COTINGA 1 Focus On Focus On: Ochre-bellied Dove Leptotila ochraceiventris Brinley J. Best

Ochre-bellied Dove Leptotila ochraceiventris (Richard Thewlis)

Of the 10 Neotropical doves in the genus The Ochre-bellied Dove L. ochraceiventris Leptotila, three are classified as globally is confined to western Ecuador and adjacent threatened in the Americas Red Data Book6, north-west Peru, a range it shares with 54 namely: Grenada Dove L. wellsi of the West other species of bird restricted to the so- Indies, Tolima Dove L. conoveri of central called Tumbesian centre of endemism or Colombia, and Ochre-bellied Dove L. Tumbesian western Ecuador and Peru En­ ochraceiventris of western Ecuador and demic Bird Area3,4,7. This area is one of the adjacent north-west Peru (the Tumbesian most topographically and vegetationally region). Recent fieldwork in the Tumbesian complex regions in South America, support­ region has yielded important new informa­ ing no fewer than 17 different vegetation tion on L. ochraceiventris which has con­ types (including 13 forest types) in an area firmed its status as one of the most threat­ of c.50 000 km2. This region possesses a ened of the Neotropical columbids. The strongly seasonal climate, with the period main aim of this article is to highlight some from January to April usually very wet in of the recent findings on the ecology and contrast to the rest of year10. Due to this habitat preferences of L. ochraceiventris. seasonality the region supports tropical de­ Information is also provided on the known ciduous and semi-deciduous forests, rare at altitudinal and geographical range of the such latitudes elsewhere in South America. species. It is hoped that the data presented here will help ornithologists visiting Ecua­ First described in 1912, L. ochraceiventris dor and Peru to find this shy and enigmatic was collected in small numbers during the species, thereby increasing our knowledge first two decades of the twentieth century in of it. the lowlands of south-west Ecuador and

30 COTINGA 1 Focus On adjacent north-western Peru, by ornitholo­ West Peru Biosphere Reserve (extreme gists from the American Museum of Natural north-western Peru)11. History. Their early fieldwork was summa­ rized by Chapman5 who stated that the dove Leptotila ochraceiventris inhabits a broad was “not uncommon in the deciduous for­ spectrum of habitat types, ranging from Dry ests of south-west Ecuador”. Between 1926 Ceiba trichistandra-dominated Deciduous and 1980, however, there were only two Forests to the most humid forest type of the records: one each from Ecuador and Peru, Tumbesian region, very humid premontane though this was most probably as a result of cloud-forest. Although by far the majority of the lack of survey effort within its range. It records come from within forest, the species was only in the 1980s when intensive sur­ has also been recorded in scrub, but usually veys recommenced that it became clear that in areas adjacent to forest. Leptotila a large population decline had occurred ochraceiventris is a forest understorey spe­ since the 1920s, leaving the bird uncommon cies, favouring the zone from the forest floor or rare at a few scattered sites in the to about 3 m up, although it has occasionally Tumbesian region, and missing from many been found feeding higher up in fruiting of its historical localities. There were no trees. Areas with a thick covering of leaf records between 1961 and 1979, and the litter seem especially favoured, and wher­ species was “rediscovered” near Piñas in El ever the understorey has a more evergreen Oro Province, Ecuador, in August 198012, character, the chances of a sighting are but was not found again until January- increased. Although virtually nothing is March 1986, this time in Peru at Canchaque known of its feeding habits, birds have been (Piura Dept.) and Campo Verde in the observed taking the marble-sized fruits of Tumbes National Forest (Tumbes Dept.) (M. the Trichilia tree1. Kessler in litt. 1991). It is thought that L. ochraceiventris may be Surveys since 1989, especially in the El Oro a local migrant, moving from one forest type and Loja provinces, south-west Ecuador, to another seasonally (and only in certain have gathered much needed basic data on years). For example, the dove was recorded the birds habitat preferences and ecology; at Buenaventura (El Oro Province, 3°40'S they have also extended the known range of 79°44'W) in August 1980 and August 198812, L. ochraceiventris both geographically and but not during detailed surveys at other altitudinally. It has now been recorded months (e.g. February and March 19912, from 23 localities in the Ecuadorian prov­ September 199113) nor during intensive field­ inces of Manabí, Los Ríos, Guayas, work in June and July 198512 (data suggest Chimborazo, El Oro and Loja, and the Peru­ that these movements occur at other sites in vian departments of Tumbes and Piura, south-west Ecuador). Movements within embracing elevations from sea-level to 2,625 forests also seem to occur, with birds mov­ m, though most reports come from between ing to the more humid areas, such as those 500 and 1,800 m. The centre of abundance found along water courses, during the dry appears to be the south-western provinces season. of Ecuador, although it should be noted that this region has received the most intensive Leptotila ochraceiventris is usually encoun­ survey effort. The recent sightings of the tered walking quietly on the forest floor. species in montane cloud-forest above They can be a difficult species to observe, Ayabaca in northern Peru at 2,625 m1 unex­ since the sound of an approaching observ­ pectedly extended the known upper er’s footsteps in crunchy leaf litter usually altitudinal limit of L. ochraceiventris by al­ causes the bird to move quickly away from most 1,000 m into the temperate zone. The the observer to safety, or to a low branch or only sites at which the species has recently other suitable perch, only descending to the been reported numerous are within the ground again if the observer remains still Tumbes National Forest, part of the North- and silent. Frequently the only view ob­

31 COTINGA 1 Focus On tained is of the upperparts as birds fly away. The future for the Ochre-bellied Dove L. At such times, the tail pattern is the key to ochraceiventris is uncertain: 95% of lowland correct identification (see below). One of the south-western Ecuador has been deforested, best strategies to find the species is simply and those patches which remain continue to to wait quietly overlooking suitable areas. be at risk8 — the picture is little better in These include water holes or streams where Peru. Although the bird occurs inside sev­ the species often gathers to drink along with eral officially protected areas, notably the larger numbers of commoner doves, espe­ Machalilla National Park in Ecuador and cially White-tipped Doves L. verreauxi, and the Tumbes National Forest in Peru, these fruiting trees which are known to be attrac­ areas are far from secure at present, owing tive to doves. One continuous 12 hour to their inadequate staffing levels and the water-hole survey at Tambo Negro in south­ park authorities’ meagre resources. Forest ern Loja Province, Ecuador in September understorey is especially vulnerable in the 1989 revealed a minimum of 26 L. verreauxi Tumbesian region, since it is trampled and and seven L. ochraceiventris drinking at grazed by cattle and semi-wild goats and just one tiny pool. Individual L. mules, and is exploited by local people for ochraceiventris were visible for up to sev­ firewood, even inside the existing reserves. eral minutes as they made their way down The species will only survive if extensive the valley sides, drank, and quietly retreated. areas of forest still with an intact understorey can be safeguarded through the designation Field identification is not difficult given a of further reserves and the adequate protec­ clear view. Leptotila ochraceiventris is the tion of existing reserves. Each local commu­ most beautiful and brightly coloured of the nity has a role to play in protecting their Leptotila doves, yet has still never been own forest patches. Such action is not only illustrated in colour. The overall pattern necessary to protect the unique avifauna of bears some resemblance to L. verreauxi, the region but also to regulate water and except the upperparts are duller in colora­ nutrient supply to cultivated areas and to tion and the crown, nape and mantle are prevent soil erosion. The conservation ofL. suffused with purple. The underparts are ochraceiventris should be seen as only a rich buff (“ochre”) from the belly downwards small part of an integrated environmental becoming suffused with purple on the breast, plan for the Tumbesian region aimed at creating a two-tone effect. In contrast to the preserving its biodiversity, but also maxi­ dark iris of L. verreauxi, the iris of L. mizing the benefits local people derive from ochraceiventris is pale. They are smaller future conservation action. The founda­ and more slightly built than L. verreauxi; tions of such a plan will soon be published by this being especially noticeable when the BirdLife International3. two are together. The tail of L. ochraceiventris is distinctively marked and R eferences is a valuable identification feature given a 1. Best, B. J. and Clarke, C. T., eds. (1991) brief flight view: only the outer two tail The threatened forests of the Sozoranga feathers on each side have white spots; in L. region, south-w est Ecuador. Cam­ verreauxi most of the tail feathers show bridge, U.K.: International Council for Bird white spots in flight. The call of L. Preservation (Study Report 44). ochraceiventris is a deep resonant “whoouur” 2. Best, B. J., ed. (1992) The threatened of about one second duration, rising and forests of south-west Ecuador. Leeds, U.K.: then falling in pitch. Birds have been heard Biosphere Publications. mainly during January to April at the ma­ 3. Best, B. J., ed. (in press) Biodiversity and jority of sites (the breeding season for many conservation in Tumbesian Ecuador and bird species in south-west Ecuador2,9), al­ Peru. Cambridge, U.K.: BirdLife Inter­ though at a few sites calls were heard only national. during August. 4. Stattersfield, A. J., Crosby, M. J., Long, A. J. & Wege, D. C. (in prep.) A global

32 COTINGA 1 Focus On directory of Endemic Bird Areas. Cam­ in S.W. Ecuador. Ibis 101: 137–152. bridge, U.K.: BirdLife International. 10. Munday, G. and Munday, M. (1992) The 5. Chapman, F. M. (1926) The distribution climate of south-west Ecuador. Pp. 7–78 in B. of bird-life in Ecuador. Bull. Amer. Mus. J. Best, ed. The threatened forests of south­ Nat. Hist. 60: 1– 784. west Ecuador. Leeds, U.K.: Biosphere Pub­ 6. Collar, N. J., Gonzaga, L. P., Krabbe, N., lications. Madroño Nieto, A., Naranjo, L. G., Parker, 11. Parker, T. A., Schulenberg, T. S., Kessler, T. A. and Wege, D. C. (1992) Threatened M. and Wust, W. (1989) Species limits, natu­ birds of the Americas: the ICBP/IUCN Red ral history and conservation of some en­ Data Book. Cambridge, U.K.: International demic birds from north-west Peru. Unpub­ Council for Bird Preservation. lished report. 7. Cracraft, J. (1985) Historical biogeogra­ 12. Robbins, M. B. and Ridgely, R. S. (1990) phy and patterns of differentiation within The avifauna of an upper tropical cloud the South American avifauna: areas of forest in southwest Ecuador. Proc. Acad. endemism. Pp. 49–84 in P. A. Buckley, M. S. Nat. Sci. Phil. 142: 59–71. Foster, E. S. Morton, R. S. Ridgely and 13. Williams, R. S. R. and Tobias, J. A. (1992) F. G. Buckley, eds. Neotropical ornithology. Preliminary report of the Amaluza 1991 Washington, D.C.: American O rnitholo­ project. Unpublished. gists’ Union. (Orn. Monogr. 36). 8. Dodson, C. and Gentry, A. H. (1991) Biological extinction in western Ecuador. BRINLEY J. BEST Ann. Missouri. Bot. Gard. 78: 273–295. 118 Aldwyn Crescent, Hazel Grove, 9. Marchant, S. (1959) The breeding season Stockport, Cheshire, SK7 5HX, U.K.

33