Archived 2014 discussion: Mauritius Kestrel (Falco punctatus): uplist to Endangered?

This discussion was first published as part of the 2013 Red List update, but remains open for comment to enable reassessment in 2014.

BirdLife species factsheet for Mauritius Kestrel

Mauritius Kestrel Falco punctatus is restricted to Mauritius. It is currently listed as Vulnerable under criterion D1+2 because it has a very small population, susceptible to a variety of threats.

This species has undergone a spectacular recovery from just four wild birds (including one breeding pair in 1974 (Safford and Jones 1997, Anon. 2005). By the end of the 1994 breeding season there were an estimated 222-286 birds in the population, following a successful recovery programme launched in 1973 (Nicoll et al. 2004). At the end of the 1999-2000 season, the population was estimated at the time to number 145-200 breeding pairs and a total population of 500-800 individuals (C. Jones in litt. 2000), divided into three subpopulations on mountain chains in the north, east and south-west of Mauritius (Jones and Swinnerton 1997). In 2007-2008 the population was estimated at 500-600 individuals by Dale (2008); 800-1,000 individuals were estimated in 2005 (Anon. 2005, Mauritian Wildlife Foundation in litt. 2006) but it is now thought unlikely that the population ever approached 1,000 (V. Tatayah in litt. 2012), and may have only peaked at 350-500 individuals at the end of the 1990s (C. Jones in litt. 2012).  By 2011-2012 the population was estimated to number c400 individuals on the east and west coast, with the small subpopulation in the Moka Range in the north of the island apparently extinct (possibly due to low habitat quality, predators and safe nesting sites), and declines observed in the south-western population, particularly in suboptimal habitat on the periphery of the range, since 2007-2008 (V. Tatayah in litt. 2012). The eastern population is described as stable and appears to be limited by nesting sites to around 40 pairs, although the population remains dependent on conservation measures (Groombridge et al. 2001) and there is no record of dispersal to other locations despite intensive monitoring through colour ringing (Ewing et al. 2008, Senapathi et al. 2011). Nevertheless, attempts are being made to revert the decline on the west coast with the provision of nest boxes in this area and maintaining those on the east of the island (V. Tatayah in litt. 2012).

Confirmation that the global population is undergoing a continuing decline will warrant the uplisting of this species to a higher category of threat. The Extent of Occurrence (EOO) is estimated to be 160 km2; it is now found in 2 locations (since the northern range is now extinct) and with sufficient evidence of a continuing decline in this species’s number of mature individuals and area, extent and/or quality of habitat, it would qualify as Endangered under criterion B1ab(iii,v) of the IUCN Red List. The number of mature individuals in the largest subpopulation is estimated to be 150 individuals (V. Tatayah in litt. 2012). As the population is estimated to be <2,500 mature individuals, if all subpopulations are confirmed to be ≤250 mature individuals and the global population is declining, this species could also qualify as Endangered under criterion C2a(i).

Subpopulations are defined by the IUCN as geographically or otherwise distinct groups in the population between which there is little demographic or genetic exchange (typically one successful migrant individual or gamete per year or less). The term ‘location’ defines a geographically or ecologically distinct area in which a single threatening event can rapidly affect all individuals of the taxon present. The size of the location depends on the area covered by the threatening event and may include part of one or many subpopulations. Where a taxon is affected by more than one threatening event, location should be defined by considering the most serious plausible threat. (IUCN 2001). For example, where the most serious plausible threat is habitat loss, a location is an area where a single development project can eliminate or severely reduce the population. Where the most serious plausible threat is volcanic eruption, hurricane, tsunami, frequent flood or fire, locations may be defined by the previous or predicted extent of lava flows, storm paths, inundation, fire paths, etc.

Further information is requested on this global population size and trends of this species and any additional comments on the proposed uplisting are welcome.


Anon. (2005) 30 years working with the Mauritius Kestrel. MWF Newsletter: 4.

Dale, R. (2008) In Search of Mauritius Kestrels. The Peregrine Fund Newsletter 39: 10-11.

Ewing, S. R., Nager, R. G., Nicoll, M. A. C., Aumjaud, A., Jones, C. G. and Keller, L. F. (2008) Inbreeding and loss of genetic variation in a reintroduced population of Mauritius Kestrel. Conservation Biology 22(2): 395-404.

Groombridge, J. J., Bruford, M. W., Jones, C. G. and Nichols, R. A. (2001) Evaluating the severity of the population bottleneck in the Mauritius kestrel Falco punctatus from ringing records using MCMC estimation. Journal of Animal Ecology 70: 401-409.

Jones, C. G. and Swinnerton, K. J. (1997) A summary of the conservation status and research for the Mauritius Kestrel Falco punctatus, Pink Pigeon Columba mayeri and Echo Parakeet Psittacula eques. Dodo: Journal of the Jersey Wildlife Preservation Trust 33: 72-75.

Nicoll, M.A.C., Jones, C.G. and Norris, K. (2004) Comparison of survival rates of captive-reared and wild-bred Mauritius kestrels (Falco punctatus) in a re-introduced population. Biological Conservation 118(4): 539-548.

Safford, R. J. and Jones, C. G. (1997) Did organochloride pesticide use cause declines in Mauritian forest birds? Biodiversity and Conservation 6(10): 1445-1451.

Senapathi, D., Nicoll, M. A. C., Teplitsky, C., Jones, C. G. and Norris, K. (2011) Climate change and the risks associated with delayed breeding in a tropical wild bird population . Proceedings of the Royal Society B Published online before print March 23, 2011.

This entry was posted in Africa, Archive and tagged . Bookmark the permalink.

3 Responses to Archived 2014 discussion: Mauritius Kestrel (Falco punctatus): uplist to Endangered?

  1. Carl Jones. says:

    Below are the relevant sections from the Handbook of the Birds of the Malagasy Region, in press Safford and Hawkins. this covers several of the points raised in your discussion about the species.

    Status and conservation.
    POPULATION AND DENSITY: The species was probably common and widespread on pristine Mauritius and its decline in distribution and numbers was primarily due to the destruction of the native forest, although some birds occurred in areas of secondary forest. The species was common in the 19th century in suitable habitat. The population declined in the twentieth century and was a very rare bird by the 1930’s and 1940’s and this may have been due to predation by the Lesser Indian Mongoose introduced in 1900 (Cheke and Hume, 2008). From the late 1950’s it is believed to have been limited to the south west of the island in and around the Black River Gorges where the population further declined after the widespread use of DDT in the late 1950’s. The population was in single figures in the early 1970’s and by 1974 there were only four known wild birds. In 1975, only one of these pairs successfully bred and it is suggested that the current population may be derived from one breeding pair (Groombridge et al. 2000, 2001).

    Breeding density increased in the south-west as the population grew during the years 1973-1996, although breeding range also expanded. The population in the south-west had a breeding density of one pair per 1.5 km2 in 1995-1996 (O’Brien, 1997). The number of non-paired floating birds during the early breeding season in the south-west population during the 1980’s and 1990’s was consistently estimated at between 0.6 -1.0 per territorial pair (O’Brien, 1997, CGJ).

    The population recovered in the 1980’s and 1990’s due to the conservation management of the wild population and the reintroduction of captive reared birds into the three mountain ranges. An analysis of available habitat and an assumed mean territory size of 1 km2 suggested that the island could hold 200-250 pairs and allowing for a floating population of about one or more per pair this gave a maximum of 600-800 birds but realistically it was suggested that the eventual population of about a hundred pairs and 500-600 birds was achievable (Jones et al 1994). The population growth of the kestrels was modelled using re-sighting data of colour ringed birds and it suggested that in 1997 about 50% of the breeding birds in the south-western population were being undetected, 61 pairs were known or suspected and it suggested that there were therefore about 120 pairs. In the east there were 35 known pairs and it was suggested to be about 41 pairs, and both populations were still growing (Groombridge et al., 2001). There was also a small population at the time in the north not considered by Groombridge et al., (2001). Allowing for one floating bird per pair this gave a possible total pre breeding population of 483 in the south-west and east and a total including the northern population of over 500 birds. The upper possible population levels of 145-200 breeding pairs and 500-800 birds have been widely quoted (BirdLife, International, 2000) although it is unlikely that the population would have exceeded 500 birds in the late 1990’s and was likely to be less than this since the computation of about 120 pairs in the south-west in the late 1990’s is now believed to have been over optimistic. Realistically in the late 1990’s when the kestrel reached its asymptote the population was probably in the region of 350-500 birds (CGJ).

    With the relaxation of conservation management the population declined in the 2000s and in 2012 the population is thought to be 300-350 birds. The Northern population in the Moka Mountains died out and the population in the south west has declined to a lower holding capacity of about 40-50 pairs and a pre-breeding season population of about 120-150 birds. The eastern population has grown to a stable population of about 45-50 pairs and a total of 130-150 birds pre-breeding season (MACN). The population data from 1970’s-2010 are being reanalysed to clarify what the asymptotic population was and what level of decline has occurred (RD).

    The primary cause of the kestrel’s decline was habitat destruction from the 17-20th centruries which reduced the amount of available native forest habitat by 95% (Jones and Owadally, 1988, Safford, 1997, Cheke and Hume, 2008). The remaining areas of native forest are being invaded by exotic plants, mainly Strawberry Guava Psidium cattleianum, and Privet Ligustrum robustum, that forms a denser under-story, making hunting beneath the canopy either difficult or impossible. In heavily degraded forests the density of their favoured prey Phelsuma geckos is much less (CGJ).

    The loss of many of the large canopy trees with hollow trunks and limbs, has resulted in a shortage of tree nest cavities. Natural high quality nest cavities are at a premium (contrary to earlier thinking (Jones, 1987)) and many of the cliff cavities that the kestrels use are sub-optimum being accessible to predators and are too small and prone to overheating or flooding (CGJ).

    Competition for nest sites is severe and suitable cavities are often taken over by Black Rats, Common Mynahs, White-tailed Tropicbirds and feral pigeons precluding the kestrels from occupying them.

    Once a whole clutch is laid predation is infrequent with only three whole clutches out of 184 laid were taken by predators in the eastern population where they nest mainly in nest-boxes (Jones et al. 1994). Predation from natural cavities is much higher. Eggs are occasionally lost during laying and not being closely attended and up to about 5% may disappear probably taken by rats (Jones et al., 1994) and Common Mynahs (CGJ). The exotic Long-tailed Macaque and Black Rat are also predators of the eggs and young. Mongooses commonly occur on rocky slopes and can readily access some kestrel nest-sites and may kill incubating females (CGJ). Recently fledged kestrels often descend to the ground where they are vulnerable to predators especially mongooses but also monkeys, feral cats Felis cattus and people (Jones and Owadally, 1988, Jones et al., 1991, Burgess et al., 2009). Mongooses were thought to have killed <10% of released young (Jones et al., 1991) and it is suspected that in some parts of the south-west to be higher than this (CGJ).

    The contraction of range, and their loss from drier rocky areas in the south-west in the early 20th century is thought to be due to mongoose predation. The inability of the reintroduced kestrels to establish a population in the dry coastal areas of Rempart and Trois Mammelles mountains, Tourelle de Tamarin, Black River and Le Morne, in the late 20th century is probably due to high mongoose, and other predator densities (CGJ).

    Areas of good native forest can support kestrel source populations with the production of surplus young. Secondary forest and agricultural land tend to be sink areas and although the kestrels can survive and breed in these areas productivity and subsequent recruitment is not high enough to replace adult mortality (Burgess et al., 2011, CGJ). The reason why these are sink areas is likely to be due to higher rates of predation, the quality of the food and disease.

    Kestrels living in secondary forest, in mixed exotic forest and on the borders of agricultural land and habitation, when compared to kestrels living in native forest, feed more on Agama Lizards and House Shrews (Jones and Owadally, 1988). The adults can feed on these with no problems but they are unsuitable if fed to young chicks. The chicks cannot deal with the high roughage content and the fur and scales cause fatal impactions of the stomach which kills them. The kestrels have evolved feeding primarily on Phelsuma geckos that are lower in roughage (Jones et al., 1994).

    Kestrels living in these sub-optimal habitats also feed on juvenile Barred Ground Doves that are more common in these disturbed habitats and around human habitation. The doves carry the flagellate protozoan parasite Trichomonas gallinae that causes the disease trichomonosis, which Mauritius Kestrels are very susceptible to (CGJ). In a study on this disease 58.5% of doves sampled were carrying trichomonosis (Bunbury et al. 2007). This disease causes the build up of hardened pus lesions in the throat and mouth that obstruct feeding and breathing and kill both adults and young. Although this cause of mortality is not high, combined with high predation rates and sub-optimal food quality it compounds the problems of living in secondary habitats and built up areas.

    Several adult (< 1%) and nestling kestrels (< 1%) have been found with avian pox lesions in both the western and eastern populations, which are of unknown pathogenicity.

    Pesticides are believed to have been responsible for wiping out the populations in the Moka and Bambous Mountains in the1940s and 1950s and in reducing the population in the south- west. Organochlorine pesticides were used extensively in the 1940’s and 1950’s for agricultural purposes and Malarial Control (Cheke and Hume, 2008). Eggs laid by captive kestrels in 1978-1979 contained appreciable quantities of DDE and dieldrin and were significantly smaller and thinner shelled than eggs laid 1985-1992 (see EGGS). Eggs laid by both wild and captive birds during the 1980’s are significantly smaller than the eggs collected before 1940 suggesting that low levels of contamination was still affecting the recovering population (Jones, 1987, Safford and Jones, 1997).

    Inbreeding. Since the Mauritius Kestrel has recovered from such a small bottleneck the current population must be very inbred. The pre-bottleneck (<1974) and post bottleneck populations were sampled by looking at 12 microsatellite DNA loci in museum skins up to 170 years old and from living birds. The ancestral variation was remarkably high comparable to continental species. Post-bottleneck the allelic diversity, across all loci, fell by 55% and heterozygosity by 57% (Groombridge et al., 2000).

    Inbreeding in the reintroduced eastern population has been closely studied (Ewing et al., 2008). The population has accumulated inbreeding at a substantial rate since its reintroduction from 1987-1994 and about 25% of all matings were between closely or moderately closely related birds (F≥ 0.125 level). The mean inbreeding value was 0.077 in 2006. Genetic diversity has been lost from the population but this has been less rapid than the rate of increase of inbreeding that has accumulated at a rate of 2.6%/generation (Ewing et al., 2008).


    The captive population on Mauritius was established between 1973 and 1988, although the first breeding attempt in the 1970s failed (Jones et al. 1980) and a project was re-launched in 1981-1982 which was successful (Jones 1983, Jones et al. 1991). For both attempts five adult, four fledglings and three nestlings were harvested and fourteen birds reared from harvested eggs were retained for the captive breeding project (Jones et al, 1994). From 1978-1992, 190 fertile eggs were laid, 129 (68%) hatched, 103 (80%) were reared and 84 (82%) were released to the wild.

    Three pairs were sent to the World Centre for Birds of Prey, Boise Idaho USA and some of the young returned to Mauritius and 18 were released to the wild.

    The most successful method of increasing the productivity of the wild pairs was the harvesting of eggs as whole clutches, the wild birds then usually laid a replacement clutch that they were left to rear. 388 eggs were harvested from the wild, 292 (75%) were fertile, 242 (83%) hatched and 233 (96%) of these were reared and most returned to the wild (Jones et al., 1991, 1994). Harvesting clutches of eggs enhanced the size and number of clutches laid by managed females and improved mid-life male and female adult survival relative to unmanaged adult kestrels (Nicoll et al., 2006). Management resulted in an increased effort in egg production but, it reduced parental effort during incubation and the rearing of offspring, which could account for these observed changes (Nicoll et al., 2006).
    331 birds were released between 1984-1993, either by fostering to wild nests or the soft release of fledglings and older birds. 105 birds were fostered to 46 different wild nests at 5-18 days old, 1-4 (2.3) were fostered per nest 96 (91%) fledged and 78 (81%) of these became independent (Jones et al., 1994).
    50 groups of 1-7 (4.2) young kestrels (25-34 days old) were soft-released and of the 208 released 164 (79%) became independent. 18 fully grown juveniles were released and fourteen became independent (Jones et al., 1994). These went to establish three populations in the south west and south, 201 released (117 independent), east 120 (90) and north 40 (30). The survival of these released kestrels was comparable to those that had been reared naturally by their parents (Nicoll et al., 2004). Predators, especially mongooses, cats and rats were controlled around release sites and some active nest-sites (1984-1993) (Jones et al., 1991, 1994)
    In the eastern population 53% (n=85) of birds that became independent were subsequently located in territorial pairs (Jones et al., 1994).
    All the birds that were soft-released were provisioned with supplemental food until they were independent and some of those that were fostered were provisioned with supplemental food during the nestling and post- fledging stages of development until independence. Some adult pairs that were being used for egg harvesting were provided with additional food to enhance egg production (Jones et al., 1991).

    Nest site enhancement and nest-boxes have been an important part of the kestrel’s management. In the eastern population where there is a paucity of natural cavities, 73% of documented nest attempts (n=815, 1988 – 2010) have been in boxes with the remaining 24% in cliffs and 3% in trees. Natural cavities have been improved by removing rocks from the cavity floor, replacing or providing suitable substrate and providing suitable perches for the kestrels to land on in front of the cavity. Others that were accessible to predators were blocked (Jones et al., 1991, 1994).

    It is proposed to genetically manage the populations. Some diversity has been located in the MHC gene with alleles present in the south-western population that are not known from the eastern population (Jim Groombridge pers. comm.). It is intended to translocate birds between these populations to facilitate gene flow.

    To improve productivity in the south-western population additional nest-boxes are to be erected in suitable areas in the lowland Black River Gorges, Chamarel and Bel Ombre where there are shortages of high quality nesting sites. Monitoring of both the western and eastern populations will continue.

    Authors: Carl G. Jones, Malcolm B. Burgess, Jim J. Groombridge, Richard Dale, Vikash Tatayah, Nicolas Zuel, Malcolm A. C. Nicoll.

  2. Andy Symes says:

    Preliminary proposals

    Based on available information and comments posted above, our preliminary proposal for the 2014 Red List would be to treat Mauritius Kestrel Falco punctatus as Endangered under criteria B1ab(iii,v); C2a(i).

    There is now a period for further comments until the final deadline of 31 March, after which recommended categorisations will be put forward to IUCN.

    The final Red List categories will be published on the BirdLife and IUCN websites in mid-2014, following further checking of information relevant to the assessments by both BirdLife and IUCN.

  3. Andy Symes says:

    Recommended categorisation to be put forward to IUCN

    Following further review, there has been no change to our preliminary proposal for the 2014 Red List status of this species.

    The final categorisation will be published later in 2014, following further checking of information relevant to the assessment by BirdLife and IUCN.

Comments are closed.