Coral Polyp Cutaway: Why Heat and Light Make Corals Expel Their Algae

A single coral polyp, cut open. Its golden-brown partner algae (zooxanthellae) feed it until heat, bright light and oxidative stress make it expel them. Raise the water temperature and watch the colour drain, then see whether the polyp starves or recovers.

Scenario

Preset

Time warp

Environment

Heatwave pulse

The heatwave button adds a pulse on top of the water temperature: ramp up, plateau, ramp down.

Live numbers

Temperature against the bleaching threshold

━ water °C · ━ this polyp's bleaching threshold (dashed)

Symbionts, oxidative load and reserve

━ symbionts (% of healthy) · ━ oxidative load (% of half-saturation) · ━ reserve (%)

Experiments (measured vs closed form)

Run an experiment. They use private copies of the model and do not touch your polyp.

Independent numerical checks

Also run outside the browser: node src/check.js.

Educational model with illustrative parameters. The polyp is drawn about 700 times life size, and the symbiont cells are shown far too big and in the hundreds instead of the millions; their number follows the simulated density.

Controls

Drag or one finger: orbit; wheel or pinch: zoom; the Cutaway, Mouth and Colony buttons change the view.
Pick a preset or move the sliders, press Marine heatwave, and watch the golden cells leave through the mouth.

The science

The partnership. A reef-building coral hosts photosynthetic dinoflagellates (Symbiodiniaceae, zooxanthellae) inside its gut-lining cells. The algae pass most of the sugars they make to the polyp; the polyp supplies shelter, CO₂ and nitrogen. The brown colour of a healthy coral is mostly the algae; the white of a bleached coral is the limestone skeleton showing through clear tissue.

Why heat bleaches. Above a local threshold (here the maximum monthly mean + 1 °C, the rule NOAA Coral Reef Watch uses for bleaching alerts) the algae's photosystems leak electrons and make reactive oxygen species. The model keeps one lagged stress variable, the oxidative load dx/dt = σ - x/τ, with σ = (T - T_c)⁺(0.4 + 0.6 · light) and τ = 10 days: heat and light add up, and the damage lags the heat by about a week, which is why bleaching shows up weeks after a heatwave starts.

Expulsion and regrowth. The polyp expels or digests algae at a rate m(x) = m₀ + m_max x²/(x² + x₅₀²) while they regrow logistically: dS/dt = r S (1 - S/K) - m(x) S. Without stress the density settles at S* = K(1 - m₀/r); after the stress ends it recovers along S = K / (1 + (K/S₀ - 1) e^-rt). With a known stress history the whole equation has an exact solution (Bernoulli substitution u = 1/S), used in the checks.

Starvation. The host's energy reserve changes as dR/dt = r_R(φ S - c) + g f: translocated carbon (a fraction φ of what the algae make) minus maintenance, plus feeding on plankton (f from 0 to 1). If the reserve reaches zero the polyp dies and the skeleton is overgrown by algae turf. Feeding can carry a bleached coral through a heatwave (reported in field and tank studies; here only qualitative).

Nutrients. Extra dissolved nitrogen lets the algae reach a higher density (K rises) but they keep more of the sugar (φ falls), and the bleaching threshold drops by 0.7 °C per unit of enrichment above 0.2: a simplified version of reported experiments.

Degree Heating Weeks. DHW is the sum, over the last 12 weeks, of the daily excess of sea temperature above the maximum monthly mean when that excess is at least 1 °C, divided by 7. About 4 °C-weeks warns of significant bleaching, about 8 of severe bleaching and mortality. The maximum monthly mean used here is a generic 28.5 °C.

Honest simplifications. One polyp, not a colony; one symbiont type (real corals host several, with different heat tolerance, and can shuffle them); every rate (r = 0.05 per day, τ = 10 days, m_max = 0.15 per day, x₅₀ = 8 °C day, r_R = 0.03 per day) is an illustrative value chosen to give weeks-long dynamics, not a measurement; real reef recovery takes years and involves the whole colony; calcification, disease, acidification and symbiont genetics are not modelled. Check the literature before quoting any number from this page.