Abstract
Coral reefs are increasingly subjected to major disturbances threatening the health of marine ecosystems. Substantial research is underway to develop intervention strategies that assist reefs in recovery from, and resistance to, inevitable future climate and weather extremes. To assess potential benefits of interventions, mechanistic understanding of coral reef recovery and resistance patterns is essential. Recent evidence suggests that more than half of the reefs surveyed across the Great Barrier Reef (GBR) exhibit deviations from standard recovery modelling assumptions when the initial coral cover is low ($\leq 10$%). New modelling is necessary to account for these observed patterns to better inform management strategies. We consider a new model for reef recovery at the coral cover scale that accounts for biphasic recovery patterns. The model is based on a multispecies Richards' growth model that includes a change point in the recovery patterns. Bayesian inference is applied for uncertainty quantification of key parameters for assessing reef health and recovery patterns. This analysis is applied to benthic survey data from the Australian Institute of Marine Sciences (AIMS). We demonstrate agreement between model predictions and data across every recorded recovery trajectory with at least two years of observations following disturbance events occurring between 1992--2020. This new approach will enable new insights into the biological, ecological and environmental factors that contribute to the duration and severity of biphasic coral recovery patterns across the GBR. These new insights will help to inform managements and monitoring practice to mitigate the impacts of climate change on coral reefs.
Introduction
As climate change increases the frequency of disturbances like coral bleaching, storm swells, and crown-of-thorns starfish outbreaks, predicting how fast a reef can bounce back is critical for management. Historically, researchers used "single-phase" models like the Logistic or Gompertz curves. However, recent data suggests that when coral cover is very low (below 10%), reefs often enter a "biphasic" recovery pattern: a transient period of very slow growth (averaging four years) before shifting into a faster, more traditional growth phase. Ignoring this initial lag can lead to overestimating future coral cover by as much as 22%, potentially resulting in poorly allocated restoration resources.
Datasets and Spatial Context
The study utilizes benthic survey data from the Australian Institute of Marine Science (AIMS) Long Term Monitoring Program (LTMP) and Marine Monitoring Program (MMP), spanning 1992 to 2020. The scope covers 373 sites across 135 reefs. The authors identified 737 specific recovery trajectories—sequences of monitoring visits between major disturbance events—to calibrate and validate the model.
Methods
The authors developed a mathematical framework that generalizes the Richards’ growth model—a flexible sigmoid curve that can mimic logistic or Gompertz growth depending on a shape parameter ($\gamma$). The core innovations include:
- Biphasic Piecewise Modeling: The model uses a "change point" ($T_d$) and a recovery scale factor ($\alpha_d$). Before the change point, growth is suppressed; after it, the reef reverts to standard growth rates.
- Multispecies Extension: Recognizing that different corals grow at different speeds, the model can split the reef into groups (e.g., fast-growing Acroporidae vs. others) that compete for the same total available space.
- Bayesian Uncertainty Quantification: Instead of providing a single "best guess," the authors use Bayesian inference to produce a range of probable outcomes (credible intervals).
- Adaptive MCMC: Because the "change point" creates a discontinuity in the model's gradient, the authors employed a robust adaptive random-walk Metropolis-Hastings algorithm to sample the posterior distribution.
Key Results and Insights
The Bayesian framework successfully characterized parameters for 95% of the available recovery trajectories. The model demonstrated "excellent agreement" with observed data across the GBR. In representative case studies, the majority of observed data points fell within the model's 95% credible intervals.
Key findings include:
- Identifiability as a Diagnostic: If the data doesn't support a biphasic pattern, the change point becomes "non-identifiable," allowing the model to naturally collapse into a simpler single-phase model.
- Drivers of Biphasic Recovery: Multispecies analysis revealed that biphasic patterns are emergent properties of specific local ecosystems. At Lady Musgrave, the Acroporidae family drives the lag, while at Thetford, the lag is driven by other hard coral groups.
- Quantifying the Lag: The model confirmed that the suppressed growth phase typically lasts between 1 to 4 years, providing a concrete window for managers to expect slow progress before a "boom" in recovery.
Conclusion and Future Directions
While the model shows high reliability, especially for reefs with less than 30% coral cover, limitations remain. Currently, the maximum possible coral cover ($K$) is treated as a fixed value; future versions should infer $K$ as a parameter. Additionally, moving toward Stochastic Differential Equations (SDEs) could better account for the inherent biological randomness of coral growth. This new approach provides a more realistic and robust tool for conservation management, enabling deeper insights into the factors contributing to biphasic coral recovery across the Great Barrier Reef.
Summary updated