About Oilbirds

A special attraction of the Asa Wright Nature Centre is a breeding colony of the nocturnal Oilbird, or Guacharo (Steatornis caripensis). Located in Dunston Cave, a beautiful riparian grotto on the property, this is the only easily accessible colony known of this remarkable species. The World Wildlife Fund made a substantial contribution toward the establishment of the Centre in order to protect the colony. This protection has been very successful, and Dunston Cave maintains a stable colony. On the last Christmas Bird Count in December 2015, the colony had 183 adult birds. In April 2015 there were 172 adults, 7 eggs, and 11 chicks.

The Oilbird, first described by Alexander von Humboldt in 1799 after visiting a colony at Caripe in Northeastern Venezuela, is the only nocturnal, fruit-eating bird in the World. Found only in Northern South America and Trinidad, the Oilbird spends the daylight hours roosting or nesting in caves. At night they forage in the surrounding forests for the fruit of palms, laurels, incense and camphor. They often fly as far as 75 miles from their cave in search of food.

The name Oilbird comes from the young birds which become quite fat, often 50% heavier than their parents. These fat nestlings were collected and rendered down for their oil by indigenous people, and early settlers.

May and June are the months when the highest number of nestlings are to be found.
The Oilbird Colony at the Asa Wright Nature Centre is the most accessible known colony of these unusual birds in Trinidad & Tobago.


Dunstan Cave Oilbird (Steatornis carepensis) Analysis 1969 to 2013

By AWNC Staff

Click here for the complete study including all data, figures, and tables


The Dunstan Oilbird cave is located on property owned by the Asa Wright Nature Centre (AWNC) with exact location 10o 42 55.29 N and 61o 17 51.31. Annual counts of the bird population have been conducted since 1969. From 1998 counts are done more frequently however there has not been any year when counts were done for all 12 months. Because of natural and external pressures including quarrying within the birds’ foraging range, it was necessary to analyse the existing data


An analysis on data for the population of oilbirds in the Dunstan Cave was conducted. The analysis looked at the overall population trend for the entire period. From 1998 further analysis on the eggs and chick population trends was conducted.

The overall population growth was found to be 3.5% per year. However, it was found that from the period 1971-1984 there was an unexplainable rapid increase in population growth followed by a rapid decrease and then another increase before the population growth became stable. There no clear modal figure derived due to the fact that the population experienced active growth. Ten percent (10%)  of the study time (54 years)  the population was found to be less than 100 birds; 53% of the time the population was between 100 and 150 birds: 36% of the time the population was between 150 and 200 birds. On two occasions only, the population was more than 200 birds; in both cases it was 203.

During the calendar year, the first half of the year was shown to be more important for the bird as there were more adults, eggs and chicks observed during this time. In spite of this, there was no clear correlation of statistical consequence between adult, eggs and chicks for any calendar year, suggesting that there were errors in the egg and chick counting exercises.

Information regarding the effect of environmental factors on the population for any particular year was not available for consideration. During the study period, 2013 was the only year when no eggs were observed for the first three months. This must be cause for great cause for concern since on average 48% of the total eggs counted was between January to March.

It is recommended that the counting process be improved to recognize the physiological states of juvenile and immature birds also.


Annual population averages for the period 1998 – 2013 were used together with annual counts from 1969 to 1996 to determine adult population trends.

Sample average was used to determine average annual population of adult bird, egg and chick population from 1998-2013. while data for 1969-1996 was obtained from direct counts.

The average growth rate was determined by Nt =No e rt      

A comparison of averages was done for each calendar month from1998 to 2013 for adult birds, chicks and eggs. Regression analysis with Gretl 32 was used to determine correlation among the adult birds, chicks and eggs .


Click here for the complete study including all data, figures, and tables



The rapid increase and decrease in the population numbers during the years 1971-1987 remained largely unaccounted for as there is no supporting evidence for an increased population growth from 1976-1980 as seen in Figure 1.

The effect of a predator–prey interaction in this scenario seems plausible from the view that the rapid population growth from 1974 – 1980. This could be possible if the predator population had greatly declined for a period. As the bird population increased, the predator population followed by an undetermined lagged value. As the predator continuously preyed on the bird population and its numbers began to decrease rapidly, there was also a decrease in the predator population until it both populations reached an equilibrium level.

This rational is supported in particular by the data from 1983, 1984 and 1985 when the population count decreased from 48 birds in 1983 to a single bird in 1984 and increased back 50 birds in 1985. It is possible that the predator entered the cave causing the birds to take flight temporarily prior to the 1984 count leaving one single bird and returning prior to the 1985 count. The doubling of the population from 1984 to 1989 would appear that there was an annual growth rate of 18%. Even though there is no egg and chick data for that period, the available egg and chick count which is available suggests that such rate of growth is highly unlikely. Therefore one has to conclude that as the predator problem declined, birds gradually returned to the cave over time.

During the first half of the year the counting exercises revealed increases in the adult birds, eggs and chicks.  However, the increase in adult birds was not as prominent as the increase in eggs and chicks.

Most eggs were counted in February, January, April and May respectively with an average of 4 eggs per month. The months in which the least eggs were counted were August, September, October and November respectively with an average of less than one egg. No eggs were found during August, September and October while there was an average of 1 egg for counts done in November (as seen in Figure 3).

Most chicks were found during the months of February, March, April, May and June respectively with an average of 9 chicks. September, October, November and December were the months where the least chicks were found averaging 2 chicks per count.

February was found to be the month for which egg production peaked, remaining consistent with peak chick production in April as an egg takes 33 days to hatch (figure 4).

Even as it appeared that the egg chick relationship was more evident than that of adult – chick relationship and the adult – egg relationship, there was no clear significant link among these three production parameters.

Regression analysis where each production parameter was used as a dependent variable and regressed against the other two parameters used as a independent variables supports this (Model 1 ,2 &3).

The negative correlation between egg and chick resulted from the fact that egg numbers counted were most likely underestimated. As the counting process took place, many eggs were missed as they were laid in locations which made it difficult to be observed during the exercise.

No calendar year during the study period seem to have favoured the bird population in terms of the three (3) parameters. The year 2002 however was not a good one because all of the production parameters were among the four (4) least values of that category (see table 1).


The average growth rate from 1969 to 2013 was 3.5% per annum. The cave appears to have a carrying capacity approximating 200 adult birds.

Generally the birds breed more during the first 6 months of the year with the most important months being January and February. The months of August, October and November are the months of least activity.

There was no correlation between the numbers amount of adult birds with the amount of eggs or the amount of chicks suggesting that eggs and chicks were not properly accounted for.

No available information intrinsic or external was identified to be linked to the increased or decreased in numbers of adult birds, eggs or chicks for any particular year.



During the first half of the year the birds should be counted more frequently. A fourth-nightly count will provide better data.

An attempt should be made to recognize other physiological states of the bird’s life cycle between chick and adult bird (juvenile or immature birds). This will give an indication as to the quantity of eggs that were missed during the counting process.

Remarks regarding environmental factors which could have a direct and indirect impact on adult population, egg production and chick production should be recorded.

Further studies should be conducted to determine average foraging periods and its likely predators.