pacific Panama (Fig. 1A and B). This reef is ~2.5 ha in planar area, shallow (1–6 m depth) and is situated in a sheltered, NW-facing embayment. The principal framework- building corals are ramose, pocilloporid species, mainly Pocillopora damicornis and Pocillopora elegans.

A bathymetric map (Fig. 1 C) was produced in May 2010, using a downward-directed ADCP (Acoustic Doppler Current Profiler)

Results

Mean sea urchin ejection event count frequencies within the MP for the n1 = 250 test cases where damselfish are present was 0.886 ± 0.036 ejections m−2 day−1 (mean ± SD) (Fig. 8 A). Sea urchin ejection event count frequencies were binned in intervals of 0.01 ejections m−2 day−1. The mean sea urchin ejection event count frequency included ejection events resulting from damselfish aggression and from ejection events caused when a sea urchin aggregation attempted to move to another cell in the MP

Discussion

Our modeling study confirms a strong interaction between damselfish and sea urchins and that damselfish agonistic behavior towards invading echinoids contributes to a complex indirect effect on the overall CaCO3 budget of a coral reef. The bioerosion score frequencies where damselfish were present was 2.0842 ± 0.8398 g m−2 day−1, which are significantly different than the bioerosion score frequencies of 1.526 ± 0.847 g m−2 day−1 where damselfish are not present (two sample t-test, t = 45.177,

Credit author statement

Peter J. Glynn: Conceptualization, Methodology, Software, Validation, Formal Analysis, Writing – Original Draft, Formal Analysis. Peter W. Glynn: Investigation, Writing – Original Draft, Writing – Review & Editing. Juan Mate: Resources, Project Administration. Bernhard Riegl: Investigation, Writing – Original Draft, Writing – Review & Editing, Resources.

Declaration of Competing Interests

None.

Acknowledgments

P.W. Glynn acknowledges U. S. National Science Foundation support, Biological Oceanography Program grants OCE-1447306, OCE-9314798 and earlier awards. We thank Kevan Mantell (Dive Base Coiba, Pixvae, Panama), Ernesto Santos, and Alfredo de León (Mistolín) for field support. The Ministry of Environment (MiAmbiente), Republic of Panama, granted permits to work in the Coiba National Park.  Juan Maté was supported by the Smithsonian Tropical Research Institute.

References (54)

  • J.J. Alvarado et al.

    Reconstruction of Diadema mexicanum bioerosion impact on three Costa Rican Pacific coral reefs

    Rev. Biol. Trop. (Int. J. Trop. Biol.)

    (2012)
  • J.J. Alvarado et al.

    Coral reef bioerosion in the eastern tropical Pacific

  • R.C. Carpenter

    Predator and population density control of homing behavior in the Caribbean echinoid Diadema antillarum

    Mar. Biol.

    (1984)
  • J. Cortés et al.

    Pérdida de zooxantelas y muerte de corales y otros organismos arrecifales en el Caribe y Pacífico de Costa Rica

    Rev. Biol. Trop.

    (1984)
  • C.M. Eakin

    Damselfishes and their algal lawns: a case of plural mutualism

    Symbiosis

    (1987)
  • C.M. Eakin

    Where have all the carbonates gone? A model comparison of calcium carbonate budgets before and after the 1982–1983 El Niño at Uva Island in the eastern Pacific

    Coral Reefs

    (1996)
  • C.M. Eakin

    A tale of two ENSO events: carbonate budgets and the influence of two warming disturbances and intervening variability, Uva Island, Panama

    Mar. Pollut. Bull

    (2001)
  • Y. Eynaud et al.

    Herbivore space use influences coral reef recovery

    R. Soc. Open Sci.

    (2016)
  • P. Fong et al.

    Algal dynamics: alternate stable states of reefs in the eastern tropical Pacific

  • P.J. Glynn et al.

    El Niño, echinoid bioerosion and recovery potential of an isolated Galápagos coral reef: a modeling perspective

    Mar. Biol.

    (2017)
  • P.W. Glynn

    Extensive ‘bleaching’ and death of reef corals on the Pacific coast of Panama

    Environ. Conserv.

    (1983)
  • P.W. Glynn

    Widespread coral mortality and the 1982/1983 El Niño warming event

    Environ. Conserv.

    (1984)
  • P.W. Glynn

    El Niño warming, coral mortality and reef framework destruction by echinoid bioerosion in the eastern Pacific

    Galaxea

    (1988)
  • P.W. Glynn et al.

    Defense of corals and enhancement of coral diversity by territorial damselfishes

  • P.W. Glynn

    Feeding ecology of selected coral-reef macroconsumers: patterns and effects on coral community structure

  • P.W. Glynn et al.

    Sporadic disturbances in fluctuating coral reef environments: El Niño and coral reef development in the eastern Pacific

    Am. Zool.

    (1992)
  • M.A. Hixon

    Effects of reef fishes on corals and algae

  • Cited by (3)

    • Predicting shifts in demography of Orbicella franksi following simulated disturbance and restoration

      2022, Ecological Modelling
      Citation Excerpt :

      Due to these myriad threats, tropical coral cover loss is estimated at 50–75% globally (De'Ath et al., 2012; Hughes et al., 2018b; Bruno et al., 2019). Additional threats to corals include hypoxic events, disease outbreaks, algal overgrowth, severe storms, and predation, all of which may pressure a population simultaneously (Bruno et al., 2007; De'Ath et al., 2012; Glynn et al., 2020). Further, the increase in prevalence of diseases, namely stony coral tissue loss disease (SCTLD) has led to extreme colony loss of highly susceptible corals and is spreading rapidly in many regions (Muller et al., 2020; Brandt et al., 2021).

    View full text