Increasing seawater temperatures are expected to have profound consequences for reef-building corals’ physiology. Understanding how demography changes in response to chronic exposure to warming will help forecast how coral communities will respond to climate change. Here, we measure growth rates of coral fragments of four common species, while exposing them to temperatures ranging from 19 to 31˚C for one month to calibrate their thermal-performance curves (TPCs). Our results show that, while there are contrasting differences between species, the shape of the TPCs was remarkably consistent among individuals of the same species. The low variation in thermal sensitivity within species may imply a reduced capacity for rapid adaptive responses to future changes in thermal regimes. Additionally, interspecific differences in thermal responses show a negative relationship between maximum growth and thermal optima, contradicting expectations derived from the classic “warmer-is-better” hypothesis. Among species, there was a trade-off between current and future growth, whereby most species perform well under current thermal regimes but are susceptible to future increases in temperature. Increases in water temperature with climate change are likely to reduce growth rates, further hampering future coral reef recovery rates and potentially altering community composition.

We collected nine colonies of four common reef-building coral species (Acropora hyacinthus, Acropora tenuis, Pocillopora verrucosa, and Stylophora pistillata) to investigate temperature-growth relationships. The colonies were collected at Kelso Reef (-18.445˚, 146.993˚) in the Great Barrier Reef in March 2021 at depths between 2 and 8 m and subsequently transported to the National Sea Simulator (SeaSim) in the Australian Institute of Marine Science. From each colony, we obtained 10 fragments and pasted those fragments on aragonite plugs. One fragment per colony was allocated to each of 10 temperature treatments (19, 21, 23, 25, 26, 27, 28, 29, 30, and 31˚C) and allocated to a 50L experimental tank. Each temperature treatment had three replicate tanks, and all tanks had identical flow rates (0.8 L min^-1), light intensities (cumulative daily light integrals were 6.48 mol photons m^-2 d^-1), and feeding regimes (2x10⁶ cells L^-1d^-1of microalgae and 0.06 Artemia nauplii L^-1d^-1).

The fragments were buoyant weighted once target temperatures were reached, and then every two weeks (or sooner if the fragments started to die and we had to remove them from the experiment). The last measurements occurred at eight weeks. We included one additional plug without a coral fragment in each tank as control (i.e., to check if changes in weight were due to coral growth rather than algae settling on plugs).

Each plug was first weighted while dry, and a subset of 10 plugs per temperature treatment was buoyant weighed to fit a linear model predicting the buoyant weight of a plug as a function of its measured dry weight.

To calculate growth between time points, we converted the buoyant weight measurements to skeletal dry weight of the coral fragments (W_d) using Archimedes’ Principle:

• W_d  = (W_w - W_p) / (1 - D_w / D_d)

W_w is combined buoyant weight of the fragment and the plug, W_p is the buoyant weight of the plug, D_w is water density (with a salinity of 35psu), and D_d is skeletal density of the fragment.

Please refer to the published manuscript for more detailed information on data collection and processing.