Diadema Preference Experiments
Description
Uncontrolled urchin populations in temperate reefs can turn macroalgal forests into urchin barrens that can remain in an unproductive stable state for years. Such impacts currently expend with climate change and bioinvasions, as ocean warming helps spread tropical urchins, both native and invasive, into temperate areas. In this context, the Indo-Pacific-Red Sea urchin, Diadema setosum arrived twenty years ago in the eastern Mediterranean and its populations have been expanding in the past decade although its distribution is still very patchy. In a previous study, the urchin turned turf covered reef areas into bare rock within a few weeks. Here we tested if D. setosum shows preference for specific macroalgae species to assess if it may pose differential threat to different macroalgae. For that, D. setosum individuals, collected from Michmoret, Israel, 32°24'29"N 34°52'2.4"E, were offered eight different combinations of two algal species in a mesocosm system, including "No-Choice" (single species) controls during March - June 2024. The results show that the urchin feeds on all offered taxa with overall no or very minor preference to specific algae species. This means that all macroalgae communities are at high risk of over-grazing by D. setosum. If D. setosum populations continue to grow without intervention, the Levant rocky reef turf barrens, and areas still dominated by native marine forests or growing invasive species patches, might all turn into urchin barrens.
Files
Steps to reproduce
In the No-Choice experiments, consumption rates by D. setosum per urchin test size were compared between species pairs when offered alone with one-way ANOVA. In the Choice experiment, we used Wilcoxon signed rank test to compare average consumption rate on the species pairs because the assumption of independence between the replicates was violated as the two algae species were grazed by the same individual. In both types of experiments, normality of the data was tested (Shapiro-Wilk’s test), as well as the homogeneity of variance (Levene’s test). Data were statistically analyzed by R Core Team (2021), R version 4.5.1, and RStudio version 2023.06.0+421 (RStudio Team, 2020).