|Scientific Name:||Pteropus livingstonii|
|Species Authority:||Gray, 1866|
Pteropus livingstonei Andersen 1912
|Taxonomic Notes:||P. livingstonei was used on the 2000 Red List. CITES also uses P. livingstonei, and livingstonei should be the grammatically correct spelling. T. Hutson (Chiroptera Red List Authority) notes that the correct spelling should be livingstonii (also by Simmons in Wilson and Reeder 2005). Andersen’s 1912 emendation of the species name is now considered unjustified (Smith and Leslie 2006).|
|Red List Category & Criteria:||Critically Endangered A2c ver 3.1|
|Assessor(s):||Sewall, B.J., Young, R., Trewhella, W.J. & Rodríguez-Clark, K.M. and Granek, E.F.|
|Contributor(s):||Glendewar, G., Dawson, J.S., Doulton, H., Masefield, W., Halidi, A.B.A., Little, S., Reason, P.F., Tidemann, C.R., Mickelburgh, S. & Hutson, A.M. and Bergmans, W.|
Listed as Critically Endangered because of a serious population decline suspected from catastrophic habitat decline caused by cutting of trees for fuelwood and construction, and by conversion of all but the steepest upland areas to agricultural use; extensive declines in area of occupancy, extent of occurrence and quality of habitat.
True population change in P. livingstonii is unknown and data on habitat change over time is limited. However, the species qualifies for CR under the A2c criterion because best estimates indicate habitat loss has exceeded 80% over the past three generations (estimated generation length: 8.1 years/generation), because remaining habitat is increasingly degraded and fragmented, and because declines in the extent and quality of habitat are continuing.
P. livingstonii also meets the Endangered threshold under a second criterion, B1. The geographic range is small, such that the current extent of occurrence is estimated to be 1,856 km², less than the EN threshold of 5,000 km² under criterion B1. In addition, the habitat is severely fragmented, and there have been clear, continuing, observed declines in extent of occurrence, area of occupancy, habitat quality, and number of locations. Declines in number of mature individuals are also inferred from these declines in habitat. Therefore, this species would also be listed as EN under criterion B1ab(i, ii, iii, iv, v).
|Previously published Red List assessments:|
|Range Description:||P. livingstonii is endemic to the Union of the Comoros, where it is only found on the islands of Anjouan and Mohéli. The species roosts and forages primarily in native forest and underplanted forest, and avoids areas heavily affected by human disturbance. On Anjouan the species avoids all lower elevations on the island (lowest record of feeding is 300 m asl), but on Mohéli some limited forest patches extend to near sea level (lowest record of feeding is 40 m asl). However, nearly all lower-elevation native forests have been lost, and in the past five years, there have been no records of this species below approximately 450 m on Anjouan or below 200 m on Mohéli (Daniel et al. in press.).|
The Extent of Occurrence (calculated as the minimum convex polygon around area above minimum observed altitude above sea level on each of the two islands) is estimated as 1,856 km². This includes unsuitable areas of land and ocean located between patches of suitable habitat.
The total land area of Anjouan and Mohéli is a combined 635 km² (Louette et al. 2004). However, not all this area is occupied by P. livingstonii. The species is found within a combined area on the two islands ranging from 99.1 km² (minimum convex polygons around all known currently occupied long-term roost sites: Anjouan = 90.2 km² Mohéli = 8.9 km²) to 462.5 km² (minimum convex polygons around all potential but unconfirmed foraging and roosting areas in native forest, degraded forest and agroforestry habitats above the lowest elevation of recent sightings on each island [450m-1050m asl on Anjouan, and above 200m asl on Moheli]: Anjouan = 326.9 km², Mohéli = 135.6 km²) (Daniel et al. in press.). These two figures represent the possible range of this species on the two islands it inhabits. This represents a substantial decline over the past five years.
A land cover classification conducted in 2012 using high resolution satellite imagery and exenstive ground-truthing (ECDD, BCSF and Durrell, 2013) found the combined area of native forest (i.e. forest with little trace of human impact and a closed canopy) on both islands was 54.7 km² (29.6 km² in Anjouan; 25.1 km² in Moheli). The combined area of native forest and degraded forest (i.e. forest either underplanted with crops, or with localised logging or clearance for fuelwood) on both islands was 143.1 km².
Within the smallest total range on both islands (the area around all known currently occupied long-term roost sites), the Area of Occupancy or surface area of potentially occupied habitat (i.e. native and degraded forest and agroforestry habitats that could be used for either roosting and foraging) is estimated to be 65.0 km² (Anjouan = 56.8 km², Moheli = 8.2 km²). Within the largest total range on both islands, the equivalent estimate for Area of Occupancy is 188.4 km² (Anojuan = 114.07 km², Moheli = 74.34 km²). Thus, the estimated total Area of Occupancy for this species is 65.0 - 188.4 km².
|Range Map:||Click here to open the map viewer and explore range.|
|Population:||Repeated simultaneous surveys of all 23 known roosts during 1998-2006 typically recorded about 1,200 bats (Sewall et al. 2007); many of these roosts were located as a result of the implementation of a national monitoring program (Trewhella et al. 2005). Surveys have been conducted only sporadically since 2007, so it is unclear whether populations have changed recently. However, a survey conducted with sequential visits to all previously-known roosts in 2011 and 2012 estimated 1,300 bats across 22 roosts (Anjouan: 940 bats at 16 roosts, Mohéli: 360 bats at 6 roosts) (Daniel et al. in press.). Colony size typically ranges from 15 to 150 individuals (Sewall et al. 2007). The largest known roosts have sometimes reached about 250 individuals during the rainy season, but not all of these are mature individuals (Sewall et al. 2011 a,b).|
Generation length for the wild population of P. livingstonii is unknown, but several demographic parameters can be derived from the studbook for captive bats of this species (Glendewar 2014). Bats of this species can be long-lived in captivity (e.g., one founder captured as an adult in 1992 is still alive and has now lived 22 years in captivity). Females do not reproduce during the first few years of life in captivity (minimum dam age at first reproduction = 3.4 years, mean dam age at first reproduction = 5.9 years), and the average age of females giving birth is elevated (of all dams born in captivity that have since died or are >10 years of age, mean age when giving birth = 8.1 years). Thus, based on studbook data (Glendewar 2014), this last figure of 8.1 years represents our best estimate of the generation length of the captive population. This estimate may be conservative as most (75%) of these females are still alive and at least some presumably will reproduce again at a later age.
It is unclear to what extent the generation length of the captive population correlates with the generation length of the wild population of P. livingstonii. However, the generation length of captive P. livingstonii correlates well with the few published generation length values for wild populations of other Pteropus species. In a wild population of P. poliocephalus, generation length was estimated conservatively as 7.4 years (Tidemann and Nelson 2011), though the true generation length could be much higher. Further, a generation length of 5.0 years was estimated for a rapidly declining wild population of P. conspicillatus (Fox et al. 2008); however, our calculations using these same data suggest that in a stable population, generation length for P. conspicillatus would be about 7.7 years. Thus, generation length values for the captive population of P. livingstonii correlate well with the only published generation length values for wild Pteropus. Further, an estimate derived from taxonomically-adjusted allometric equations for mammals suggests generation length for P. livingstonii is 8.1 years (Pacifici et al. 2013). Thus, the captive estimate appears in line with estimates for wild populations, and we conservatively estimate the generation length in the wild population of P. livingstonii to be 8.1 years.
True changes in the population size of P. livingstonii are currently unknown, but are likely to be linked to changes in forest and underplanted forest habitat. Our estimates, derived from the little available data on forest loss rates (FAO 2010, UNDP 2013), suggest that habitat decline over the most recent three-generation (or 24.3-year) period was 83%. Thus, our best estimate of loss of P. livingstonii habitat exceeds the threshold under the A2 criterion for CR status (≥80% loss of population, as suspected by habitat change, over a three-generation period, where habitat loss is ongoing).
|Current Population Trend:||Decreasing|
|Habitat and Ecology:||This species roosts in emergent trees, primarily on steep-sided valleys with south-east facing slopes, near ridge tops and in areas generally associated with natural vegetation (Granek 2002). Ficus esperata, Girostpula comoriensis, Gambeya spp., Ficus lutea and Nuxia pseudodentata are the five most commonly used species for roosting (Daniel et al. in press.). Populations of this bat are largely confined to primary tropical moist forest (Mickleburgh et al. 1992), especially montane areas. Specifically, on Anjouan, all the major roosts are restricted to a narrow mid-altitudinal range (600-960 m asl) and are strongly associated with the presence of native and endemic trees, with most of the largest roosts located in dense canopy, old growth forest (Granek 2002, Sewall et al. 2007, Sewall et al. 2011, Daniel et al. in press.). The bats feed primarily on fruit, but also to a lesser degree on nectar and leaves (Trewhella et al. 2001, Sewall 2002). Feeding is principally on native species but also includes non-native kapok (Ceiba pentandra) (Trewhella et al. 2001, Sewall 2002); particularly important food sources are native Ficus lutea and Ficus antandronarum (Sewall 2002).|
|Generation Length (years):||8.1|
Pteropus livingstonii is threatened principally by the continuing degradation, fragmentation, and loss of its forest habitat (Sewall et al. 2007). Over the last few decades, the Comoros have undergone a sustained, rapid deforestation, resulting in the loss of nearly all its native forests. Data quality on rates of habitat loss is poor (FAO 2010), but best estimates are that during the past 20 years alone, the country has lost 75% of its remaining forests, the fastest rate of any country in the world (UNDP 2013). This habitat change is caused by cutting of trees for fuelwood and construction, and by conversion of all but the steepest upland areas to agricultural use (for subsistence agriculture and for export crops such as cloves) (Weightman 1987, Sewall et al. 2007, Sewall et al. 2011b, Doulton et al. 2015). With continuing habitat change, this species has retreated to higher elevations (above 450 m asl on Anjouan and above 250 m asl on Mohéli). Even above these elevations, forest habitat on both islands, and especially Anjouan, has undergone extensive and continuing habitat degradation, fragmentation, and loss (Sewall et al. 2007, 2011a; R. Young unpublished), leaving large areas increasingly unsuitable for this species. Human pressures on native forests are expected to continue, as the growth rate of the human population remains high, with an annual increase of 2.5% (UNDP 2013).
Roost sites represent critical habitat for Pteropus livingstonii (Granek 2002, Sewall et al. 2007, Sewall et al. 2011b), yet tree felling has destroyed a number of roosts, leading to displacement of bats to less impacted areas, often at higher elevations (Trewhella et al. 2005, Sewall et al. 2011b). Of 23 roosts occupied in 2007 (Sewall et al. 2007), three roosts have been abandoned due to clear felling of forest in the past five years, and only one new but very small roost (15 bats) has been uncovered despite extensive searching effort (Daniel et al. in prep.). This new roost was found in a patch of degraded forest at 1050 m, the first to be found at such a high elevation, and is believed to have been established very recently. On Anjouan, forest clearance, underplanting or significant soil erosion following deforestation upslope of the roost was found within 50 m of all but one of the 16 occupied roosts (Daniel et al. in prep.). Further, the replacement of native forests with agricultural lands on Anjouan and Mohéli may render formerly inaccessible roost sites on Anjouan and Mohéli more accessible, possibly exposing roosts to increased levels of human disturbance. Finally, the loss of native forests has resulted in impermanence or complete drying of nearly all rivers on Anjouan and many on Mohéli (Louette et al. 2004, Fernandez Astudillo 2012). As P. livingstonii roosts are often associated with rivers and other humid environments, a factor which may be related to their thermal sensitivity (Granek 2002), this loss of rivers may additionally affect the quality of roosting habitat.
There is no evidence that the species is hunted for food (Trewhella et al. 2005), a key factor affecting other fruit bat species (Mickleburgh et al. 1992). This may be due in part to cultural taboos among the majority of the population that surround consumption of fruit bats. However, the apparent lack of hunting may also be due in part to P. livingstonii’s habit of roosting in inaccessible areas remote from towns; some limited hunting of the other two fruit bat species in the Comoros has been occasionally recorded (Sewall et al. 2003, B.J. Sewall, pers. obs.). It is unclear if the increased access to formerly inaccessible roost areas could expose the bats to any hunting pressure in the future, but any such hunting would likely first affect P. livingstonii’s congener, P. seychellensis comorensis, which sometimes roosts and forages in visible locations in or near towns.
The mean surface temperature of the Comoros is thought to have increased about 1° C since 1900 and is expected to increase by 1-3° C more by 2100 due to global climate change (IPCC 2013). It is unclear whether or how such changes will affect P. livingstonii, but since thermal characteristics are important components of roost suitability for this and other Pteropus species (Pierson and Rainey 1992, Granek 2002) predicted temperature increases could affect availability of suitable roosting habitat.
P. livingstonii’s small population is found within a restricted range solely on two small, adjacent islands. The species is therefore susceptible to single threatening processes, like cyclones, that could simultaneously or rapidly affect its entire range (Sewall et al. 2007). Further, the species may have naturally taken advantage of elevational differences in tree fruiting phenology to compensate for seasonal variation in fruit availability (Sewall 2002). Thus, the loss of large areas of foraging habitat including all the lower parts of its elevational range, could result in seasonally restricted food availability.
A detailed conservation plan for the species, the Conservation Action Plan for Livingstone’s Flying Fox (Sewall et al. 2007) was developed by the non-governmental organizations Action Comores Anjouan and Action Comores International, and a consortium of other conservation groups, through a participatory process involving a broad range of stakeholders. The plan identifies a conservation strategy with several key elements, including habitat protection, forest management, environmental education, population monitoring, ecological research, ex-situ breeding, and conservation partnerships. This conservation plan was adopted by the government of the Union of the Comoros as the national conservation strategy for this species and has served as a guide for conservation action.
A long-term comprehensive citizen science program involving Comorian villagers in population monitoring for this species was led by Action Comores Anjouan and Action Comores International between 1992 and 2006. Population surveys and habitat evaluation around roost sites was conducted by Engagement Communautaire pour le Développement Durable of the Bristol Zoological Society and Durrell Wildlife Conservation Trust between 2009 and 2012, and by Dahari more recently. A national environmental education programme was implemented by Action Comores Anjouan and Action Comores International, and it was successful in raising local and international awareness of this species and the threats it faces, in improving training of local personnel, and in increasing local participation in conservation and science programs (Trewhella et al. 2005). A small community-based ecotourism program for this species has been implemented on Mohéli by Projet Conservation de la Biodiversité et Développement Durable and local communities and is maintained by the local community in Ouallah 2 (Sewall et al. 2011b, Doulton et al. 2015). Dahari is implementing integrated landscape management around the southern forest block of Anjouan; this involves sustainable land management at the roosting and foraging habitat of this species and a pilot program of payment for ecosystem services both within the Moya Forest area (Doulton et al. 2015). Research by Dahari and partners into feeding ecology and the genetics of roost-site populations is also planned.
There is an active captive-breeding programme underway for this species, initiated by Durrell Wildlife Conservation Trust in 1992. The program began with 17 founder individuals (7 females and 10 males) captured from the wild on Anjouan during 1992-1995 (Clark et al. 1997). The captive population of P. livingstonii subsequently expanded at a slower rate than other fruit bat species in captive breeding programs (e.g., P. rodricensis), at least initially (O’Brien 2011). However, by 2014, the P. livingstonii captive population had reached 59 individuals (22 females and 37 males) housed at four institutions (Durrell Wildlife Conservation Trust [Jersey, U.K.]; Bristol, Clifton, and West of England Zoological Society [Bristol, U.K.]; North of England Zoological Society [Chester, U.K.]; and Lisieux Cerza [Lisieux, France])(Glenewar 2014).
This species receives the highest level of legal protection available within the Union of the Comoros. It is listed as an ‘integrally-protected species’ (list 1 of RFIC 2001), which prohibits the capture or detention of P. livingstonii individuals without a permit. This law also expressly prohibits the killing of flying fox individuals; transport, purchase, sale, export or re-export of live or dead flying fox individuals or body parts; all disruption during the period of reproduction and raising of young; and the destruction of roosts (RFIC 2001). The Union of the Comoros also ratified the Convention on Biological Diversity in 1994, and in response has developed a National Biodiversity Conservation Strategy (Roby and Dossar 2000). This strategy highlights the importance of, threats to, and conservation recommendations for fruit bats of the Union of the Comoros (Sewall and Granek 2000). P. livingstonii is also listed on Appendix II of the Convention on International Trade in Endangered Species of Wild Fauna and Flora, or CITES (UN 1973), which prevents international trade in specimens of this species without a permit. In practice, however, enforcement activities within the Comoros for these laws and treaties have been very limited in scope (Sewall et al. 2007).
P. livingstonii also does not occur in any protected areas. However, critical roosting habitat at seven key roost sites that together harboured more than half the population was identified during the development of the conservation action plan (Sewall et al. 2007). Habitat conservation of these roost sites was identified as a key goal in this plan. As part of its implementation, Action Comores Anjouan and the Comoros Forest Reserves Project conducted a conservation assessment to identify the two highest-priority sites (roost sites Yiméré on Anjouan and Hassera-Ndrengé on Mohéli), where the establishment of protected areas for roost habitat would be both highly beneficial to biodiversity conservation and highly feasible (Sewall et al. 2011a). These two groups also completed conservation planning for protection of critical roosting habitat at all seven of these critical roost sites (Sewall et al. 2011b). Implementation efforts by Action Comores Anjouan and partners to conserve habitat around critical roosting habitat on Anjouan and Mohéli are under development. A project to establish larger reserves to protect rainforest and cloud forest on both Anjouan and Mohéli, which would include roosting and foraging habitat for this species, has been proposed by the United Nations Development Programme, the Global Environmental Facility, and the Comorian government. The success of habitat conservation and restoration efforts will be particularly critical to the long-term prospects for this species.
|Citation:||Sewall, B.J., Young, R., Trewhella, W.J. & Rodríguez-Clark, K.M. and Granek, E.F. 2016. Pteropus livingstonii. The IUCN Red List of Threatened Species 2016: e.T18732A22081502.Downloaded on 27 October 2016.|
|Feedback:||If you see any errors or have any questions or suggestions on what is shown on this page, please provide us with feedback so that we can correct or extend the information provided|