seasonal distribution, abundance, and nesting of puna ... - BioOne

The Condor 109:276–287 # The Cooper Ornithological Society 2007

SEASONAL DISTRIBUTION, ABUNDANCE, AND NESTING OF PUNA, ANDEAN, AND CHILEAN FLAMINGOS SANDRA M. CAZIANI1, OMAR ROCHA OLIVIO2, EDUARDO RODRI´GUEZ RAMI´REZ3, MARCELO ROMANO4, ENRIQUE J. DERLINDATI1,8, ANDRE´S TA´LAMO1, DAVID RICALDE5, CARMEN QUIROGA6, JUAN PABLO CONTRERAS3, MARIANA VALQUI5, AND HEBER SOSA7 1

Consejo de Investigacioń, Universidad Nacional de Salta-CONICET, Buenos Aires 177, 4400 Salta, Argentina 2 Centro de Estudios en Biologıá Teo´rica y Aplicada – BIOTA, Casilla de Correo 4778, La Paz, Bolivia 3 Corporacioń Nacional Forestal, II Regioń, Av. Argentina 2510, Antofagasta, Chile 4 Centro de Investigaciones en Biodiversidad y Ambiente ECOSUR, Pasaje Sunchales 329, 2000, Rosario, Argentina 5 Peru´ Verde, Calle Manuel Banõn 461, San Isidro, Lima 27, Peru 6 Proyecto: Conservacioń de la Biodiversidad, Autoridad Binacional del Lago Titicaca, Calle Cuba No. 1114, Bolivia 7 Direccioń de Recursos Naturales Renovables, Gobierno de Mendoza, Juan B. Justo 381 (5501), Godoy Cruz, Mendoza, Argentina Abstract. Of the world’s five flamingo species, the rarest and least known are the Puna Flamingo (Phoenicoparrus jamesi) and the Andean Flamingo (P. andinus). These two species coexist with the more common Chilean Flamingo (Phoenicopterus chilensis) throughout much of their range. We conducted four simultaneous surveys from 1997 to 2001 (two in summer and two in winter) to estimate the distribution and abundance of all three species in Argentina, Bolivia, Chile, and Peru, at a regional scale. Of 224 wetlands surveyed, 179 had flamingos; 63% of these were in the high Andes (above 4000 m), 25% were in the puna (3000 to 4000 m), and the remainder were in lowlands (below 3000 m). Maximum counts were 64 000 Puna Flamingos (summer 1998), 34 000 Andean Flamingos (summer 1997), and 83 000 Chilean Flamingos (winter 1998). In summer, Puna Flamingos congregated at wetlands in the high Andes, with 50% of the population in just three lakes: Colorada, Grande, and Vilama. Andean Flamingos were more uniformly distributed across a broader elevational range (2500 m), and Chilean Flamingos showed a heterogeneous distribution pattern. In winter, all species moved to lower latitudes within the high Andes and to lower altitudes on the central plains of Argentina. The most important nesting wetlands were Colorada, in Bolivia, for the Puna Flamingo, Surire and Atacama, in Chile, for the Andean Flamingo, and Surire for the Chilean Flamingo. We recommend continued monitoring through simultaneous summer surveys, and a conservation strategy that considers the large spatial and temporal scales at which these species operate, including their seasonal migrations. Key words: altiplano, Andes, flamingos, Phoenicoparrus andinus, Phoenicoparrus jamesi, Phoenicopterus chilensis, salt lakes.

Distribucioń Estacional, Abundancia, y Nidificaciones de los Phoenicoparrus jamesi, P. andinus, y Phoenicopterus chilensis Resumen. De las cinco especies de flamencos en el mundo, las ma´s raras y menos conocidas son el Phoenicoparrus jamesi y el P. andinus, las cuales coexisten con el Phoenicopterus chilensis en gran parte de su distribucioń. Nosotros llevamos a cabo cuatro relevamientos simultańeos desde 1997 al 2001 (dos en verano y dos en invierno), para estimar la distribucioń y abundancia de tres especies de flamencos en Argentina, Bolivia, Chile y Peru´, a una escala regional. De 224 humedales relevados, 179 tenıán flamencos; 63% de estos se encontraron en los Altos Andes (por encima de los 4000 m), 25% en la Puna (3000 a 4000 m), y los restantes en tierras bajas (por debajo de 3000 m). Los conteos ma´ximos fueron, 64 000 Phoenicoparrus jamesi (verano de 1998), 34 000 P. andinus (verano de 1997), y 83 000 Phoenicopterus chilensis (invierno de 1998). En verano, los Phoenicoparrus jamesi se congregaron en humedales de los Altos Andes, con el 50% de su poblacioń en solo tres lagos: Colorada, Grande y Vilama; el P. andinus estuvo ma´s uniformemente distribuido a lo largo de un rango altitudinal (2500 m), y el Phoenicopterus chilensis mostro´ un patroń ma´s heterogeneo. En invierno, todas las especies se movieron Manuscript received 4 May 2006; accepted 29 January 2007. 8 Corresponding author. Present address: Facultad de Ciencias Naturales, Universidad Nacional de SaltaCIUNSa, Buenos Aires 177, 4400, Salta, Argentina. E-mail: [email protected] [276]

DISTRIBUTION AND ABUNDANCE OF SOUTH AMERICAN FLAMINGOS

277

a latitudes ma´s bajas dentro de los Altos Andes y a altitudes ma´s bajas en las planicies centrales de Argentina. Los humedales ma´s importantes de nidificacioń fueron Colorada, en Bolivia, para el Phoenicoparrus jamesi, Surire y Atacama, en Chile, para el P. andinus y Surire para el Phoenicopterus chilensis. Recomendamos continuar los manitoreos a trave´s de relevamientos estivales simultańeos, y una estrate´gia de conservacioń que considere la gran escala espacial y temporal a la que estas especies operan, incluyendo sus migraciones.

INTRODUCTION Flamingos (Phoenicopteridae) are gregarious, long-lived birds that inhabit saline water environments and breed in colonies. They can travel long distances between nesting and feeding sites, even during incubation and chick-rearing, making international cooperation key for their study and conservation (Parada 1990, Simmons 1995, Johnson 1997, Grupo de Conservacioń de Flamencos Altoandinos 2001). Of the five known species, the rarest are the Puna Flamingo (Phoenicoparrus jamesi) and the Andean Flamingo (P. andinus; Rose and Scott 1994). In summer, these species breed in salt lakes high in the Andes of South America (3000–4800 m). In winter, many individuals apparently move to wetlands at lower elevations throughout the Andes, in the central plains of Argentina, and on the Peruvian coast (Hurlbert and Keith 1979, Parada 1990). The distribution of Puna and Andean Flamingos overlaps that of the Chilean Flamingo (Phoenicopterus chilensis), which is more abundant and widespread from Peru to Tierra del Fuego (Allen 1956). South American flamingos occupy wetlands subject to natural fluctuations in size, depth, and salinity, factors that affect habitat availability and reproductive success (Vuille and Baumgartner 1993, Caziani and Derlindati 2000). Additionally, particularly in their summer ranges, these flamingos must contend with mining activity and the associated demand for surface and underground water, energy production, and transportation, as well as unregulated tourism and egg collecting by local people (Parada 1992, Romero and Rivera 1997, Van Ryckeghem 1997, Rocha and Alcoba 2000, Rocha 2002). As a result, both the Puna and Andean Flamingo are listed as threatened species (BirdLife International 2004). Conservation of these species necessitates determination of overall abundance, seasonal distribution, and key nesting sites.

Before the simultaneous surveys in Argentina, Bolivia, Chile, and Peru reported in this study, there were only vague estimates of the abundance of Puna, Andean, and Chilean Flamingos (Kahl 1975). Despite pioneering studies by Hurlbert (1978, 1981) and Parada (1990), knowledge of the summer distributions of these three species was incomplete, especially in Argentina and Peru, while winter distributions and movement patterns were mostly hypothetical. For these reasons, in 1996 the Wildlife Conservation Society sponsored the first simultaneous survey of flamingos in Argentina, Bolivia, Chile, and Peru. These first preliminary population estimates (Valqui et al. 2000) suggested that: Andean Flamingo populations were in more critical condition than Puna Flamingo populations; the Puna Flamingo population was double the size of the Andean Flamingo population; the distribution of flamingos in general was more widespread that previously thought, including not only altiplano wetlands but also several wetlands outside this ecoregion during the winter; and the wintering grounds remained largely unknown. Our objectives were: (1) to estimate the abundance and distribution and document nesting events of Puna, Andean, and Chilean Flamingos in the area where these species occur in sympatry, (2) to examine seasonal variation in abundance and distribution, and (3) based on the results, to propose guidelines for population monitoring and an adequate conservation strategy for these three species. METHODS STUDY AREA

We conducted our study at wetlands throughout the potential distribution of Puna and Andean Flamingos (Parada 1990): the altiplano in the central Andes of South America in Argentina, Bolivia, Chile and Peru, the coast of Peru, and the central plains of Argentina. An appendix with elevational and climatic infor-

278

SANDRA M. CAZIANI

ET AL.

mation for each survey unit is available upon request from the corresponding author. We classified wetlands into three groups, based on elevation (Cabrera and Willink 1973): (1) high Andes (above 4000 m), (2) puna (3000– 4000 m), and (3) lowlands (below 3000 m). The high Andes and puna are cold and arid regions exposed to intense solar radiation, strong winds, and large daily fluctuations in temperature. Precipitation occurs as rain, snow, or hail. Within this desert matrix, lakes and salt flats or salares form patches of aquatic habitat in endorrheic basins. At the highest elevations, wetlands freeze completely in winter, except those at lower latitudes (i.e., in Peru) and those with thermal springs. In contrast, in the lowlands of the central plains of Argentina, and on the coast of Peru, wetlands are exposed to temperate climates, intensive agriculture and cattle ranching, and high human population density (Reati et al. 1997, Peralta et al. 2001). SURVEY DESIGN AND COORDINATION

During a meeting at San Pedro de Atacama, Chile, in December 1996, researchers from the four countries agreed on the methods for the simultaneous surveys, based on previous experience from Chile (Rodrı´guez 1996). A general coordinator and country coordinators organized the field teams and grouped wetlands into survey units. Each field team was led by an experienced counter and was assigned from one to five survey units, for which the same team was responsible in subsequent surveys. Country coordinators compiled and analyzed the data. To minimize error due to movement of individual flamingos between wetlands, the surveys were conducted simultaneously and lasted no more than one week (three days in Chile but up to seven days in Argentina). This approach allowed us to estimate the abundance of flamingos by summing the counts over all wetlands. The first time we surveyed each wetland, we used a GPS to determine its location and an altimeter to measure its elevation. We conducted four comprehensive simultaneous surveys: two in the breeding season (austral summer; 23–30 January 1997 and 1998) and two in the nonbreeding season (austral winter; 7–14 August 1998 and 23–30 July 2000). Some survey units were not visited in the initial surveys, and others were added in

summer or winter 1998. In the winter of 2000, before the terrestrial survey, we used aerial surveys to detect and photograph the position of flamingo groups on the largest lakes: Poopo´ in Bolivia, Mar Chiquita in Argentina, and Salinas Grandes in Argentina. In addition to the comprehensive simultaneous surveys, we conducted partial summer and winter counts in Argentina to improve our knowledge of the species’ distributions, and to determine which wetlands should be included in future surveys. FLAMINGO COUNTS

During terrestrial surveys, depending on the size and shape of the wetland, we stood at one or more survey points on the shore and used spotting scopes, binoculars, and manual counters to count individuals of the three flamingo species. For a given wetland, we used the same census points during each subsequent survey. For flamingo groups of #4000 individuals, we counted all individuals of each species; for groups of .4000, we counted flamingos in blocks estimated to contain 10 or 100 individuals (Bibby et al. 1992). Each count was done by at least two people, and the average was used as the estimated abundance. The results of the simultaneous surveys were analyzed separately for each survey and species, and abundances were summed over each survey unit. We counted individuals with juvenile plumage during the winter 2000 survey and, when possible, identified them to species. To estimate production of chicks 2–3 months old, we monitored nesting colonies detected from 1997 to 2000. STATISTICAL ANALYSES

To determine whether each species’ local abundance was correlated with the abundance of the other two species, we used partial correlations at a 5% significance level. We only used this analysis for survey data from the years with greater coverage (summer 1998 and winter 2000). RESULTS WETLANDS

Our surveys covered 224 wetlands in Argentina, Bolivia, Chile, and Peru, encompassing latitudes from Lago Suches in Peru (14u479S) to La


279

TABLE 1. Results of four simultaneous flamingo surveys conducted in Argentina, Bolivia, Chile, and Peru, in summer 1997 and 1998 and winter 1998 and 2000: number of surveyed wetlands and survey units, and population estimates for the three South American flamingos. Summer January 1997

Wetlands Survey units Puna Flamingo (Phoenicoparrus jamesi) Andean Flamingo (P. andinus) Chilean Flamingo (Phoenicopterus chilensis) Flamingosa Total a

47 33 39 120

93 24 619 918 087 0 624

Winter

January 1998

64 27 25 117

126 22 101 913 777 0 691

July 1998

26 14 83 124

148 29 163 722 300 0 185

July 2000

15 24 44 13 97

117 33 398 442 245 170 255

Flamingos not identified to species.

Picasa, Argentina (34u209S), and longitudes from Lagunas de Quirno, Argentina (61u259W) to Lago Parinacochas, in Peru (73u429W). Elevations ranged from 2 m at Mejıá and Punta Bomboń on the Peruvian coast to 4800 m at Hoyitos in Argentina. Half of the wetlands, mostly those in the high Andes, were partially or totally frozen in winter. High Andes wetlands that did not freeze were in Peru, at lower latitudes. We detected flamingos at 179 wetlands, 12% of which were in the lowlands, 25% in the puna, and 63% in the high Andes. TOTAL ABUNDANCE

The highest numbers counted for each species were 64 101 Puna Flamingos (summer 1998), 33 918 Andean Flamingos (summer 1997), and 83 300 Chilean Flamingos (winter 1998; Table 1). Maximum winter estimates were 26 163 Puna Flamingos, and 24 442 Andean Flamingos. More Chilean Flamingos were counted in winter than in summer (see above). DISTRIBUTION

The Puna Flamingo (Fig. 1) was the most abundant species in all summer surveys in the high Andes, and was concentrated in three areas: Surire in the north; Colorada, Khara Grande, Tara, and Lagunas de Vilama at middle latitudes; and Grande in the south. This last survey unit was added in the summer of 1998. That summer, half of the Puna Flamingo population surveyed was found in three wetlands: Colorada (15 857), Grande (11 374), and Vilama (4865). Puna Flamingos were also found at some of these wetlands in winter. In

the northern high Andes, at the Surire survey unit, the population remained over winter (2426 in 1998 and 4850 in 2000). In the central high Andes, the species was markedly less abundant in winter, although some individuals remained in the Colorada and Khara Grande survey units (8000 to 1000, depending on the year and wetland). In the southern high Andes, where lakes froze, the Puna Flamingo almost vanished in winter, apparently moving to lower altitudes in puna and lowland wetlands in Argentina. In partial summer surveys at Los Aparejos (27u179S, 68u219W) and Brava (28u2479S, 68u569W) in Argentina, we recorded numbers equivalent to 13% of the total global estimate for the Puna Flamingo, extending the species’ summer distribution to high Andes wetlands at higher latitudes in Argentina. In summer, the Andean Flamingo (Fig. 2) was found at survey units in both the high Andes and the puna. Compared to the Puna Flamingo, it showed lower spatial concentration and a more uniform pattern of distribution. In summer 1997, 50% of the total population was concentrated in five wetlands (Surire, Negro Francisco, Maricunga, Pozuelos, and Vilama). In winter, between 3500 and 4500 flamingos remained at Surire, Huasco, Coposa, and Atacama. However, the species’ distribution shifted toward lower latitudes within the high Andes and lower elevations in the puna and the Argentine lowlands. For example, in winter 2000, 39% of the population was found at Poopo´ and Uru Uru (8924 birds) in the puna, and in lowland wetlands in Argentina (4170 birds). In partial summer surveys (2000 and 2001), we discovered birds in areas that extended the Andean Fla-

280

SANDRA M. CAZIANI

ET AL.

FIGURE 1. Distribution and abundance of Puna Flamingos for all wetlands, based on four comprehensive simultaneous surveys in Argentina, Bolivia, Chile, and Peru. Names and codes for all survey units shown on the maps: Alumbrera (Alu), Aricota (Ar), Ascotań (As), Atacama (Ata), Blanca (Bla), Brava (Bra), Colorada (Co), Chucuyo (Chu), Grande (Gra), Guayatayoc (Gty), Hombre Muerto (HM), Huasco Coposa (HC), Huayrapata (Hy), Jama (Jam), Khara Grande (Kha), Lago Suches (LSu), Lagunas de Vilama (LVi), Lagunas Mellizas (LMe), Loriscota (Lor), Mejıá (Mej), Llullaillaco (Llu), Mar Chiquita (MCh), Melincue´ (Mel), Negro Francisco (NFr), Parinacochas (Par), Pastos Grandes (Pas), Pozuelos (Po), Punta Negra (PN), Runtuyoc (Run), Sacabaya (Sac), Salinas (Sal), Salinas Grandes (SGr), Surire (Su), Tara (Tar), Titicaca (Ti), Uru Uru Poopo´ (Upo), and Vilacota (Vil).

mingo’s known distribution in the high Andes south to central Argentina. The Chilean Flamingo (Fig. 3) showed a diverse pattern of distribution. It was present at

numerous survey units in the high Andes, but at lower abundances than the other two species. The highest concentrations of Chilean Flamingos were at Surire, Parinacochas, Uru Uru, and


281

FIGURE 2. Distribution and abundance of Andean Flamingos for all wetlands, based on four comprehensive simultaneous surveys in Argentina, Bolivia, Chile, and Peru. Names and codes for all survey units as in Figure 1.

Poopo´. In winter, the species increased in abundance at lower latitudes in puna and high Andes wetlands in Peru, and at lower elevations in lowland wetlands in Peru and Argentina. INTERSPECIFIC ASSOCIATIONS

In the summer of 1998, the abundances of Andean and Chilean Flamingos were positively correlated (r 5 0.81, df 5 123, P , 0.001). In winter 2000, abundances of Chilean and Puna

Flamingos were correlated (r 5 0.49, df 5 114, P 5 0.001). INDIVIDUALS WITH JUVENILE PLUMAGE AND SUCCESSFUL BREEDING COLONIES

In winter 2000, less than 1% of the Puna and Andean Flamingos we detected were juveniles. Puna Flamingo juveniles were found only in Argentina. Andean Flamingo juveniles were found in a few wetlands in Peru.

282

SANDRA M. CAZIANI

ET AL.

FIGURE 3. Distribution and abundance of Chilean Flamingos for all wetlands, based on four comprehensive simultaneous surveys in Argentina, Bolivia, Chile, and Peru. Names and codes for all survey units as in Figure 1.

Nesting by the Puna Flamingo occurred almost exclusively at Colorada, in Bolivia. The species also nested with much lower productivity in other wetlands in Bolivia and Chile (Table 2). For the Andean Flamingo, the most successful nesting sites were in Chile, although 223 fledglings were recorded in March 2001 at Los Aparejos, in Argentina. For the Chilean Flamingo, in the area of sympatry the most

productive site was Surire in Chile, with minor production in other wetlands of Chile and Argentina. DISCUSSION DISTRIBUTION AND ABUNDANCE

During summer, Puna and Andean Flamingos congregated in wetlands in the puna and high

0 0 0 220 0 0 0 0 4000 0 0 0 0 0 0 0 4500 0 0 0 0 0 0 500 7500 0 0 0 0 0 0 0 5000 400 0 0 0 0 0 0 3000 0 400 0 0 0 223 0 3000 0 2338 0 0 0 0 0 0 0 800 0 0 0 0 0 3000 0 240 0 0 0 0 0 200 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 18 159 0 0 0 0 400 0 0 0 0 0 0 0 0 0 0 11 423 717 0 0 Surire (Chile) Huasco (Chile) Atacama (Chile) Pujsa (Chile) Colorada (Bolivia) Guayaques (Bolivia) Aparejos (Argentina) Melincue´ (Argentina)

0 0 0 0 0 0 0 0

2000 1999 1998 1997 Wetland

1998

1999

2000

2001

1997

1998

1999

2000

2001

1997

Chilean Flamingo Andean Flamingo Puna Flamingo

TABLE 2. Number of 2–3-month-old chicks in successful breeding colonies of Puna, Andean, and Chilean Flamingos in South America.

2001


283

Andes of Argentina, Bolivia, and Chile. Thus, abundance of these species could be accurately monitored with simultaneous summer surveys. Information from our partial summer surveys suggests that total abundance estimates for these populations would increase by 5%–10% if we include Aparejos and Brava survey units in Argentina in simultaneous surveys. Preliminary results of surveys from 2004 suggest that summer surveys should also include some wetlands in Bolivia (southwest of Salar de Uyuni) and Argentina (province of San Juan) that were not included in these surveys. There were differences in total abundance estimates of 41% for Puna Flamingo and 71% for Andean Flamingos between summer and winter surveys. The lower winter numbers could be due to one or more of the following factors: (1) incomplete coverage of the wintering areas, (2) movement to yet unsurveyed wetlands such as rivers and floodplains, and (3) underestimation of numbers when surveying large wetlands, where Andean and Puna Flamingos coexist with thousands of Chilean Flamingos. Preliminary data from satellite tracking of Andean Flamingos (SMC and FA, unpubl. data), and long-term records from Poopo´ (Rocha 2002) suggest that the latter two factors could have the greatest influence. In aerial surveys of the larger wetlands in winter 2000, total flamingo numbers could be estimated and large groups could be located, but individuals could not be identified to species (Bucher 1992, Woodworth et al. 1997). Surveys of extensive wetlands could be improved by having several field teams working simultaneously at different survey points, and by using boats to count groups not clearly visible from the shore (either because of their distance from the shore, or because of visual obstruction by emergent aquatic vegetation). For both Puna and Andean Flamingos, distribution patterns were highly consistent across years. In both summer and winter, the sites with the greatest concentrations of individuals coincided with those identified in previous partial surveys since 1975 (Parada 1990, Rocha 1997, Rodrı´guez and Contreras 1998, Caziani and Derlindati 2000), suggesting predictable seasonal patterns of abundance and distribution. The most important new information about the distribution of these populations comes the southern altiplano in Argentina,

284

SANDRA M. CAZIANI

ET AL.

where our estimates almost doubled initial survey numbers when only one additional survey unit was added (Laguna Grande at Catamarca). In summer, Puna Flamingos were concentrated in a few high Andes wetlands, while Andean Flamingos were more dispersed and occurred in both the high Andes and the Puna. Compared to the other two species, the Chilean Flamingo showed the greatest variability in distribution and abundance, particularly in Peru and Argentina. This variability was expected, because: (1) the study area covered only part of the species’ total range, (2) the Chilean Flamingo’s total abundance exceeds that of the other two species, probably by at least one order of magnitude, and (3) the summer surveys did not include lakes in the central plains of Argentina, which are important summer habitat for the Chilean Flamingo (Bucher 1992, Romano et al. 2002). SEASONAL MOVEMENTS

Movements of most flamingos have been described as nomadic, as individuals follow unpredictable and patchy resources (Baldasarre and Arengo 2000, Baldasarre et al. 2000). However, South American flamingos use primarily discrete environments (i.e., lakes) subject to strong seasonality, so we might expect movements to be more directional and thus predictable. Puna, Andean, and Chilean Flamingos are partial elevational migrants; thus, at a regional scale (altiplano) and a supraregional scale, movements are directional. To develop a conservation strategy, it is important to understand the movement patterns of flamingos. In species with predictable, directional movements, such as South American flamingos, a site-based conservation strategy may be more appropriate than a species-based strategy, which would be more effective for nomadic species (Roshier 2003). BREEDING AND RECRUITMENT

Breeding in flamingos is affected primarily by rainfall cycles, but also by human activities. For the Puna Flamingo, there is apparently only one important nesting colony, at Laguna Colorada, Bolivia. Although we recorded breeding colonies at Salar de Tara in 1986 and 1988, with as many as 5000 chicks, recruitment appears to depend almost exclusively on Color-

ada, rendering the Puna Flamingo extremely vulnerable to human disturbances. As flamingos are long-lived, decreased recruitment for a few years would not necessarily imply population decline. However, in addition to climate periodicity, there is evidence of human alteration of nesting habitat. For example, in Chile, where Andean Flamingo breeding colonies were concentrated and where mineral and hydrocarbon exploration and exploitation have increased in the last two decades, both the number of successful breeding colonies and the total production of chicks of Andean Flamingos have declined since the 1980s (Parada 1992, Rodrı´guez and Contreras 1998). We do not know how many juvenile flamingos survive after dispersing. Although yearly production is above 10 000 chicks, we recorded a very low proportion of juveniles in our surveys, which could indicate high mortality during dispersal. On the other hand, juveniles may have been overlooked because they could not be identified to species in large wetlands with difficult access, such as Poopo´. Greatest reproductive success of the Chilean Flamingo occurred outside the breeding range of the other two species (Sosa 1999, Bucher et al. 2000). However, since 1993, recruitment for this species has increased in Salar de Surire, where it coexists with breeding colonies of Andean Flamingos (Rodrı´guez et al. 2002), and there are important breeding colonies that we have not yet evaluated at Poopo´ (ORO, pers. obs.). STATUS AND CONSERVATION

To improve our knowledge of flamingo species and ensure their protection, international cooperation is crucial (Parada 1990). The international simultaneous surveys have increased our knowledge of high Andes flamingo populations, and the summer surveys, in particular, appear to be a useful tool for monitoring populations at regional scales. Previous estimates of abundance were not based on simultaneous counts (Hurlbert 1978, 1981, 1982, Parada 1992), and cannot be directly compared with our results to examine population trends. Two broad types of strategies could be used to conserve flamingos: site-based strategies and species-based strategies. Site-based conserva-


tion strategies may be adequate if flamingos continue to congregate in the same areas and if partial seasonal migration is taken into account. Some important summer sites are within currently protected areas, particularly in Chile and Bolivia. In Argentina, protected areas are few, and vital sites such as Vilama, Grande, and Melincue´ lack effective protection. Mining projects in the altiplano are major obstacles to creating new protected areas (Sureda 2003). Important winter wetlands such as Poopo´ (a Ramsar site) and Mar Chiquita (a provincial reserve) have complex conservation problems: there are no controls on activities in their watersheds, and their present relatively undisturbed state is likely a result of the buffering capacity provided by their large size. Multinational reserves appear to be an alternative that is more attractive to governments than the creation of new national areas. A trinational reserve including parts of Argentina, Bolivia, and Chile could integrate the management of Los Flamencos National Reserve in Chile, Eduardo Avaroa National Reserve for Andean Fauna in Bolivia, and Lagunas de Vilama Ramsar Site and Pozuelos Natural Monument in Argentina, and protect important congregation sites and breeding colonies of Andean and Puna Flamingos. At the southern limit of these species’ distributions, Nevado de Tres Cruces National Park and the proposed Ojos del Salado National Park, both in Chile, could be integrated with important Argentine survey units that remain unprotected, and with Laguna Brava Provincial Reserve, also in Argentina. The second approach consists of protecting and monitoring each species directly. For example, species may be designated Natural Monuments, a category that exists in Argentina that affords protection to species regardless of where they are found. If properly implemented, the designation of flamingos as Natural Monuments would protect the species and their colonies wherever they occur, including areas without official protection and in mining areas. As a precedent, in Chile, Andean Flamingo breeding colonies are monitored and preserved even outside protected areas. Sites with large concentrations of flamingos have been surveyed with greater frequency and for a longer period of time, particularly in Chile and Bolivia, so a subset of wetlands could be

285

selected for more frequent and detailed monitoring (for example, including habitat variables and other waterbirds). By combining historical information, surveys, banding programs, and satellite tracking, we are beginning to define a biological baseline for these birds. Flamingos are flagship species and, as such, they are being used for long-term conservation of wetlands. In the future they may be considered indicators of ecosystem health. Already, their biology is teaching us the importance of applying conservation tools at appropriate spatial and temporal scales. ACKNOWLEDGMENTS This paper is the result of the enthusiastic and generous collaboration of more than 80 biologists, students, and park rangers in Argentina, Bolivia, Chile, and Peru. The simultaneous surveys and training workshops were supported by the Wildlife Conservation Society (WCS), the Convention on Migratory Species (CMS), and the Ramsar Convention (Wetlands for the Future). The partial surveys were supported by WCS, the National Geographic Society, the Lincoln Zoological Society, and Idea Wild. We thank William Conway for promoting this initiative from its beginnings. Local institutions that supported this study were, in Chile: corporacioń Nacional forestal (CONAF), la Unioń de Ornito´logos (UNORCH), Universidad Arturo Prat de Iquique, the mining companies Escondida, Quiborax, Quebrada Blanca, Sociedad Chilena de Litio S.A., Cerro Colorado, El Abra, SQM Salar, and Donã Ine´s de Collahuasi; in Argentina: Administracioń de Parques Nacionales, Consejo de Investigacioń de la Universidad Nacional de Salta, and CONICET; in Bolivia: Servicio Nacional de Areas Protegidas (SERNAP), Direccioń General de Biodiversidad (DGB), the personnel of Reserva Eduardo Avaroa, Centro de Estudios en Biologıá Teo´rica y Aplicada (BIOTA), and WCS Bolivia; and in Peru: Inkanatura and Peru´ Verde. We thank Felicity Arengo, Kristina Cockle, Niels Krabbe, Javier Lo´pez de Casenave, Patricia Marconi, Rosendo Fraga, Ana Laura Sureda, and Carlos Trucco for comments on the manuscript, and Kristina Cockle and A. L. Sureda for the English translation. We dedicate this work to the memory of Pablo Canevari, who supported our work unconditionally from several institutions, and to the first author Sandra Caziani, who devoted a great part of her life to the study and conservation of these species and their habitats. We also thank F. Arengo, G. Baldassarre, and D. S. Dobkin, whose suggestions improved earlier versions of this manuscript. An appendix with site information is available upon request from the corresponding author.

LITERATURE CITED ALLEN, R. P. 1956. The flamingos: their life, history and survival. National Audubon Society, New York.

286

SANDRA M. CAZIANI

ET AL.

BALDASARRE, G., AND F. ARENGO. 2000. A review of the ecology and conservation of Caribbean Flamingos in Yucatań, Mexico. Waterbirds 23 (Special Publication 1):71–79. BALDASARRE, G., F. ARENGO, and K. BILDSTEIN [EDS.]. 2000. Conservation biology of flamingos. Waterbirds 23 (Special Publication 1). BIBBY, C. J., N. D. BURGESS, AND D. A. HILL. 1992. Bird census techniques. Academic Press, London. BIRDLIFE INTERNATIONAL. 2004. Threatened birds of the world 2004. CD-ROM. BirdLife International, Cambridge, UK. BUCHER, E. H. 1992. Population and conservation status of flamingos in Mar Chiquita, Co´rdoba, Argentina. Colonial Waterbirds 15:179–184. BUCHER, E. H., A. L. ECHEVARRIA, M. D. JURI, AND J. M. CHANI. 2000. Long-term survey of Chilean Flamingo breeding colonies on Mar Chiquita Lake, Co´rdoba, Argentina. Waterbirds 23 (Special Publication 1):110–113. CABRERA, A. L., AND A. WILLINK. 1973. Biogeografıá de Ame´rica Latina. Secretarıá General de la Organizacioń de los Estados Americanos, Washington, DC. CAZIANI, S. M., AND E. J. DERLINDATI. 2000. Abundance and habitat of high Andean flamingos in northwestern Argentina. Waterbirds 23 (Special Publication 1):121–123. GRUPO DE CONSERVACIOŃ DE FLAMENCOS ALTOANDINOS (HIGH ANDES FLAMINGOS CONSERVATION GROUP). 2001. Priority actions for the conservation of the high Andes flamingos. Final report to the Migratory Species Convention. High Andes Flamingos Conservation Group and Fundacioń Pachamama, Salta, Argentina. HURLBERT, S. H. 1978. Andean lakes and flamingo investigations. Technical Report 1, Department of Biology, San Diego State University, San Diego, CA. HURLBERT, S. H. 1981. Andean lakes and flamingo investigations. Technical Report 2, Department of Biology, San Diego State University, San Diego, CA. HURLBERT, S. H. 1982. Andean lakes and flamingo investigations. Technical Report 3, Department of Biology, San Diego State University, San Diego, CA. HURLBERT, S. H., AND J. O. KEITH. 1979. Distribution and spatial patterning of flamingos in the Andean altiplano. Auk 96:328–342. JOHNSON, A. R. 1997. Long term studies and conservation of Greater Flamingos in the Camargue and Mediterranean. Colonial Waterbirds 20:306–315. KAHL, M. P. 1975. Distribution and numbers: a summary, p. 93–102. In J. Kear and N. Duplaix-Hall [EDS.], Flamingos. Poyser, Berkhamsted, UK. PARADA, M. 1990. Flamencos en el norte de Chile: distribucioń, abundancia y fluctuaciones estacionales, p. 54–66. In M. Parada, J. Rottmann, and C. Guerra [EDS.], I taller internacional de especialistas en flamencos sudamericanos. Corporacioń Nacional Forestal y Sociedad Zoolo´-

gica de Nueva Cork, San Pedro de Atacama, Chile. PARADA, M. 1992. Conservacioń de flamencos en el norte de Chile. Perıódo 1985–1990. Informe de proyecto. Corporacioń Nacional Forestal and the Wildlife Conservation Society, Antofagasta, Chile. PERALTA, E., M. ROMANO, L. DELANNOY, AND N. BIASATTI. 2001. Manejo integrado de cuencas hidrogra´ficas, caso de estudio: La laguna Melincue´, problema´tica y perspectivas. Ambiental 4:90–105. REATI, G. J., M. FLORIŃ, G. J. FERNAŃDEZ, AND C. MONTES. 1997. The Laguna de Mar Chiquita (Co´rdoba, Argentina): a little known, secularly fluctuating, saline lake. International Journal of Salt Lake Research 5:187–219. ROCHA, O. 1997. Fluctuaciones poblacionales de tres especies de flamencos en Laguna Colorada provincia Sud Lı´pez, departamento de Potosı´, (Bolivia). Revista Boliviana de Ecologıá y Conservacioń Ambiental 2:67–76. ROCHA, O. 2002. Diagno´ stico de los recursos naturales y culturales de los lagos Poopo´ y Uru Uru, Oruro – Bolivia (para su nominacioń como sitio Ramsar). Convencioń Ramsar, WCS Bolivia, La Paz, Bolivia. ROCHA, O., AND J. R. ALCOBA. 2000. Un ana´lisis sobre el ecoturismo en la Reserva Nacional de Fauna Andina Eduardo Avaroa, p. 53–67. In O. Rocha [ED.], Planificacioń y regulacioń del Uso Pu´blico Sustentable de los Humedales Altoandinos. Talleres de Capacitacioń para el Manejo Integrado de los Humedales Altoandinos de Argentina, Bolivia, Chile y Peru´. Wetlands for the Future–Ramsar Convention, La Paz, Bolivia. RODRI´GUEZ, E. 1996. Manual para el censo simultańeo de flamencos en el norte de Chile. Corporacioń Nacional Forestal, Antofagasta, Chile. RODRI´ GUEZ, E., AND J. P. CONTRERAS. 1998. Conservacioń de flamencos en Chile, p. 37–56. In V. Valverde [ED.], La conservacioń de la fauna nativa en Chile, logros y perspectivas. CONAF, Ministerio de Agricultura de Chile, Santiago de Chile, Chile. RODRI´GUEZ, E., J. P. CONTRERAS, N. A. POOL, A. SANTORO, I. VALENZUELA, AND B. CAQUEO. 2002. Conservacioń de flamencos en el norte de Chile. Corporacioń Nacional Forestal, Antofagasta, Chile. ROMANO, M., F. PAGANO, AND M. LUPPI. 2002. Registros de Parina grande (Phoenicopterus andinus) en la laguna Melincue´ , Santa Fe, Argentina. Nuestras Aves 43:15–17. ROMERO, H., AND A. RIVERA. 1997. La fragilidad de los ecosistemas de montanã del altiplano de Antofagasta frente a la modernizacioń econo´mica de Chile, p. 343–355. In M. Liberman and C. Baied [EDS.], Desarrollo sostenible de ecosistemas de montanã: manejo de a´reas fra´giles en los Andes. United Nations University, La Paz, Bolivia. ROSE, P. M., AND D. A. SCOTT. 1994. Waterfowl population estimates. 2nd ed. Wetlands Interna-


tional Publication 44, Wageningen, The Netherlands. ROSHIER, D. A. 2003. On animal distributions in dynamic landscapes. Ecography 26:539–544. SIMMONS, R. E. 1995. Population declines, viable breeding areas, and management options for flamingos in southern Africa. Conservation Biology 10:504–514. SOSA, H. 1999. Descripcioń del evento reproductivo del Flamenco Austral (Phoenicopterus chilensis) en Laguna Llancanelo, Malarguë, Mendoza. Multequina 8:87–89. SUREDA, A. L. 2003. Patrones de diversidad en aves de lagunas altoandinas de Catamarca, noroeste de Argentina. Undergraduate thesis, Facultad de Ciencias Naturales, Universidad Nacional de Salta, Salta, Argentina.

287

VALQUI, M., S. M. CAZIANI, O. ROCHA, AND E. RODRI´GUEZ. 2000. Abundance and distribution of the South American altiplano flamingos. Waterbirds 23 (Special Publication 1):110– 113. VAN RYCKEGHEM, M. 1997. Contaminacioń minero metalu´rgica y salud pu´blica en la cuenca del Lago Poopo´. Eco Andino 2:27–66. VUILLE, M., AND M. F. BAUMGARTNER. 1993. Hydrologic investigations in the north Chilean altiplano using LANDSAT – MSS and TM data. Geocarto International 3:35–45. WOODWORTH, B. L., B. P. FARM, C. MUFUNGO, M. BORNER, AND J. O. KUWAI. 1997. A photographic census of flamingos in the Rift Valley lakes of Tanzania. African Journal of Ecology 35:326–334.