DistributionRead full entry
Range DescriptionAmazilia castaneiventris was formerly known from the slopes of the Serrana de San Lucas, Colombia where one specimen was taken in 1947 on the east slope of the serrana in Bolvar, but this population has not been rediscovered despite rapid assessment searches in 1999-2001 (T. Donegan in litt. 2008). Today, it appears to be restricted to the drier parts of the Magdalena Valley, Colombia, with a core range in the Chicamocha, Suarez and Chucuri valleys. Although there are various sites where the species is now known to be found, it is somewhat unpredictable in occurrence (partly influenced by poorly understood seasonal movements) and, at least in the Yariguies area, not locally abundant. Recent records from the rio Chucurri basin and La Paz are outside of and generally more humid than the dry valley system that forms the core of its range (J. O. Cortes in litt. 2009; D. C. Sabogal in litt. 2009). Extensive survey work by Fundacion ProAves has recorded the species at 14 sites. There are historic records from two sites in Santander (in 1962 and 1963), and three in Boyac including the 1977 specimen from Tipacoque; recent work has again recorded the species at Tipacoque (J. Zuluaga in litt. 2009), it has now been recorded in eight municipalities including a rediscovered population in the environs of Soat (Chavez and Cortes 2006, Cortes-Herrera et al. 2006, Cortes-Herrera et al. 2007, Parra et al. 2006) and there was a sighting in 2000 at Villa de Leyva (Lpez-Lans 2002), although this has not been confirmed, and subsequent visits to Villa de Leyva have failed to find the species (J. Cortes in litt. 2011). Historically, it was locally common, but trends are difficult to assess owing to a lack of baseline data. The species is often inexplicably rare in apparently suitable habitats and may go unrecorded for periods (J. Zuluaga in litt. 2009); elsewhere it appears to be resident. The increase in records of the species owes much to increased observer effort, but also may reflect nomadic movements in recent years linked to flowering events on which the species relies to some degree (J. Cortes in litt. 2009). The global population is roughly estimated at 3,780 individuals by extrapolating the species's known territory size by the area of suitable habitat (D. C. Sabogal in litt. 2009); however, it is unlikely to be evenly distributed throughout suitable habitats and this figure may represent an overestimate. Based on several years of surveys, J. Cortes (in litt. 2011) roughly estimated the global population at 1,200 individuals by extrapolating densities from Soata and Tipacoque in 2008-2011 to known habitat and suitable habitat of the species (J. Cortes in litt. 2011).