Abstract :
[en] Since the beginning of the industrial era, human activities have resulted in a significant increase in the concentration of atmospheric CO2. A part of this CO2 accumulates in the atmosphere, resulting in an increase of the natural greenhouse effect and of the temperature on the Earth's surface. This process is known as the climate change or global warming. Moreover, about 25 % of CO2 produced is absorbed by oceans. The dissolution of CO2 in seawater increases the concentration of protons and bicarbonate ions (HCO3-) and decreases the concentration of carbonate ions (CO32-). This results in a decrease in pH and in the calcium carbonate saturation state. All of these processes are called ocean acidification (OA). The pH of surface ocean waters has already fallen by 0.1 pH units since the beginning of the industrial era. This phenomenon is expected to intensify over the next century. According to average IPCC emission scenarios of future emissions of greenhouse gases, the average temperature of the surface ocean water is expected to increase by 2 to 4 °C and pH should decrease by 0.3 to 0.4 units by 2100.
Over the past two decades, many studies have highlighted the negative impact of OA on marine organisms. The first studies were mostly conducted in artificial environments and highlighted major impact on the physiology of organisms, mainly at the individual level. However, recent studies in the field have stressed the importance to conduct long-term experiments at the ecosystem scale, and in conditions closer to the natural environment. This kind of study can take into account ecosystem interactions and acclimation processes to better predict the direct, but also indirect, effects of the decrease of the pH of the ocean waters.
The persistence of tropical coral reefs is dependent on the rate of constructive processes (mainly through the reef-building coral calcification) which must exceed its (bio) erosion. On one hand, several studies have shown that the rate of reef-building coral calcification decreases with increasing pCO2. On the other hand, sea urchins are important bioeroders of coral reefs and thus contribute to the loss of reef material. However, sea urchins also prevent, by their grazing, the overgrowth of corals by algae which are favored by OA Therefore, the effect of elevated pCO2 on sea urchins and their bioerosive ability can be decisive for the future of tropical coral reefs during the century, especially where the density

of these bioeroders is important. Such a prediction is even more complex if one takes into account possible acclimation of different contributors at the long-term.
The aim of this work was therefore to evaluate the long-term effect of the increase of pCO2 expected in 2100 on the physiology and bioerosive activity of a key sea urchin in some coral reefs, Echinometra mathaei, using an artificial device which reproduces the coral reef ecosystem.
The first step was to set up an experimental tool which allows the maintenance at long- term (more than one year) of a simplified coral reef ecosystem in control condition and at the pH expected in 2100 while keeping other physico-chemical parameters identical and close to natural conditions (including in their daily variations). The developed system is composed of scleractinian hermatypic corals as reef builders, sea urchins (E. mathaei) as bioeroders and grazers and coral reef substrate with its diverse communities of algae, bacteria, archae, fungi and meiofauna. Daily variations in pH and temperature reproduce those measured in situ in the lagoon of La Saline, Reunion Island, from where some of the organisms originate. The average pH of control aquaria was successfully maintained at a mean of 8.09 ± 0.04, and the high pCO2 aquaria at 7.63 ± 0.02. The total alkalinity of the system was maintained between 2350 and 2450 μmol kg-1.
The impact of the OA expected in 2100 (pH 7.7) on the physiology of E. mathaei was studied at short-term (seven weeks). Sea urchins fed principally on algae that grow on the reef calcareous substrate, as in natural conditions. This study highlighted the resistance of this sea urchin to a moderate OA at short-term. Indeed, the growth and the metabolism were not significantly affected. This was associated to the ability of sea urchins to regulate the acid-base balance of their extracellular fluid, the coelomic fluid, through bicarbonate compensation.
The same experiment was then performed at long-term. The decrease in pH was induced gradually during six months until reaching a mean pH of 7.65, which was then maintained at this value during seven more months. The ability to regulate the acid-base balance of the coelomic fluid and the resistance of E. mathaei to OA was confirmed at long-term. Growth, metabolism and mechanical properties of the skeleton were not affected. This resistance appears to be related to the capabilities of acid-base regulation of E. mathaei, an apparently genetic trait. This resistance may also depend on the quantity and quality of the available
food (calcareous or not). We suggest that the bicarbonate ions involved in acid-base regulation are partly mediated by food.
Simultaneously to physiological measurements, the erosive activity of E. mathaei was measured. Results indicate that the rate of bioerosion triples under acidified conditions (pH 7.65). This increase is mediated by the increased activity of grazing of sea urchins and biological dissolution of the substrate, as the mechanical properties of the teeth of the sea urchins and coral skeletons do not appear significantly affected. We suggest that this increased bioerosion could have an impact on the dynamic balance between bioerosion and bioaccretion of corals and could determine the future of coral reefs where E. mathaei is the main bioeroder. It should be noted that the erosive activity of the sea urchin is associated with increased consumption of macroalgae that compete with corals and coralline algae, favoring the latter.
The results, according to those compiled from the literature, indicate that global change may cause a profound change in tropical reef ecosystems. Indeed, all major bioeroders studied so far seem to be resistant to global climate change and show an increase in erosive activity. For corals that currently show a low net ecosystem calcification, increased bioerosion could lead to net erosion and to the reduction and disappearance of the reef. However, prediction of the future of tropical coral reefs globally must take into account many parameters: acclimatization, tolerance/sensitivity and interactions of the various actors of the reefs. Other comparable studies to those carried out in the present work should be conducted in order to test these factors. The data obtained could then be used in the construction of a mechanistic model to implement, locally, reef conservation measures, in addition to the required massive reduction of atmospheric CO2 emission worldwide.
