Final Project Report:
              Ecology of an Established Population of Reintroduced Scarlet Macaws (Ara macao)
in Costa Rica

    Greg Matuzak, M.Sc.

1.0 Introduction

Captive breeding and reintroduction programs for parrots and macaws have been slow to develop in response to the decline in their numbers throughout the Neotropics (Beissinger and Snyder 1992, Snyder et. al 2000). Initial efforts to reintroduce two endangered parrots met with  little success, including 0% two-month survival rate of released captive-raised thick-billed parrots (Rhynchopsitta pachyrhyncha) in 1991, and only one of three released captive-raised Puerto Rican parrots (Amazona vittata)  in September 1985 survived the first week after release (Meyers et al. 1996, Snyder et. al 1994). Since then, many of the essentials for a successful reintroduction project for psittacines have been identified. These essentials include: sufficient genetic diversity, control of health and disease risks, sufficient resources of time and money to maintain captive breeding populations to provide suitable birds for release, and the development of structured pre-release protocols to minimize risks associated with predation and non wild behaviors (IUCN/SSC 1995, Snyder et. al 2000). Recent reintroduction programs for Amazona parrots and Ara macaws addressed these issues directly and found greater success with first-year post release survival rates ranging between 35% and 92%, including 45% to 50% first-year survival of recently released Puerto Rican parrots (Sanz and Grajal 1998, Collazo et. al 2003, Brightsmith et al. 2005, White et al. 2005).

Identifying the link between food resource availability, diet, and the ranging patterns of reintroduced parrots and macaws on a seasonal basis is an important aspect of parrot and macaw ecology that has implications for the conservation and management of these species (Matuzak 2004). If an area does not contain sufficient food at certain times of the year, the birds may follow food resources outside of a protected area, raising special management concerns and possibly putting the long-term viability of the population at risk (Matuzak 2004, Matuzak and Boyd 2005). The establishment and conservation of a viable population of parrots and large macaws, such as the scarlet macaw (Ara macao), through reintroduction depends on the conservation of the most important seasonal food resources for the species, a sufficiently large and genetically diverse group of birds to release, and successful reproduction by that population once established in the wild (Brightsmith et. al 2005, Matuzak 2004).

Supplemental feeding programs have proven to be important to the survival of released scarlet macaws and other parrots (Brightsmith et al. 2005). Supplemental food provides crucial nourishment during the weeks or months it takes for the birds to learn to forage well on their own and it acts as a control mechanism drawing birds back to the release area on almost a daily basis where they can be effectively monitored (Brightsmith et. al. 2005, Matuzak 2004). This is especially important when the release area contains no native or established reintroduced population of the target species. In some cases, such as with the current release program of the Puerto Rican parrot, supplemental food can be provided for only the first few weeks post release until the parrots can forage on their own and move out of the release area and socialize with other parrots (White et al. 2005). However, in the current Puerto Rican parrot release program all parrots carry radio transmitters and can be tracked during their first year so their survival, diet, health, ranging patterns, and social behavior can be monitored.     

2.1 Study Area – This study was conducted on the southern Nicoya Peninsula, in western Costa Rica. The primary area of investigation included Curú NWR, but fieldwork was also conducted in areas adjacent to Curú NWR, including the Tortugas Islands and small to large private farms (Figs. 1 and 2). Curú NWR is a private wildlife refuge and working farm located in the Province of Puntarenas (09° 47' N, 84° 56' W). The site covers 1,492 ha (3,700 acres) in an area of fragmented habitats and agricultural use, containing 70% forested habitats and 30% human-created habitats such as pastures and plantations (Schutt and Vaughan 1995).
Figure 1.  Costa Rica and the Nicoya Peninsula, location of the study site at Curú National Wildlife Refuge.
Figure 2.  Map of Curú National Wildlife Refuge and environs.
Rainfall totals approximately 2,000 mm (79 inches) per year and is strongly seasonal, with a wet season from June to November and a dry season from December to May. The average temperature throughout the year is 27.3° C (81° F) with the highest monthly mean of 29° C (84° F) occurring in March and April. The coolest months are November and December when temperatures average 26° C (79° F) (Vaughan et al. 1994). The site is located at the boundary of tropical dry forest, tropical moist forest and tropical premontane life zones (Tosi 1969). Being located at the confluence of three distinct life zones has created a diverse number of habitats on the site, which consists of a mosaic of mangroves, dry deciduous and semi-deciduous forests, mixed coconut forest, beaches, and evergreen/gallery forest along the Curú River. Pastures and plantations of teak (Tectona grandis) and mango (Mangifera indica) have also been created on the site (Vaughan et al. 1994).

2.1 Fruit Abundance and Phenology, Diet, and Foraging Ecology – The main food resources for the Scarlet Macaw in Curú NWR were well documented prior to this project from almost five years of observations by the owners of the refuge and their employees. Following the methodologies outlined by Chapman et al. (1992), we selected a subset of the 25 known tree species in the macaws’ diet to study their food resource availability throughout the year through direct counts.  The study began in August 2003.  A total of 150 trees from 16 selected species were included in the first year direct count study.  Each tree was marked with a numbered stainless steel tag, mapped using GPS, and the diameter of each was measured at breast height (dbh at 1.3 meters – 52 inches) (Fig. 3). In July 2004, an additional 27 individual trees (5 additional species) were added to the study. In total, 177 trees from 21 tree species were individually marked along nine transects. The food availability and phenology transects lay in 6 of the 7 main habitat associations in the refuge no transects through the mangrove habitat).

Each marked tree was evaluated twice a month, except for March and December 2004, and August 2005 when counts were conducted once a month. The number of fruits and seeds were counted directly using 10x50 powered binoculars in order to estimate the number of fruits and seeds each contains on a monthly basis. For trees in which the entire crown was not visible, the observed fraction was estimated and the number of fruits or seeds extrapolated to the whole crown (Chapman et al.1992). For months where surveys were conducted twice, counts were averaged for those months. In addition, the nutrient content of 21 of the primary food items was analyzed. Data were collected from August 2003 through August 2005.
Figure 3.  As part of the study, 177 trees from 21 species were marked and GPS location and dbh were recorded.  Food availability for the macaws for each tree was evaluated twice a month from August 2003 through August 2005.
Various Types of  Habitats  in Curu NWR
Mixed Coconut (human created)
Pasture  (human created)

The foraging ecology, habitat usage, and ranging patterns of the Scarlet Macaw in Curú NWR were evaluated by directly observing foraging of macaws within and outside the wildlife refuge. Foraging transects were walked morning and afternoon through the main habitats of Curú NWR a minimum of three days each week. Each habitat and season of the year was sampled equally. Whenever a single or group of macaws was found feeding, the following data was recorded: date, time, transect #, habitat type, species of parrot, species of plant or tree being eaten, # of birds observed foraging, and the plant part that was eaten. During opportunistic sightings of foraging, each location was identified by either a transect # or the general location. These data were also collected August 2003 through August 2005.

2.0 Methods 

The project methodology was based on successful projects with Amazona parrots and Ara macaws (Powell et al. 1999, Vaughan 2002, Snyder et al. 1987). Aspects of successful conservation programs for the scarlet macaw in Costa Rica, Peru, and Belize were also integrated into the project (Brightsmith 2001, Brightsmith et al. 2005, Vaughan 2002, Renton 2006). The data collected are comparable to existing data from wild populations of scarlet macaws, allowing assessment of whether the reintroduced macaws in Curú NWR are using food resources and habitats in a manner similar to that of native wild macaws.
2.3 Supplemental Feeding Study -  When the scarlet macaws were first released in 1999, they were provided suppliemental feeding for a number of months. Eventually the supplemental feeding was reduced to a few cupfuls of sunflower seeds put out each afternoon as a way of keeping the macaws coming back to the site for observation.  The objective of this supplemental feeding study was to measure the consumption of sunflower seeds by 1) the number of macaws eating supplemental food each day, and 2) the amount of time each macaw ate sunflower seeds during monitoring. Monitoring took place from 2:30 to 5:30 pm an average of 10.6 times per month from July 2003 to August 2004. During each monitoring session the following data was taken: time of first arrival and arrival of each subsequent macaw, how long each macaw ate, time last macaw left the area, and the direction of arrival and departure of each macaw.

2.4 Nesting Study –  The scarlet macaw nesting season in Costa Rica begins in January and ends between April and late May. We conducted nest monitoring in 2003, 2004, and 2005. We searched for nests by either following observed macaws flying in and outside of Curú NWR or by receiving information from adjacent landowners that macaws had begun to actively nest on their property. We monitored each active nest a minimum of two times a week, mixing up monitoring times from 7:00 to 11:00 am and 2:30 to 5:30 pm (Fig. 4). During the monitoring of each active nest, data was taken on the following macaw behaviors: the presence of either or both parents in the nest, either or both parents in the area but out of the nest, and both parents completely out of the area.

3.0 Results

3.1 Fruit Abundance and Phenology, Diet, and Foraging Ecology - The main food resources available to the scarlet macaws changed between the dry/nesting season and the rainy season over the observation period March 2004 – August 2005 (Table 1).  Of the 21 tagged tree species, a maximum of 15 species were in fruit during February and March 2005. Ten tree species were in fruit and seed in December 2004 and January 2005, while 13 were in fruit and seed in April 2005 and 10 in May 2005. In comparison, during the rainy season, the highest number of tree species in fruit and seed was 9 in June, while July, August, and September had 8 species in fruit and seed, and October and November each had 7 tree species in fruit and seed. Total food availability, the number of fruits and seeds counted on all marked trees combined per survey, showed monthly and seasonal changes. There was a very large surge in the number of fruits and seeds in March 2004 and July 2005. During both dry seasons, February and March had the highest available food, and July and August account for the highest available food during the entire rainy season in 2004. October and November had the lowest overall food availability in the study.

Figure 4.  Nesting attempts by the scarlet macaws were monitored in 2003, 2004, and 2005.  Here a pair investigates a potential nesting site.
Beach almonds are in seed year round, peaking from January to April, while the flamboyant tree is in seed from August to the end of December, making it an important resource at a time of year when overall food resources are low. Other important diet items include coconuts (Cocos nucifera), teak, wild plum (Spondias mombin), African palm (Elaeis guineensis), and cenizaro or rain tree (Pithecellobium saman). The macaws prefer small, immature coconuts (4 – 6 inches long) that are easily opened at the soft top allowing access to the water and soft pulp inside. Macaws were never found opening and eating large, mature coconuts; however, on one occasion, a single macaw was seen foraging on a mature coconut that had been previously opened by a variegated squirrel (Sciurus variegatoides atrirufus). Gavilan (Pseudosamanea guachapele) and Pride of India (Lagerstroemia speciosa) were also important food items for the macaws, both of which were found in low numbers in beach and mixed coconut habitats. 
The birds were observed foraging on 32 food plant species in 15 families, most of which are important diet items of the native, wild scarlet macaws in Carara National Park in Costa Rica (Marineros and Vaughan 1995). Only a few trees made up the most important diet items on a seasonal basis (Table. 1). A total of 600 foraging bouts were recorded during the study, and the diet comprised of seeds (73.33%), fruits (10.17%), bark (5.83%), flowers (5.17%), leaves (5.17%), dead wood (0.17%) and lichen (0.17% - Table. 2). Beach almond (Terminalia catappa) and the flamboyant tree (Delonix regia) comprised over half of all foraging bouts recorded.

Plant Parts Consumed     # of Foraging  % of Foraging                                                Bouts            Bouts
Seeds                              440       73.33
Fruits                                61        10.17
Flowers                             31          5.17
Leaves                              31          5.17
Bark                                 35          5.83
Dead Wood                                1         0.17
Lichen                                1         0.17

Table. 2 Food Plant Parts Consumed  by scarlet macaws
in Curú

Habitat Type  All Seasons    Dry Season   Wet Season

Mangrove      1.5          0.40      2.40
Beach         57.5        52.30     61.54
Deciduous     1.33        3.10      0.00
Evergreen      5.17        9.50      1.80
Semi-deciduous     0.2         0.80      0.30
Mixed Coconut    13.3         9.90     16.00
Pasture       21.0          24.0     18.05

Table. 3 Habitat usage observed during foraging bouts by scarlet macaws in Curú NWR, Costa Rica, on a yearly and seasonal basis
Throughout the year we observed the scarlet macaws using mangroves, beach, mixed coconut, pasture, evergreen, semi-deciduous and deciduous habitats, with beach, mixed coconut, and pasture habitats being the most commonly preferred for foraging (Table. 3). Habitat usage varied between seasons, with evergreen habitat and pasture usage increasing during the dry season and deciduous forest usage decreasing and mangrove usage increasing during the wet season (Table. 3). Though semi-deciduous habitat in Curú NWR is the most common habitat on the project site (that is, has the largest area within the wildlife refuge), the scarlet macaws rarely use this heavily forested habitat. The scarlet macaws appear to prefer more open habitats and the beach area where two of their favorite foods, beach almond and the flamboyant tree, are located.
3.2 Ranging Patterns and Movements of Scarlet Macaws

We observed the scarlet macaws a maximum of 8 linear kilometers (5 miles) from the release area within the wildlife refuge. We estimate that the population of scarlet macaws has a yearly home range of approximately 35 square kilometers (13.5 square miles). Interestingly, they have not been documented to the immediate southwest of the project site where the small town of Valle Azul is located, nor have they been documented in Paquera, the largest town in the area located only 5 kilometers (3.1 miles) from the project site (see Fig. 2).

3.3 Supplemental Feeding Study

The maximum number of macaws observed eating sunflower seeds was 10 during a single survey, while on 3 days not a single macaw came to the feeding area. The macaws generally arrived between 3:00 and 3:30 pm and the last macaw(s) to eat sunflower seeds left between 4:30 and 5:00 pm. After leaving the feeding station, the macaws generally flew along the beach area to look for beach almonds, and in some cases, balsa flowers and flamboyant tree seeds to eat before returning to roost in the mangroves or nesting cavities for the night. Seasonally, the numbers of macaws coming to the supplemental feeding area each day and how much they ate per visit varied. In general, fewer macaws ate sunflower seeds and they consumed less per visit between February and May when compared to the months of October and November.  

3.4 Nesting Study

The nesting season of the scarlet macaw in Curú NWR begins in mid to late January and continues until May. Between 2002 and 2005, six cavities were observed being examined by the scarlet macaws. We are certain that no chicks fledged from four out of six nests. One active nest in 2002 was identified in a large, mature ceiba tree (Ceiba petandra) and it is unclear whether a chick fledged. Macaw feathers were found from a dead bird within 300 m of the nest once it was abandoned, and it is unknown whether the feathers were from a fledged chick or from a dead adult macaw (Curú NWR Unpubl. data). Three active nests were identified between 2003 and 2004 in gallinazo trees (Schizolobium parahyba), 2 were in dead trees that eventually fell over and the other was in a live tree on a neighboring private property.

In July 2004, two young macaws without numbered bands appeared in one of the large pasture areas. Within a week, the fledglings appeared at the supplemental feeding area with their parents. It appeared that the young macaws were taught how to eat the supplemental food by their parents, and their parents would act aggressively towards any other macaws that attempted to approach the feeding area while the young macaws were eating. The fledglings had large black pupils, very smooth facial patches, were smaller in size to the adults, and had patchy plumage (Fig. 5). They also spent much more time than the adults cracking and eating the sunflower seeds.   We suspect the young macaws fledged in May 2004, but we could not identify the area or tree from which they fledged. This is the first documented case of captive-bred macaws released to the wild successfully fledging offspring. In 2005, we documented one active nest in a cavity of an extensive red mangrove (Rhizaphoramangle) forest; however, the pair of macaws abandoned the site soon after first observation.

Figure 5.  One of the two fledgling scarlet macaws produced from an unknown nest about May 2004.
4.0 Discussion

The focus of this project was to study the links between food resource availability for the scarlet macaws in Curú NWR, their diet, habitat usage, and ranging patterns in order to make recommendations for the management of the population and for future macaw releases. This study showed that the macaws shift habitats seasonally to exploit available food resources and they change their use of supplemental food based on the availability of wild foods in the environment. This has important management implications for the macaws and the areas they use that lie outside of the refuge, which include both foraging areas and potential nesting sites. Once the macaws leave the refuge in search of food or nesting cavities, their level of protection declines and refuge managers and researchers have less of an ability to monitor the macaws. Since areas adjacent to the refuge are populated, there is a higher risk of poaching and harassment since the macaws come into contact with more humans outside of the protected area.   
Habitat usage appears to be correlated more with the total number of fruits and seeds available to the macaws than the number of tree species in fruit and seed in each habitat type. The macaws also use the mangrove areas in the refuge for roosting during the afternoons and nights, but our method of looking at food resources underestimates the usage (and probably importance) of the mangrove habitat by the birds. During the nesting season, some of the macaws would abandon roosting in the mangroves in favor of sleeping in or near nesting cavities. Nightly roosting of wild scarlet macaws in mangroves has also been reported in Carara National Park (Marineros and Vaughan 1995).

Figure 6.  Beach almond, almendro, or tropical almond is one of the keystone food resources of the macaws at Curú NWR, even though it is a naturalized species originally from the Malay Peninsula.
During the late wet season (September to November), the time of lowest food availability in Curú NWR, the macaws feed primarily on beach almond (non-native/naturalized species) (Fig. 6), teak (non-native)(Fig.7), flamboyant tree (non-native)(Fig. 8), and wild plum (Fig.9). These are the main food items that sustain the macaws at a time of year when there is little other food available. The fruit and unripe seeds of the wild plum tree and the seeds of the beach almond tree are the two most important resources for the macaws and are considered “keystone” resources. As a result of these project findings, the owners of Curú NWR have begun to restore wild plum trees within and adjacent to the large pasture areas in the wildlife refuge so this food source will be in larger supply in the future if the population of macaws expands naturally, and hopefully, through additional releases.
Figure 7.  Teak or teca is another important food resource for the Curu´macaws, particularly during the heaviest of the rainy season, August through November.  It is originally from Madagascar but is widely planted in plantations to supply  fast growing wood.
Beach almond is very important to the diet of the two wild, viable scarlet macaw populations in Carara and Corcovado National Parks along the Pacific coast of Costa Rica, but it can’t be restored to new areas of Curú NWR because it only grows well along beaches. The few beach almond trees that are found in areas besides the beach do not grow as well and they have fewer seeds per tree than the trees found along the beach area. Teak and the flamboyant tree are also important resources for the macaws, but since both are non-native to Costa Rica neither will be restored within the refuge. Currently, there are extensive areas of teak in and surrounding the wildlife refuge, while the flamboyant tree grows along beaches similarly to the beach almond.
Figure 8.  Flamboyant tree, or malinche, produces a plethora of foot long, bean-like seed pods that are an important food resource for the Curú scarlet macaws even though it is a naturalized species originally from Madagascar.
Knowing the ranging patterns of the macaws has helped us determine the macaws’ use of adjacent farms and forests, which has allowed the project team to try to mitigate potential problems such as nest poaching on adjacent private property. When active nests have been found on adjacent private properties, we have met with the landowners and have asked permission to monitor these nests closely. We have also attempted to get landowners involved with protecting active nests on their properties. Though the macaws showed no use of commercially important crops in areas adjacent to the wildlife refuge, it is important to continue to monitor the ranging patterns of these macaws in order to determine potential problems with adjacent landowners before they arise. Teak, a cultivated species, was eaten in adjacent areas to Curú NWR, but macaws only forage on the seeds of this species and do not limit the growth of teak trees.

Unknown prior to this study, the macaws' use of the supplemental food varies seasonally. There are two potential reasons for this. First, the time of year of lower supplemental food usage corresponds to the nesting season, so macaws may be staying away from the supplemental food area when they need to dedicate more time to finding and defending nests. Second, an increase in the abundance of wild food during the dry season may give the macaws less reason to eat supplemental food, but during times of lower food availability in the wet season, the macaws may rely on supplemental food more heavily. Prior to this study, certain facts about the usage of the supplemental sunflower seed feeding were unknown to researchers and managers of the wildlife refuge. During the times of low wild food availability and high supplemental food intake, only 50 - 80% of sunflowers seeds were usually eaten, so clearly the macaws are not dependent on them and use them simply as a supplement to their wild diet. During times of high wild food availability, the macaws would consume 0 – 60% of the seeds left for them.

Prior to this study, it was also presumed that the macaws fed on the sunflower seeds immediately before heading off to roost for the night in nearby mangroves.  Our observations show that the timing of the macaws’ attendance at the feeding stations has more to do with the time of the afternoon the seeds are typically put out rather than time of going off to roost. We observed on most days that after the last macaws left the feeding area by 4:30 to 5:00 pm, they would fly along the beach looking for beach almonds, and in the late wet season and early dry season, the macaws would look for seeds of the flamboyant tree and balsa flowers along the beach until the sun went down. They would fly off to roost after that.

Figure 9.  Jobo or wild or hog plum is a natural Costa Rican species eaten by the scarlet macaws, humans, and other types of wildlife.  It is an important food resource for the macaws during the heaviest part of the rainy season and is being used in reforesting efforts in Curú as a result of the findings of this study.
We suggest that any subsequent releases of captive-raised macaws in Curú NWR should include a supplemental feeding program for at least the first 6 – 12 months until they are established with the resident population and are feeding entirely on wild food items. We also suggest that future releases of macaws in Curú NWR be conducted with radio telemetry (or potentially satellite telemetry, a technology that we are currently testing on captive scarlet macaws with funds granted by the Loro Parque Foundation) so that the movements and ranging patterns of these macaws can accurately be assessed. Radio telemetry can be used to help locate and rescue newly released macaws that become weak and are in need of being saved from starvation or predation. In southwestern Costa Rica, a scarlet macaw released with a radio telemetry unit became very weak and would have died of starvation, but the macaw was saved when researchers were able to locate the macaw with radio telemetry far from the release area (Hilburn and Higgins 2000).      

Based upon the results of this study, several management options have been recommended to managers of Curú NWR and several are currently being implemented. Recommendations have included enhancing reproduction with artificial nest boxes (none have been used to date); reducing nest poaching (through monitoring of nest sites, expanded education programs, and better enforcement of laws in the area); reforestation with native species of trees that enhance macaw habitat and increase food availability at times of food scarcity in the area; and additional scarlet macaw releases to augment the small established population with the goal of establishing a viable population in the project area. Reforestation has been implemented and we hope that it will continue. Close nest monitoring and some environmental education programs have been developed in local communities adjacent to Curú NWR. We recommend that the current management of teak plantations within and adjacent to Curú NWR include rotational harvests where large areas of mature trees are allowed to continue to flower and seed each year. Though this species is non-native, it is being more widely planted on the Nicoya Peninsula and it provides an important food source, not only for the macaws, but for the entire parrot community and other seed eating wildlife during a time of overall low food availability. Given the fact the areas surrounding Curú NWR are fragmented by villages and small to large farms, it will be important to maintain and restore wild food resources for the scarlet macaws if a viable population is to be established in the area.


We thank Fiona Dear, Bernadette Bezy, Laura Palasz, Sigrid Collado, Olivier Horiot, Jerome Hillaire, Cynthia Shafer, and all project assistants for their dedication to assisting with the fieldwork. Thanks to the Ministry of Environment and Energy in Costa Rica (MINAE) for providing a research permit to conduct this study in its entirety. We thank Curú Wildlife Refuge for their support of the project and permission to conduct the study on their property. Funding provided by Loro Parque Foundation, Lincoln Park Zoo, Bird Clubs of Virginia, Columbus Zoo, Kaytee Avian Foundation, Amigos de las Aves USA Parrot Conservation and Research Fund, and the following private donors: Tracey Coryell and Jeff Kidston.

Brightsmith, D. J. 2001. The Tambopata Project: Annual Report 1999-2000. AFA Watchbird 28(3): 24-32.

Brightsmith, D., Hilburn, J., del Campo, A., Boyd, J., Frisius, M., Frisius, R., Janik, D. and Federico Guillen. 2005. The use of hand-raised Psittacines for reintroduction: a case study of scarlet macaws (Ara macao) in Peru and Costa Rica. Biological Conservation 121: 465 - 472.

Chapman, C. A., Chapman, L. J., Wrangham, R., Hunt, K., Gebo, D. and L. Gradner. 1992. Estimators of  fruit abundance of tropical trees. Biotrpica 24(4): 527-531.

Collazo, J. A., White, T. H., Vilella, F. J. and S. A. Guerrero.  2003. Survival of captive-reared Hispaniolan Parrots released in Parque Nacional del Este, Dominican Republic. The Condor 105: 198-207.

Hilburn, J. and Katie Higgins. 2000. Nature Restoration Foundation's Scarlet Macaw Conservation Project Survivorship Update. Nature Restoration Foundation. 4 pp.

IUCN/SSC. 1995. Guidelines for re-introductions. Approved by the 41st Meeting of the IUCN Council, Gland, Switzerland. May.

Marineros, L., and C. Vaughn. 1995. Scarlet macaws in Carara. Pages 445-468 in J. Abramson, B. L. Speer
and J. B. Thomsen. The Large Macaws: Their Care, Breeding and Conservation. Raintree Publications, Fort Bragg, CA.

Matuzak, G. 2004. Scarlet Macaw (Ara macao) Restoration and Research Program in Curú National Wildlife Refuge, Costa Rica. Proceedings from the 29th Annual Convention of American Federation of Aviculture, San Francisco, California.

Matuzak, G. and J. Boyd. 2005. Ecology and Diet of Six Parrot Species in Costa Rica. Proceedings from the 30th Annual Convention of American Federation of Aviculture, Miami, Florida.

Matuzak, G. and D. Brightsmith. 2006. Yellow-naped parrot in Costa Rica: estimating the size and recruitment of threatened populations. Submitted to the Journal of Field Ornithology.

Meyers, J. M., Arendt, W. J. and G. D. Lindsey. 1996. Survival of radio-collared nestling Puerto Rican Parrots. Wilson Bulletin 108(1): 159-163.

Powell, G., Wright, P., Aleman, U., Guindon, C., Palmenteri, S. and R. Bjork. 1999. Research Findings & Conservation Recommendations for the Great Green Macaw (Ara ambigua) in Costa Rica. Tropical Science Center. San Pedro, Costa Rica. 28 pp.

Renton, K. 2006. Diet of adult and nestling Scarlet Macaws in southwest Belize, Central America. Biotropica 38(2): 280-283.

Schutt, Adelina and Christopher Vaughn. 1995. Incorporating wildlife into development: The case of the Curu wildlife refuge and farm, Costa Rica. In Integrating people and wildlife for a sustainable future. Proceedings of the first international wildlife management congress. John A. Bissonette and Paul R. Krausman. The wildlife society. Bethesda, Maryland.

Sanz, V. and A. Grajal.  1998.  Successful reintroduction of captive-raised Yellow-shouldered Amazon Parrots on Margarita Island, Venezuela. Conservation Biology 12(2): 430-441.

Snyder, N. R. F., J. W. Wiley, and C. B. Kepler 1987. The parrots of Luquillo: natural history and conservation of the Puerto Rican Parrot. Western Foundation of Vertebrate Zoology,
Los Angeles, CA.

Snyder, N, Koenig, S, Koschmann, J, Snyder, H, and T. Johnson. 1994. Thick-billed Parrot releases in Arizona. The Condor 96(4): 845-862.

Snyder, N, P. McGowan, J. Gilardi and A. Grajal, eds.  2000.  Parrots:  Status Survey and Conservation Action Plan 2000-2004. International Conservation Union, Gland, Switzerland. 

Tosi, J. 1969. Ecological life zone map of Costa Rica. Tropical Science Center, Costa Rica.

Vaughan, C., Gamboa, J. F., Hernandez, D., Mora, J. M., Sanchez, P. E. and Henry Chaves Kiel. 1994.

Evaluacion de los recursos biologicos y diseno de corredores para los distritos de Lepanto, Paquera, y Cobano del Canton Central de la Provincia de Puntarenas. Universidad Nacional, Heredia, Costa Rica. Informe Final. April. 159 pages.

Vaughan, C. 2002. Conservation strategies for a scarlet macaw (Ara macao) population in Costa Rica. Ph.D. diss., University of Wisconsin - Madison, Madison, Wisconsin.

White, T. H., Collazo, J. A. and F. J. Vilella.  2005. Survival of captive-reared Puerto Rican Parrots released in the Caribbean National Forest. The Condor 107: 424-432.

Scarlet Macaws at feeding station.
Here we report on the post-release (2003 – 2005, year 5 to 7) ecology of a small population of scarlet macaws in Curú National Wildlife Refuge (Curú NWR) in western Costa Rica, Central America (Fig. 1). The current population of scarlet macaws resulted from an experimental release of young captive-bred, hand-reared birds in a joint project by the Costa Rican non-profit organization Asociación Amigos de las Aves and Curú National Wildlife Refuge in 1999. Originally, 13 scarlet macaws aged 1.7 to 3.7 years (average age 2.7 years) were released in Curú NWR. The release technique was a “soft release” where the macaws were first kept in a large flight for a number of months, transitioned from their captive rations to natural foods during that time, and then provided supplemental natural foods for several months post-release at a feeding station.  The amount of supplemental food was gradually reduced until only a few cups of sunflower seeds were placed at a feeding station to attract the birds back for monitoring.  First-year survival of the macaws was 92% (one macaw was lost the first year) and annual survival post first-year was 96% up to 4.2 years post-release (Brightsmith et al. 2005).

Within several months of release, the macaws were well established in the wildlife refuge and foraging independently. During the first year post-release, some of the macaws began to pair bond and even entered wooden barrels and small wooden nest boxes placed in the release area. In total, 3 pairs formed and active nesting attempts in natural cavities began in 2002, 3 years after release. However, no active nest was known to fledge young successfully in 2002 or 2003 (Brightsmith et al. 2005, Matuzak 2004). As of 2006, the population consisted of 11 macaws, including 9 adults and 2 juveniles fledged in 2004, 5 years after release.  

The native parrot community along the Nicoya Peninsula included the scarlet macaw until the late 1960’s when human persecution drove them to local extinction (Julietta Schutt Valle, local landowner since 1950’s - personal communication).  The area contains 5 additional wild, native parrots, including the threatened yellow-naped parrot (Amazona auropalliata auropalliata). Prior to this work some expressed the opinion that the yellow-naped parrots bred prolifically in Curú NWR, but our work has shown that they breed mostly on offshore islands and remote areas of the peninsula, while foraging in Curú NWR most of the year (Matuzak and Brightsmith 2006).
In 2003, Greg Matuzak, a conservation biologist, began a 3-year research project to study the ecology of the scarlet macaws to determine their diet, foraging ecology, habitat usage, ranging patterns, nesting ecology, and use of supplemental food. The major goals of the research were to: 1) evaluate the baseline ecology of the macaws, 2) determine any special conservation needs of the small population, and 3) determine whether future releases would be feasible and warranted given the initial success of the project. With the backing and support of Amigos de las Aves USA Parrot Conservation and Research Fund, a US 501 (c)(3) corporation, Curú NWR, and MINAE (the Costa Rican government Ministry of Environment and Energy), the project attracted interest and funding from several American zoos, bird clubs, and foundations, as well as the Loro Parque Foundation in Tenerife, Canary Islands, Spain. During the 3-year project, three students from the University of Costa Rica and Latin University of Costa Rica were trained and worked on the project for research experience and University credit, while 23 additional international volunteers and 7 Costa Ricans were also trained and worked on the project.

Curú macaws exploring old wooden barrel.
Dry Season (December to May)

Scientific NameCommon Name    Part Eaten
Terminalia catappa (nn)  Beach Almond       Seeds
Delonix regia (nn)   Flamboyant Tree    Flowers, seeds
Cocos nucifera (nn)Coconut  Water, young coconut
Pithecellobium saman          Rain Tree, monkey pod  Leaves, flowers, seeds
Elaeis guineensis (nn)    African Palm          Fruits, seeds
Guazuma ulmifolia Guacimo        Seeds

Wet Season (June to November)

Scientific NameCommon NamePart Eaten
Terminalia catappa (nn)         Beach Almond    Seeds
Delonix regia nn)           Flamboyant Tree  Flowers, seeds
Cocos nucifera (nn)       Coconut       Water, young coconut
Pithecellobium saman          Rain Tree, monkey pod        Leaves, flowers, seeds
Spondias mombin  Wild Plum    Fruits, seeds
Tectona grandis (nn)      Teak    Seeds
Pseudosamanea guachapele GavilanSeeds, flowers, bark
Cavity in dead tree that attracted scarlet macaw investigation in 2003. Macaw head is just visible at lower edge of cavity.  No young were produced.
Table. 1 Main Seasonal Food Items in the Diet of the Scarlet Macaw in Curú (nn: non-native).  Notice how important non-native species are.
Amigos de las Aves USA --
The Parrot Fund
"Wild Parrots --
Keep Them Flying!"