Search

International audience; The largest area of intact vegetation in the Mascarene's archipelago is found on Reunion Island, but the remaining natural areas are under threats. Biological invasions are one of the main threats to biodiversity loss on this island and globally worldwide. This study aimed to quantify invasion degree by alien plants species in Reunion Island. This work was possible thanks to a large partnership between researchers and managers. All existing spatial data on invasion pattern were combined and completed by expert knowledge to develop the first 250 x 250 m map of invasion degree at the island scale. To fill the gaps where no field survey data or expert knowledge was available, we used a Random Forest model using nine climatic, landscape and anthropogenic variables. This model also provides a preliminary assessment of drivers of invasion at Reunion Island. Results showed that 85% of the extant native vegetation was invaded in different proportions; 38% are slightly invaded, 26% moderately invaded and 22% very heavily invaded. Despite the high levels of invasion in some places, more than 50% of the extant vegetation is not invaded or slightly invaded. Most of the invaded areas are located in the lowland and in the leeward coast although alien plants invade all types of vegetation from the coast to the top of the island. These results highlight a clear increase in the distribution of alien species over time. This study constitutes a key first step for about the ongoing prioritisation of management interventions on Reunion Island.

The largest area of intact vegetation in the Mascarene Archipelago is found on La Reunion Island. In 2007, a national park has been established where most protected habitats and rare species are targeted by some conservation actions. Invasions by exotic plants are the main threat to the biodiversity within this park and several management programs aim at controlling the main invasive alien plant species. For decades, inventories of alien species have been carrying out at La Reunion, constituting a large amount of data but these data never been synthetized, assembled and exploited to assess the level of invasion. Here, data from different organizations were combined to quantify the overall level of invasion by exotic plants island-wide. These data were extrapolated to grid cells of 250 x 250 m and categorized into four invasion categories ranging from intact to heavily invaded areas. Here, we present the spatial pattern of invasion on La Reunion Island. Only 12% of the island remain intact while 60%is moderately to heavily invaded. Inside the national park, more than 55% is lightly invaded or intact. We then discuss the invasion rate according to several factors: altitude, habitats, geomorphology, land cover and the presence of pathways or streams.

Biological invasions threaten biodiversity worldwide, especially on tropical islands such as Reunion Island. Here we present the results of a collaborative effort to map biodiversity priorities and to prioritise alien plant clearing. We first combined all existing data on habitats, threatened and endemic species and the distribution of alien plant species. Using expert-knowledge where necessary, we provide the first map of invasion level for the whole island. We developed a decision-making process using Zonation, a conservation-planning tool, to identify biodiversity conservation priorities. Conservation priority areas were selected to maximise biodiversity levels in areas lightly invaded. We then included criteria to identify priority areas for alien plant clearing based on clearing costs, accessibility and history of previous clearing programmes. Almost 40% of the remaining natural habitats on Reunion Island are moderately to heavily invaded. We identify 58,500 ha as conservation priority areas of which 30%are considered to be not invaded at all. We discuss how this will impact future alien plant clearing programmes and propose the basis for an integrated clearing and restoration plan for Reunion Island.

Conservation planning requires the identification of habitat that may support focal species, and an assessment of how to prioritise lands to protect sustainable populations. Because conservation plans often guide the allocation of limited resources in the face of habitat loss, there is little room for error in the prioritisation process. However, incomplete information on the distributions of species and habitats makes prioritisation an enormous challenge, particularly if surveys cannot be conducted due to land inaccessibility or other constraints. The use of predictive habitat distribution models by land managers in the conservation management of threatened species is therefore increasing. Reunion Island (western Indian Ocean) is one of the few islands at global scale holding two endemic species of petrels, the Barau's Petrel (Pterodroma baraui), and the Mascarene Petrel (Pseudobulweria aterrima), both of which are poorly known and endangered. The principal threats to both species are introduced mammalian predators (rats and feral cats) and light pollution. In this study, we focused on the very rare Mascarene petrel whose remained mysterious for 160 years. Thanks to the recent discovery of breeding sites in 2016, it is vital, to face conservation urgency, to quickly identify all potential breeding areas and engage actions at a broader scale. Thus, we used environmental parameters mixed with anthropic pressures (predators density and level of invasive plants) to investigate and model the distribution of suitable habitat of this rare and endemic species. Predictive distribution maps revealed habitat refuges suggesting a high sensitivity of the species to anthropogenic pressures. This approach is very helpful for long-term management to 1) identify priority conservation areas, 2) design concrete actions to be implemented according to the environmental / anthropic characteristics of the habitat.

Contexte : Les invasions biologiques constituent une des menaces principales de l'érosion de la biodiversité dans le monde et notamment sur l'île de La Réunion. Cette étude a pour but de coordonner et de spatialiser les enjeux de conservation, en rassemblant toutes les données existantes pour établir un état des lieux spatialisé à une échelle utile aux gestionnaires afin de prioriser les actions de gestion des plantes exotiques envahissantes. Une approche de priorisation spatiale basée sur un algorithme d'aide à la décision a été utilisée. Cette approche est participative et co-développée avec les gestionnaires afin de répondre à une demande du département de La Réunion. Approche : Nous avons rassemblés un ensemble solide de données quant à la distribution de la faune et de la flore indigènes, la répartition des habitats et leur niveau d'invasion. L'ensemble de ces données ont été insérées au sein d'un logiciel d'aide à la décision, Zonation, permettant l'identification des enjeux de conservation. L'utilisation de ce logiciel permet l'identification de priorités objectives et basées sur des concepts clés. Ainsi, les priorités identifiées mettent en avant les zones peu envahies, les habitats et espèces rares et sont représentatives de l'ensemble de la biodiversité. Une fois les enjeux de conservation identifiés, il est possible de définir les priorités d'actions de lutte contre les plantes exotiques. Pour cela, trois critères opérationnels ont été utilisés : l'accessibilité de la zone, les priorités d'intervention et l'historique des chantiers. Résultats : Ces analyses nous ont permis d'identifier 58 500 ha d'enjeux de conservation parmi lesquels 30% sont considérés comme non envahis, ce qui implique des actions de surveillance. A l'inverse, 53 714 ha de zones potentielles d'intervention ont été identifiées. Parmi ces zones, 4392 ha sont considérés comme étant de forte priorité. Trois résultats principaux y sont présentés : la carte du degré d'invasion, les enjeux de conservation et les priorités d'actions de lutte. Les données spatiales sont téléchargeables sur le site du GEIR : https://www.especesinvasives.re/en-pratique/documents-et-outils/article/documents-et-outils