An inventory has been made of the decapod fauna of the Glorieuses Islands, western Indian Ocean (WIO), following the BIORECIE 2 Expedition to the Islands, 5–17 December 2012. Field data are complemented by a review of taxonomic studies for these islands. Overall 157 species are now reported from the Glorieuses Islands, including 61 new records. The presence of the coconut crab, Birgus latro, is confirmed from these islands, for the first time since 1884, and the WIO distribution of this endangered species is updated, based on observations made in the region since 2006.

In many atolls of the South Pacific, a shift has occurred in the past couple of decades from traditional to more intensive fishing practices. Increasing fishing pressure on coral reefs raises the need for long-term studies to understand how fish communities react to fishing practice changes. The objective of this study was to analyse the variations of reef fish communities (in terms of species richness, diversity, density and structure) during a long time lapse at Tikehau atoll (Tuamotu archipelago, French Polynesia). The same eight lagoon pinnacles were sampled in 1987 and 2003 at 6 and 12 m depth. Quantitative data were collected by visual census techniques (UVC) on 50 m × 5 m belt transects. Analyses were conducted using a multiple spatial scale (depth, pinnacle and whole lagoon) approach with a particular attention on commercial species such as Scaridae (parrotfish), Acanthuridae (surgeonfish), Lutjanidae (snapper), Lethrinidae (emperor), Serranidae (grouper). Despite an increased sampling effort between 1987 and 2003, the mean species richness per transect decreased significantly from 26.2 ± 6.2 to 21.6 ± 9.6. The mean diversity varied similarly. A stronger decrease happened at pinnacles close to the village, where fish density also decreased. Conversely, fish density, species richness and diversity increased at pinnacles less visited by inhabitants. The community structure shifted from commercial species to small site-attached species e.g. Pomacentridae (damselfish), Ptereleotridae (dartfish) as fishing reduced the abundance (and biomass) of targeted stocks. We argue that these spatiotemporal variations resulted from a drastic change in fishing practices over the 16 years period, that shifted a sustainable fishery using traditional fish traps to an unbalanced, species-threatening, selective fishery. This study underlines the need for management and for the implementation of marine protected areas (including no-take zones) in order to protect the coral reef ecosystem and favor sustainable fisheries at Tikehau atoll.

An unusual mass settlement event of the grouper Epinephelus merra occurred on the fringing coral reefs of Reunion Island (south-western Indian Ocean) in April 1994. This major event coincided with stormy conditions associated with the new moon period. The abundance of new settlers was surveyed for 45 days following settlement on both the reef flat and the back reef area of the coral reefs of St Gilles, La Saline and St Leu. Mean densities of settlers were higher on reef flats (maximum 474 ind/20 m2 at La Saline) than on back reef areas (maximum 156 ind/20 m2 at La Saline). Seven weeks later, densities of settled E. merra had drastically decreased, and did not differ with site and zone (18.5 ind/20 m2). Percentages of post-settlement mortality at that period ranged from 85 to 88% in back reef areas and from 93 to 95% on reef flats. A strong correlation between density of recruits and per capita mortality suggests density-dependent early post-settlement mortality for E. merra. The limitation of food resources and shelters, perhaps combined with an increase of diseases due to high population densities of settlers, was probably the main cause of mortality just after the mass settlement. Predation by reef fish, including cannibalism between settlers, is probably responsible for most subsequent mortality.

This article aims to review 1) the major and most frequent human-induced physical disturbances and their consequences on coral reef habitats using a multi-scale approach, and 2) the scale-related indicators and conceptual aspects used to detect and measure the effects of these physical impacts. By physical disturbances, we mean direct perturbations that lead to the destruction/erosion of the carbonate framework. Human-induced direct physical disturbances are numerous from coastal development, tourism, harvesting, accidents and nuclear/weapon testing. Since methods for monitoring and measuring indicators are generally scale-implicit, coral reefs are first presented according to different ecological-spatial scales of organization, from colony to region (colony, reefscape, reef zone, whole reef, island and region). In this way, it is easier to link a couple {habitat, disturbance} to their potential indicators and to the descriptors they target. Three classes of descriptors, related to the response of the living component of coral reef ecosystem, are considered here: stony coral, reef fishes and the human uses. A synthesis of the different options for coral habitat assessments is proposed. We sort them according to their objectives (monitor, initial status or improvement of knowledge), their specificities (identification or not of a specific disturbances) and their scale of investigation (small, meso- or large scales). Usually, the majority of the indicators of human-induced disturbances are non-specific. They reveal that something is happening but not the actual causality and can only detect differences across time or space. A major weakness lies in the difficulty in deconvoluting the signals from a conjunction of stressors occurring at different scales. As such, a hierarchical concept of disturbances in coral reefs would be the next logical step to enhance our capabilities in monitoring and forecasting coral reefs status.