Abstract Inbreeding depression (the reduction in fitness associated with inbreeding) has been demonstrated in a wide range of animals, but despite its ubiquity, is not an inevitable consequence of inbreeding. As a result, there is uncertainty about the extent to which inbreeding depression poses an ongoing risk to endangered species currently experiencing significant demographic recovery. Quantifying inbreeding depression will be critical if we want to understand these risks. A comprehensive quantification of the fitness costs of inbreeding requires detailed individual-based longitudinal data so lifetime impacts can be assessed. Here, we use an extraordinarily detailed long-term dataset on Mauritius kestrels ( Falco punctatus ) to explore inbreeding depression in a population currently experiencing significant demographic recovery. To do so, we constructed a social pedigree of 1,758 individuals and combined this with 1,240 nest records and 1,411 individual resighting histories to explore lifetime fitness effects over a 30-year period. Inbreeding increased significantly over time as the population recovered before stabilising. Inbred eggs were less likely to survive to fledging. Inbred adult male and female birds had significantly lower annual reproductive success than outbred individuals because of a lower annual egg-to-fledgling survival probability. This resulted in significantly lower lifetime reproductive success in inbred females but not males, which showed a negative trend. Population growth was negative and extinction risk increased slightly at current levels of inbreeding. These impacts will become more severe should inbreeding levels increase in the future, which is highly likely given ongoing genomic erosion. Taken together, our results demonstrate significant fitness costs associated with inbreeding in Mauritius kestrels, which pose an ongoing risk to population viability. This suggests that monitoring and managing inbreeding risks in endangered species will likely be required even in populations that are showing significant demographic recovery in response to conservation interventions.

ABSTRACTGame bird species were introduced to several insular systems during the 18th and 19th centuries. The Mascarenes are a relevant example of such introductions, including the grey francolin (Francolinus pondicerianus). In the case of Mauritius and Réunion Island, although this bird has been present for more than 300 years, limited information is available concerning its behavioural adaptations in insular systems, especially regarding its feeding behaviour. The first aim of this study was to describe the diet of the grey francolin, a well‐established game‐bird species in the Mascarenes. From this novel information, we also wish to discuss how the grey francolin can be positioned in insular food chains, and thus, its potential impact on the biodiversity of Mauritius and Réunion Island. We provide an analysis of 9115 food items, including both insects and seeds harvested from stomach contents of 13 grey francolin. The birds were harvested in Mauritius, from two distinct zones, and their stomach contents were sorted and identified using binocular magnifiers. The results revealed a total diversity of 20 seed species and 23 insect species, with a majority of exotic species. Solanum sp., Amaranthus sp. and Chamaesyce hirta were the dominant seeds, whereas insect food items were widely dominated by ant species (Pheidole megacephala, Solenopsis geminata). Important variability was observed in the composition of stomach contents, resulting in relevant amplitudes of species‐diversity indicators and the animal‐versus‐plant ratio, throughout the 13 birds. Furthermore, the amount of certain food items was particularly high for one stomach content, whereas their global occurrence among the 13 birds was rather low, suggesting that the grey francolin has a versatile and opportunistic diet. Despite the lack of monitoring of food availability or diversity, the variation in stomach contents suggests that the diets of these generalist birds are dependent on the availability of resources in the environment.

The radiated tortoise and Aldabra giant tortoise were used as ecological replacements for the extinct Mauritian tortoise Cylindraspis inepta. Aldabra giant tortoises were better adapted as seed dispersers and grazers, and have become the species of choice. It is intended to remove the radiated tortoises from Round Island. It is suggested there was an open grazing climax plant community maintained by tortoises with heliophilous native plants that have adaptations to avoid being grazed or browsed and respond to grazing with prostrate growth form. The Aldabra giant tortoises fed on fallen fruits and spread the seeds of hardwood trees, screw-pines and palms in their droppings that subsequently demonstrated enhanced germination and growth rates. Preliminary work was done on Ile aux Aigrettes before the release on Round Island. The tortoises have established grazed areas colonised by the native tussock grass Chrysopogon argutus that the tortoises do not graze and is benefiting from reduced competition from the fast-growing non-native grasses.

The Mauritius Kestrel Falco punctatus, once the rarest kestrel worldwide, became an icon of bird conservation after it recovered from four to six individuals in 1974 to some 800 by 2005 following intense conservation management. Its population however then halved within about a decade prompting a re-evaluation of the IUCN status and up listing of the species in 2014 and an increased conservation attention. Drivers of this new decline are unclear and the influence of habitat structure and diet on breeding success may be important contributors but have received relatively little attention, particularly in the way they may interact to influence production of new fledglings. We address this knowledge gap by studying whether breeding success is influenced by habitat structure (in terms of cover of the invasive Ravenala in native habitats, an alien plant causing strong structural shift in the forests that it invades, and extent of cleared area), diet composition and food pass frequency (as a proxy for food intake) and food quality at 28 nests of a re-introduced kestrel population in south east Mauritius during the 2015–2016 breeding season. The kestrel's diet comprised native and alien birds, reptiles, insects, and small alien mammals, with a disproportionately high proportion of Phelsuma gecko. A higher frequency of food provisioning and percentage cover of Ravenala both contributed to higher breeding success. Ravenala may increase gecko density or increase gecko detectability and predation by the kestrel, or both, while changed land use (pasture and sugar cane fields) may increase prey diversity in the form of non-forest prey known to be eaten by Kestrels (e.g. alien agamids, small mammals and birds). These prey related influences on breeding suggest that the Bambou mountain range provides a human-generated novel ecosystem altering food availability and increasing the kestrel's breeding success. However, Ravenala is an invasive alien species harmful to the wider forest biodiversity. Progressive weeding of Ravenala and concurrent re-introduction and augmentation of native palms and Pandanus species which geckos can use at comparable densities to Ravenala, is recommended. This would likely improve the kestrel's hunting habitat quality and maintain high gecko density or detectability and the vegetation structure required for hunting manoeuvrability and prey availability without the negative consequences of invasive Ravenala.

Recovery programmes for endangered species can become increasingly demanding over time, but managers may be reluctant to change ongoing actions that are believed to be assisting recovery. We used a quantitative risk assessment to choose support strategies for a reintroduced population of Mauritius olive white-eyesZosterops chloronothos. Facing increasing costs, managers considered changing the ongoing supplementary feeding strategy, but at the same time worried this could jeopardize the observed positive population trend. We used a feeding experiment to compare the current feeding regime and a cheaper alternative (a simple sugar/water mix). Results suggested the cheaper alternative would only marginally reduce population vital rates. We assessed the influence of these results and the associated uncertainty on population recovery and management costs using two decision-analytic criteria, incremental cost-effectiveness ratio and stochastic dominance. The new feeding regime was expected to be, on average, more cost-effective than thestatus quo. Moreover, even negative outcomes would only likely mean a slower growing population, not a declining one, whereas not changing feeding regime actually entailed greater risk. Because shifting from the current regime to a cheaper sugar/water mixture was both a risk-averse and a cost-effective choice, we decided to implement this change. Four years after the experiment, the population continues to grow and costs have been contained, matching predictions almost exactly. In this case, the field experiment provided useful empirical information about prospective actions; the risk analysis then helped us understand the real implications of changing the feeding regime. We encourage managers of recovery plans facing a similar situation to explicitly recognize trade-offs and risk aversion, and address them by combining targeted research and formal decision analysis.

Conservation translocations are commonly used in recovery programmes for threatened species from a wide range of taxa but outcomes can vary considerably both within and between programmes, and the causes of success or failure are often unclear. Central to understanding translocation success is the implementation of an accompanying monitoring programme, enabling the drivers of population establishment and persistence to be explored within a population ecology framework. Here we review and assess the outcome of a translocation programme for the Endangered Mauritius Kestrel, which involved the initial translocation of captive‐reared Kestrels into four isolated populations and long‐term nest‐site management and monitoring. We show that after 20 years these four populations have different recovery trajectories including local extinction, recent decline and comparative stability. We explore the demographic drivers behind these trajectories and how they have been influenced, and could potentially be manipulated, by conservation management actions. Metrics of breeding performance differed between populations and in part this was driven by nest‐site selection, with Kestrels nesting more frequently, laying larger clutch sizes and rearing larger broods in nestboxes. We found no compelling evidence for inter‐population variation in survival rates. Simulating population trajectories under a range of conservation management scenarios, including further conservation translocations or a scaling up of nest‐site management, suggested that the latter would be a more effective, practical long‐term solution for the population currently in decline. Our findings provide valuable insights into the merits of monitoring, population demographic reviews and the challenges associated with identifying and mitigation for the drivers of rarity in threatened species.

In our case study on the critically endangered Mauritius olive white-eye Zosterops chloronothos, we demonstrated how difficult decisions about ongoing support of actively managed species can be approached by carefully combining experiments and risk analysis (Ferri ere et al., 2020). The thoughtful commentaries by Converse & Sipe (2021) and Stojanovic & Crates (2021) offer opportunities for further insights. When we conceived the study described in Ferri ere et al. (2020), the conservation management we were carrying out for the olive white-eye had become unsustainable. Adding more feeding stations would exceed staff capacity and budget. It became clear that if, as we believed, supplemental feeding was essential for the population to grow, we had to find a way to make the workload sustainable, as well as finding a potential long-term minimal management strategy. This should be the aim of most conservation projects which are limited by funding and must look for ways to reduce project costs without jeopardizing species recovery; that is, to find a solution to what is often believed to be a zero-sum game as described by Converse & Sipe (2021). Facing this problem, we turned to the experiences of managers elsewhere in the world who dealt with similar issues for similar species. In particular, the solution to our problem was inspired by those used to help the recovery of the hihi Notiomystis cincta in New Zealand (Chauvenet et al., 2012). In our case, our experimental approach was limited by uncertainty and our willingness to take calculated risks. We were unable (and unwilling, see below) to use a control group with no feeding, because we felt the risk to the population and the species would be too great; although the Ile aux Aigrettes population is increasing, the one on mainland Mauritius is still decreasing (Maggs et al., 2015). Moreover, previous experiences suggest that some drastic management decisions can have immediate impacts (Armstrong, Castro & Griffiths, 2007). A catastrophic decline in the population might leave us no time to react. Therefore, we had to get our decisions right; a trial might help, but within limits. By creating alternatives, articulating uncertainty and clearly recognizing our objectives, including how we see benefits and risks, we effectively embraced those limitations, took control of the decision process and made room for success (Converse & Sipe, 2021). Our approach in Ferri ere et al. (2020) underlines two key steps for success in changing a conservation management strategy, which highlight both the challenges and opportunities pointed out by Converse & Sipe (2021) and Stojanovic & Crates (2021). The first step was to think openly about uncertainty. Endangered species may face a wide variety of interacting threats. Small populations also mean small sample sizes and greater chances of extreme events, making inference difficult, compounded by species traits that reduce observability. We were able to monitor our study species closely, thanks to well-known territories and previous ringing efforts. As Stojanovic & Crates (2021) point out, other programs may not have such advantages; however, when uncertainty is high, learning can be harder (e.g. if there are multiple confounding factors) but it may also be easier (e.g. if it is easier to shift current beliefs). In such cases it is even more critical to articulate uncertainty very clearly: where does it come from, how does it affect our decisions, how can we reduce it? This can be a fundamental challenge to the assumption, often implicit in conservation, that more data are always better. A useful method we increasingly turn to for decisions on monitoring is calculating the value of information provided by more data (Canessa et al., 2015; Canessa et al., 2020). The second vital step was to be clear on how we felt about uncertainty. We agree with Stojanovic & Crates (2021) that risk aversion is common in conservation programs, but it is not in itself irrational, particularly where worst-case scenarios may involve the extinction of an entire species. In our case, ongoing vegetation restoration efforts and our ability to closely monitor the species, and respond if necessary, led us to believe that even worst-case outcomes would not be as catastrophic as in an immediate crisis phase (Stojanovic & Crates, 2021). Given worse possible

Emerging Infectious Diseases (EIDs) are recognised as global extinction drivers of threatened species. Unfortunately, biodiversity managers have few tested solutions to manage them when often the desperate need for solutions necessitates a response. Here we test in situ biosecurity protocols to assess the efficacy of managing Psittacine beak and feather disease (PBFD), one of the most common and emergent viral diseases in wild parrots (Psittaciformes) that is currently affecting numerous threatened species globally. In response to an outbreak of PBFD in Mauritius “echo” parakeets (Psittacula eques), managers implemented a set of biosecurity protocols to limit transmission and impact of Beak and feather disease virus (BFDV). Here we used a reciprocal design experiment on the wild population to test whether BFDV management reduced viral prevalence and viral load, and improved nestling body condition and fledge success. Whilst management reduced the probability of nestling infection by approximately 11% there was no observed impact on BFDV load and nestling body condition. In contrast to expectations there was lower fledge success in nests with added BFDV biosecurity (83% in untreated vs. 79% in treated nests). Our results clearly illustrate that management for wildlife conservation should be critically evaluated through targeted monitoring and experimental manipulation, and this evaluation should always focus on the fundamental objective of conservation.

The provision of supplementary food is widely used in the management of endangered species. Typically, food is provided ad libitum and often without a planned exit strategy, which can be costly. The role supplementary food plays within population demography can be challenging to identify and therefore any reduction must be carefully considered to avoid negative impacts. Here we investigate the role supplementary food plays within a reintroduced population of a Critically Endangered passerine species by quantifying its use alongside intrinsic and extrinsic factors. Specifically, we illustrate how the provision of supplementary food could be refined in response to breeding stage and the time of food provisioning and, via habitat restoration, create a long-term exit strategy based on influential plant species. The consumption of supplementary food increases during energetically expensive phases of the breeding cycle, during the morning provision of food and when natural plant resource availability is low. We also show a pattern whereby supplementary food could act as a buffer during periods of low natural resource availability during breeding. Based on these findings short-term management could take a reactive approach; refining supplementary food supply in response to breeding stages of pairs and potentially removing the provision of food in the afternoon. In the long-term key plant species, found to correlate with a decrease in supplementary food consumption, could be incorporated into habitat restoration efforts which could create a continuous natural food supply and contribute to creating a self-sustaining population and a potential exit strategy.

When making decisions about reintroducing a species, practitioners need to consider whether the release site contains habitat suitable for those species, whether past extinction drivers have been remedied and whether reintroduction is the best option for the species to recolonize the release site. These concerns are captured within two paradigms; the habitat and metapopulation paradigms. We use cost‐distance analysis to assess the need for reintroduction of two bird species, Rodrigues Fody and Rodrigues Warbler, to Anse Quitor reserve on Rodrigues Island, testing hypotheses based on these underlying paradigms. Given a lack of detailed field studies of dispersal across the landscape on either species we rely on expert judgment. Our results show that experts believe Rodrigues Fody will naturally colonize Anse Quitor but that Rodrigues Warbler may not, at least within a time frame of 10 years. This information and treatment of expert judgment allows greater justification in reintroduction planning. Our method shows one way to assist in reintroduction decision making in poorly studied systems.

The Mascarene Islands (Mauritius, Rodrigues, and Réunion) of the western Indian Ocean are home to a diversity of endemic flora and fauna, often linked through close evolutionary and ecological relationships. Owing to a turbulent history of anthropogenic exploitation, contemporary research and habitat restoration efforts are underway on Mauritius and numerous surrounding islets, notably Round Island. The Round Island stick insect, Apterograeffea marshallae (Phasmatodea: Platycraninae), was described in 2002 and information about the species’ biology has been wanting. In this short communication we confirm the importance of the endangered Blue Latan Palm (Latania loddigesii) in the diet of the Round Island stick insect and include observations on reproduction and predation of this insect species. These observations are relevant for informing ongoing habitat restoration and conservation efforts in the Mascarenes, as well as the biology of the stick insect A. reunionensis on Réunion. Despite their roles as prey, predators, and pollinators, insects are disproportionately overlooked as conservation priorities. We suggest that conservation projects would greatly benefit from directed study of oft-neglected insects to supplement knowledge of species biology, reconstruct ecological histories, re-establish lost ecological interactions, and restore integrity of ecological communities.

The diversity of introduced mammals and their introduction history varies greatly across the Western Indian Ocean (WIO) islands, from ancient introductions in the past millennia on islands off the East coast of Africa where extant terrestrial native mammal communities exist, to very recent invasions in the past decades on islands in the Mascarene archipelago. We compile the distribution of 16 introduced mammal taxa on 28 island groups comprising almost 2000 islands. Through an exhaustive literature review and expert consultation process we recorded all mammal eradications, and species recoveries which could be attributed to introduced mammal eradication or control. All island groups have been invaded by mammals, and invasive cats and rats in particular are ubiquitous, but cultural contingency has also led to regional invasions by other mammals such as lemurs, civets and tenrecs. Mammal eradications have been attempted on 45 islands in the WIO, the majority in the Seychelles and Mauritius, and where successful have resulted in spectacular recovery of species and ecosystems. Invasive mammalian predator eradication or control in association with habitat management has led to improved conservation prospects for at least 24 species, and IUCN red-list down-listing of eight species, in the Mascarene Islands. Future island conservation prioritisation in the region will need to take account of global climate change and predicted sea-level rises and coastal inundation. Greater investment and prioritisation in island conservation in the region is warranted, given its high biodiversity values and the extent of invasions.

Conservation projects have limited resources and an ever expanding to-do list, presenting a significant leadership challenge. Mid-term evaluations can be informative tools to check how short term activities and resources are achieving long term conservation outcomes. This research study involved a programme evaluation of a successful species recovery project in Mauritius, using a systems-thinking approach to conservation management, and utilising a Theory of Change to assess the effectiveness of short term activities on long term impacts. This systematic method of evaluation gave greater clarity on resource planning, performance indicators and supporting processes, with observations that could be incorporated into ongoing plans. Such an approach could be used by funding organisations or by local management teams to review project performance without the need for a comparator, extensive benchmark data, nor a prescriptive management standards framework.

Responding to allospecific territorial signals may result in the unnecessary expenditure of energy and time and is therefore, in general, not adaptive. The signals of closely related species are often similar owing to common ancestry, especially when species evolve in allopatry. However, selection for species recognition in sympatry tends to result in distinct territorial signals. The endangered pink pigeon, Nesoenas mayeri, endemic to Mauritius, occurs in sympatry with the invasive Madagascan turtle dove, Nesoenas picturata, and their perch-coos are remarkably similar. We conducted playback experiments to test whether pink pigeons respond to coos from Madagascan turtle doves. Pink pigeons responded similarly to playback of conspecific and Madagascan turtle dove coos but not to those of the more distantly related and sympatric Stigmatopelia chinensis. This study stresses the need to identify the impact of N. picturata on the breeding success and conservation of N. mayeri. Crown Copyright (C) 2014 Published on behalf of The Association for the Study of Animal Behaviour by Elsevier Ltd. All rights reserved.

Invasive species are a major threat for island biodiversity, causing species decline and extinction globally. Of all invasive mammals rats are one of the most detrimental and have been the target of numerous control and eradication programmes. In Mauritius rats have contributed to the extinction of 50% of the island's fauna and are thought to be the main threat to the endemic Mauritius olive white-eye (Zosterops chloronothos), a critically endangered passerine. Assessing the impact of rats and suitable control strategies is often problematic in such cases because of the lack of replicate populations for experiments. Here, we illustrate how to overcome this issue by combining a small-scale rat management experiment on olive white-eyes with demographic models that provide estimates of the potential effects of management on vital rates and population growth. We established poison and trapping grids within breeding territories, and show that rat management significantly decreased rat abundance and increased nesting success. An individual-based stochastic simulation model suggested that rat control could produce a 5–6 fold increase in the annual productivity of female olive white-eyes, which in turn would be sufficient to stabilise population growth. In the absence of rat management, our analysis suggests the olive white-eye population will decline by about 14% per annum. By combining low cost field experiments with widely available demographic models we highlight the value of targeted, effective rat management techniques for both short and long-term population management in threatened passerines.

Global and project-level biodiversity indicators have received considerable attention, but indicators of the conservation actions and impacts of programmes and institutions appear to be under-developed. The IUCN Red List Index (RLI) has potential to be a useful indicator at an organizational-level to evaluate long-term impact of conservation on the extinction risk of species, thereby supporting institutional decision-making and communications. However, it has not yet been tested for its utility in tracking changes in extinction risk of a set of species targeted specifically by an individual conservation agency. Here, we examine the feasibility of using the RLI as one metric of the conservation impact of the Dun-ell Wildlife Conservation Trust, a conservation charity which runs multi-decadal programmes on a modest number of globally threatened terrestrial vertebrate species. Of 17 target amphibian, bird and mammal species, eight underwent improvements in Red List category (reductions in extinction risk) owing to conservation. This drove a 67% increase in the value of the Red List Index between 1988 and 2012. This contrasts with a 23% decline in a counterfactual RLI showing projected trends if conservation had been withdrawn in 1988. For organizations that target sets of species with circumscribed geographic distributions and that are regularly assessed by the IUCN Red List, the RLI is a useful indicator for measuring and demonstrating long-term conservation impact to technical and non-technical audiences. (C) 2014 Elsevier Ltd. All rights reserved.

The extinction of large herbivores, often keystone species, can dramatically modify plant communities and impose key biotic thresholds that may prevent an ecosystem returning to its previous state and threaten native biodiversity. A potentially innovative, yet controversial, landscape-based long-term restoration approach is to replace missing plant-herbivore interactions with non-native herbivores. Aldabran giant (Aldabrachelys gigantea) and Madagascan radiated (Astrochelys radiata) tortoises, taxonomically and functionally similar to the extinct Mauritian giant tortoises (Cylindraspis spp.), were introduced to Round Island, Mauritius, in 2007 to control the non-native plants that were threatening persistence of native species. We monitored the response of the plant community to tortoise grazing for 11 months in enclosures before the tortoises were released and, compared the cost of using tortoises as weeders with the cost of using manual labor. At the end of this period, plant biomass; vegetation height and cover; and adult, seedling, flower, and seed abundance were 3-136 times greater in adjacent control plots than in the tortoise enclosures. After their release, the free-roaming tortoises grazed on most non-native plants and significantly reduced vegetation cover, height, and seed production, reflecting findings from the enclosure study. The tortoises generally did not eat native species, although they consumed those native species that increased in abundance following the eradication of mammalian herbivores. Our results suggest that introduced non-native tortoises are a more cost-effective approach to control non-native vegetation than manual weeding. Numerous long-term outcomes (e.g., change in species composition and soil seed bank) are possible following tortoise releases. Monitoring and adaptive management are needed to ensure that the replacement herbivores promote the recovery of native plants.

Aims The use of exotic species as taxon substitutes to restore lost ecological interactions is currently hotly debated. Aldabrachelys gigantea giant tortoises have recently been introduced to three islands in the Mascarene archipelago (Ile aux Aigrettes, Round Island and Rodrigues) to resurrect herbivory and seed dispersal functions once performed by extinct giant tortoises. However, potential unintended impacts by frugivore substitutes on native ecosystems, e.g. whether they will facilitate the germination of exotic plant species, are largely unknown. We investigated whether A. gigantea introduced to Rodrigues in 2006 could enhance the germination percentage of four widespread fleshy-fruited exotic species on the island. Using germination trials to forecast unintended impacts that could arise from the introduction of a frugivorous taxon substitute enables conservation managers to limit potential adverse negative interactions before they occur. Methods In germination trials that ran over 4 months, we investigated the effects of ingestion (gut passage and deposition in faeces) by subadult and adult A. gigantea on the germination percentage of four exotic fleshy-fruited plant species introduced to Rodrigues. We fed fruits of these plant species to sub-adult and adult A. gigantea to test how variation in age and size of the frugivore would affect seed germination. Feeding of distinctly coloured plastic pellets together with the fruits allowed us to test for individual tortoise effects on seed germination. Important Findings Ingestion by A. gigantea increased the percentage of seeds germinating of Mimusops coriacea and Lantana camara ,b ut not percentage of germination of Veitchia merrillii or Wikstroemia indica. Seeds were more likely to germinate following ingestion by sub-adult rather than adult tortoises, which may be a consequence of the shorter gut passage time observed for sub-adults. Our results demonstrate that introduced frugivorous taxon substitutes could facilitate germination of exotic and invasive plants and highlight the need for conservation managers to weigh the risk of taxon substitutes potentially facilitating the germination and recruitment of exotic fleshy-fruited plants against the benefit of restoring lost seed dispersal functions of threatened indigenous plants. Our findings also highlight the importance of considering age and size variation in frugivores, in particular in long-lived taxa such as giant tortoises, when studying ingestion effects on the germination performance of plants.