Coral Toxicity Testing Methods: Life Stages, Endpoints and Standardisation
Coral toxicity tests expose a defined coral life stage, from gametes and larvae to adult fragments or cultured cells, to a substance or product under controlled conditions and measure effects such as bleaching, loss of photosynthetic efficiency, tissue loss, failed settlement or death. No OECD Test Guideline exists for corals yet, so methods still differ between laboratories in species, life stage, light, exposure duration and statistics, and results are only comparable when those choices are reported.
Published 26 October 2026. Sources checked on that date.
Which coral life stage is tested?
Coral tests fall into four families, defined by the biological material exposed. Each answers a different question, and none replaces the others.
| Material | Typical endpoints | Strength | Limit |
|---|---|---|---|
| Gametes (fertilisation assays) | Fertilisation success over a few hours | Short, sensitive to reproduction effects | Only available during spawning, a few nights a year |
| Larvae (planulae) | Survival, deformity, settlement and metamorphosis | Captures the recruitment stage reefs depend on | Seasonal supply; wild spawn varies batch to batch |
| Adult fragments (nubbins) | Bleaching, photosynthetic efficiency, tissue loss, mortality | Available year-round from cultured colonies; many replicates per genotype | Adults can be less sensitive than early stages for some chemicals |
| Cultured coral cells | Cell viability, mechanistic markers | Fast, useful for mechanism and ranking | No whole-organism or symbiosis response |
The adult-fragment approach goes back to work such as Shafir et al. (2003), who used hundreds of small nubbins cut from Stylophora pistillata colonies to run replicated exposures. Larval assays, such as those developed for Acropora larvae by Negri and colleagues and by Flores et al. (2020), test settlement and survival. Downs et al. (2014, 2016) combined planulae with cultured cells from several coral species.
Sensitivity across life stages is chemical-specific. For copper, Hédouin et al. (2016) found Pocillopora damicornis larvae only slightly more sensitive than adults, and both became more sensitive with season and with a 3 °C rise in temperature. For some UV filters the gap between early stages and adults is much larger, which is covered in UV filters and corals.
What do coral toxicity tests measure?
Endpoints range from early, sub-lethal signals to death. A good test reports several, because they move at different speeds.
- •Bleaching: loss of the symbiotic algae or their pigments, scored visually or from calibrated photographs.
- •Photosynthetic efficiency: chlorophyll fluorescence of the symbionts (for example Fv/Fm measured by PAM fluorometry), an early and quantitative signal.
- •Tissue loss and mortality: partial or total loss of living tissue from the skeleton.
- •Behaviour and morphology: polyp retraction, mucus production, filament ejection, deformities.
- •Early-life endpoints: fertilisation, larval survival, settlement and metamorphosis.
- •Growth: calcification or buoyant weight, usually only meaningful over weeks to months.
In the adult-fragment method described by Brefeld et al. (2024), bleaching and mortality were the most robust endpoints across the four test substances, which included oxybenzone and copper.
Acute vs chronic coral tests
Acute tests are short, commonly 96 hours for adult fragments, and capture effects that appear quickly. Chronic tests run for weeks to months, follow slow endpoints such as growth, and can include a recovery phase to see whether effects reverse once exposure stops.
Neither is better in general. An acute test is the practical first answer for a product or substance; a chronic test is the relevant one when exposure in the field is continuous, as it can be for discharges or persistent substances. ReefTox runs both designs, described in the ReefTox methodology.
Which design choices change the result?
The same substance can look harmless or toxic depending on how it is delivered. Reviews of the field, including Mitchelmore et al. (2021) for UV filters, single out the following.
- •Light: Downs et al. (2016) reported a 24-hour LC50 of oxybenzone for S. pistillata planulae of 139 µg/L in light against 779 µg/L in darkness.
- •Nominal vs measured concentration: poorly soluble, lipophilic substances stick to glass and tissue. Fel et al. (2019) measured concentrations of several filters well below the nominal values they had dosed.
- •Static, semi-static or flow-through: renewing the test medium keeps exposure closer to the intended level over 96 hours.
- •Carrier solvent: substances that do not dissolve in seawater need a carrier, which needs its own control.
- •Temperature and season: Negri and Hoogenboom (2011) and Hédouin et al. (2016) both found that contaminants and heat interact, so test temperature should be reported and justified.
- •Species and genotype: closely related species can respond differently, which is the subject of multi-species coral testing.
NOEC, LOEC, EC50 and EC10: which number to trust?
Coral studies report two kinds of values. Hypothesis-testing values (NOEC, the highest concentration with no significant effect, and LOEC, the lowest with one) can only take the concentrations the experimenter chose. Regression values (EC50, EC10) are estimated from a fitted concentration-response curve.
OECD statistical guidance (2006) and Warne and van Dam (2008) set out the weakness of NOEC and LOEC: they depend on spacing and replication, and a noisier experiment can return a higher, less protective NOEC. Regression-based ECx values are generally preferred where the data allow.
The practical consequence: a multi-concentration design is needed to derive an EC50 or EC10. A single-concentration test answers a different question, whether a defined dose causes an observable effect, and should be reported as such.
Where does standardisation stand?
As of October 2026, there is no OECD Test Guideline for corals, and no coral project appears on the OECD Test Guidelines work plan. Marine ecotoxicity programmes therefore rely on algae, crustaceans, molluscs and fish, with coral data added from non-guideline studies.
Proposed methods exist. Brefeld et al. (2024) described a 96-hour semi-static assay with adult Montipora digitata fragments, run in a single laboratory and presented by its authors as suitable for further validation. Validation across several laboratories is the next step any such method needs before regulators can rely on it.
The data already published are substantial but uneven. A systematic review by Ouédraogo et al. (2023) compiled 847 studies from 181 articles and derived 105 toxicity thresholds for 56 chemicals or mixtures. Reference values set to protect marine organisms appeared protective of corals for all but three: copper, diuron and irgarol 1051.
Frequently asked questions
Is there an official standard method for coral toxicity testing?
Not yet. No OECD Test Guideline exists for corals as of October 2026. Published methods, such as the 96-hour adult-fragment assay described by Brefeld et al. (2024), are proposed for further validation, and laboratories still differ in species, life stage and conditions.
Which coral life stage is most sensitive?
It depends on the chemical. Larvae were far more sensitive than adults for some UV filters, while for copper Hédouin et al. (2016) found larvae only slightly more sensitive than adults of the same species. That is why the life stage must be stated next to any result.
What is the difference between a NOEC and an EC10?
A NOEC is the highest tested concentration with no statistically significant effect, so it depends on which concentrations were chosen and on replication. An EC10 is estimated from a fitted concentration-response curve as the concentration causing a 10% effect. Statistical guidance has favoured regression-based ECx values for this reason.
Why does light matter in coral tests?
Some UV filters become more toxic under light. Downs et al. (2016) reported a 24-hour LC50 for oxybenzone in coral planulae of 139 µg/L in light against 779 µg/L in darkness, so a test run without realistic light can understate effects.
Can results from different coral studies be compared directly?
Only with care. Reviews of the field note that species, life stage, light regime, duration, carrier solvent and nominal versus measured concentrations vary between studies, and each of these can move the result.
Sources
Peer-reviewed and agency sources, checked 26 October 2026. Values quoted on this page are specific to the species, life stage and conditions of each study and should not be read as general coral thresholds.
- Shafir S, Van Rijn J, Rinkevich B (2003). The use of coral nubbins in coral reef ecotoxicology testing. Biomolecular Engineering 20:401-406
- Downs CA et al. (2014). Toxicological effects of the sunscreen UV filter, benzophenone-2, on planulae and in vitro cells of the coral Stylophora pistillata. Ecotoxicology 23:175-191
- Downs CA et al. (2016). Toxicopathological effects of the sunscreen UV filter, oxybenzone (benzophenone-3), on coral planulae and cultured primary cells. Archives of Environmental Contamination and Toxicology 70:265-288
- Puisay A, Pilon R, Hédouin L (2015). High resistance of Acropora coral gametes facing copper exposure. Chemosphere 120:563-567
- Negri AP, Hoogenboom MO (2011). Water contamination reduces the tolerance of coral larvae to thermal stress. PLoS ONE 6(5):e19703
- Flores F et al. (2020). Toxicity thresholds of three insecticides and two fungicides to larvae of the coral Acropora tenuis. PeerJ 8:e9615
- Hédouin LS, Wolf RE, Phillips J, Gates RD (2016). Improving the ecological relevance of toxicity tests on scleractinian corals: influence of season, life stage, and seawater temperature. Environmental Pollution 213:240-253
- Fel J-P et al. (2019). Photochemical response of the scleractinian coral Stylophora pistillata to some sunscreen ingredients. Coral Reefs 38:109-122
- Brefeld D et al. (2024). Acute toxicity assays with adult coral fragments: a method for standardization. Toxics 12(1):1
- Mitchelmore CL et al. (2021). A critical review of organic ultraviolet filter exposure, hazard, and risk to corals. Environmental Toxicology and Chemistry 40(4):967-988
- Ouédraogo D-Y et al. (2023). What are the toxicity thresholds of chemical pollutants for tropical reef-building corals? A systematic review. Environmental Evidence 12:4
- OECD (2006). Current Approaches in the Statistical Analysis of Ecotoxicity Data: A Guidance to Application. OECD Series on Testing and Assessment No. 54
- Warne MStJ, van Dam R (2008). NOEC and LOEC data should no longer be generated or used. Australasian Journal of Ecotoxicology 14:1-5
See how these choices apply to your product
Coral testing for cosmetic formulations and SPF ingredients, or coral as a complementary species in a marine ecotox programme.