Ollie Bartlett - Coordinator of Environmental Research
Over the course of the last 100 years, jaguars (Panthera onca) have seen a decline in their overall range by approximately 54%. They have been locally extirpated in many parts of their current range states, whilst in others, such as Uruguay, they are now completely extinct. The four leading causes of this decline: conversion of natural habitats into either monocultures (soy, rice, maize) or grazing lands, retaliatory killing by ranchers for acts of depredation, hunting of prey species for either commercial or subsistence use, and finally illegal trophy/sport killing.
To combat this decline, during the last years conservation scientists have identified 90 core jaguar populations, which if linked via 182 dispersal corridors may help to save the jaguar from extinction. Eight of those populations were classified as critical, both biologically and for the integrity of the overall network. One of those eight populations is in Bolivia; JCU 77, a vast area comprised of two protected areas, Reserva Rios Blanco y Negro and Parque Nacional Noel Kempff. However, the authors of the original document accepted that these corridors were based on models, and that actual field surveys were required to document the presence of both jaguars and their prey species within corridors. The objectives of this study were to do just that.
Whilst WCC Ambue Ari is not actually within one of the dispersal corridors, its close proximity (150m) means that species are likely to be homogenously distributed between the reserve and the corridor, and as such it may be considered a surrogate for surrounding forest health. We wanted to provide an abundance index of potential jaguar prey species over the course of three years, estimate the likely number of jaguars that might be observed within the reserve based on previous studies within the body of scientific literature, and finally estimate the actual number of jaguars within the reserve over two consecutive years. WCC Ambue Ari, and by proxy the neighbouring corridor, could be considered to be suitable location for future jaguar conservation efforts if a relatively complete body of prey species could be documented, with no significant reduction between years, and if final estimates of jaguar abundance were equal to or above that which could be predicted by previous studies, with no reduction in their estimated abundance between years
The reserve of WCC Ambue Ari, owned and operated by Comunidad Inti Wara Yassi, lies on the borders of the states of Santa Cruz and the Beni. It is located at the meeting point of three distinct ecoregions; the northern Chiquitano dry forests, the southern Amazon moist forests, and the Beni savannahs. Covering an area of approximately 9.5km2 , the reserve is surrounded by seasonally flooded Pampas grasslands and dry forested hills.
However, since the reserve’s purchase in 2002, intensive forms of mechanised agriculture have increased within the local area, with year round production of soy, rice, and maize destined for internal markets. Cattle ranching is also on the increase. Yet the land of WCC Ambue Ari has so far been spared this fate, due to its role as a refuge for wild animals sequestered from the illegal wildlife trade. The reserve cares for a variety of both carnivorous and herbivorous species, including 22 small-to-large bodied felids.
In order to document the number of potential jaguar prey species utilising the reserve, 26 and 46 camera traps were installed throughout the reserve (2016 & 2018 respectively) over periods of six to eight weeks. A relative abundance index (RAI) was calculated for each species using the formula above, taken from O’Connell et al. (2011). Finally, differences in prey species RAI (for all species observed in both survey years) were modelled using a paired samples t-test with 10,000 bootstrap resamples. Mean differences were considered significant if 𝑎<.05.
In order to predict how many jaguars might be found within the reserve, data were extracted from a review of jaguar abundance estimates, conducted by Tobler & Powell (2013). Only studies conducted in an area of <250km2 were considered. Five variables were found to be constant throughout 46 studies. Data on camera minimum convex polygon (MCP), number of camera stations (stations) and average camera spacing (spacing) per survey were considered predictors of both number of individual jaguars captured (captures) and estimated abundance (estimate). Seven multiple regression models were constructed for each dependent variable, and then ranked according to their AIC value, adjusted for small sample size. Parameter estimates were averaged according to the relative weight of model support, and final predictive models were constructed using all parameters found to be significant predictors of the dependent variables.
In order to estimate actual jaguar abundance within the reserve, 13 camera stations (paired) were installed throughout the reserve in both 2016 and 2017. Individual jaguars were identified by variations in their pelage, and their detection histories pooled into sampling periods of 3 and 4 days (2016 & 2017 respectively). Abundance was estimated suing the program “Capture”, modelling abundance as a function of seven different models describing variations in detection probability as a result of time, behavioural response, and individual heterogeneity, as well as all possible combinations plus a null model. Finally, resulting abundance estimates were compared to predicted estimates based on derived predictive models
Over the two survey periods, 29 potential jaguar prey species were observed. 16 potential prey species were observed in both series. Seven of these were found to have an increased RAI in 2018 compared to 2016, whilst the remaining nine all decreased in their RAI value. However ,the average decrease of -1.86 photographic events per 100 camera trap days was not statistically significant.
Of the 46 reviewed jaguar abundance estimates, 14 were derived from study areas located within Bolivia. The average number of camera trap stations deployed, the average spacing between cameras (km), the average area of the camera trap polygon (km2 ), average number of individual jaguars captured on camera, and the average abundance estimate, are displayed in the bottom table. Averages for studies based in Bolivia are presented in bold. Key figures for Bolivia; the average size of the study areas was approximately 70km2 , with each study observing an average 5.26 individual jaguars, with an average estimated abundance of 6.21 jaguars.
Following the modelling process, two models from each series received strong to moderate support. The remaining five models from each series received little to no support. Following the averaging of parameters based on each model’s relative weight of support, two variables (area of camera trapping polygon and distance between cameras) were found to be significant predictors of both the number of jaguars observed, and resulting estimated abundances, from the sample of 46 camera trapping studies. These two parameters were used to construct final predictive models.
Results 2: Model predictions
Using these two models, the predicted number of jaguars that might be observed within the reserve was 0.56-0.70, while the predicted estimated abundance of jaguars utilizing the reserve was 0.75-1.11. Essentially, no more than one jaguar could be expected to be found within WCC Ambue Ari. Variations in these predictions are caused by varying size of camera polygon, and distance between camera stations, between survey years.
However, in 2016 five individual jaguars were observed, 28 times over the course of 33 days. Following the modelling of detection probability, this resulted in an estimated abundance of six jaguars utilizing the reserve. These results are 8.00-8.92 times higher than that that could be predicted by the highly significant models. Additionally, two individuals were observed mating on camera, while eight ocelots were observed 94 times, resulting in an estimated abundance of 9 individuals.
In 2017 six individual jaguars were observed, 43 times over the course of 68 days. Following the modelling of detection probability, this resulted in an estimated abundance of eight jaguars utilizing the reserve. These results were 8.57-8.79 times higher than that that could be predicted by the highly significant models. Additionally, nine ocelots were observed 177 times, resulting in an estimated abundance of 11 individuals.
Within WCC Ambue Ari exists a near complete assemblage of medium-to-large bodied jaguar prey species, including key ungulate prey species such as collared peccary and red brocket deer. Additionally, there is evidence of breeding populations of Brazilian tapir and giant anteater, both classified as endangered by the IUCN. There is a large population of ocelots (relative to the size of the study area), and two other species of felid (puma and jaguarundi) are regularly observed. However, while their relative abundance would seem stable in time due to their being no significant reductions between years, certain key prey species (collared peccary, red brocket deer, and spotted paca) were all found to have declined in relative abundance between years. While this may have been the result of some kind of systematic variation due to camera locations or seasonality, these three prey species are known to be among the most hunted by man within the local area. Any actual change in their abundance may have negative consequences for jaguars. Additionally, hunting of endangered Brazilian tapir is known to be locally common.
While no significant changes were detected in species’ RAI between years, the use of RAI as a measure of abundance has been debated. An RAI is not a measure of true abundance, but rather a measure of the rate of photographic capture, and the index itself is open to a number of sources of bias. It can be calibrated, however the process in incredibly labour intensive, demanding multiple camera trapping and distance sampling surveys for each individual species. Alternative methods are available for example the Royle/Nicholls abundance induced heterogeneity model, however this requires all survey sites to be independent from one another. Given the small area of WCC Ambue Ari and the large home ranges of species such as collared peccary and red brocket deer, this method was deemed not suitable for the study. In recent years random encounter modelling has come to the fore, however the methods are still relatively untested, demand high investment in top of the range cameras, and are statistically intensive. As such they were deemed outside the scope of this study.
In terms of the local jaguar population, nine individuals were observed over a collective 101 days. Five of these individuals have been observed across multiple years. One camera station observed four different individuals at the same location within a 48 hour period. Additionally, a 10th individual has recently been observed, with six individuals having been observed during the last two months. If their abundance were to be estimated at present, we would likely see similar estimates as 2017. All this points to a stable and resident population. However in spite of all this, we still know almost nothing about this population. We have no idea as to their home ranges, how they respond to anthropogenic pressures both within and outside the reserve, or their vital rates. A much expanded, long-term camera trapping survey within the wider landscape is required so that we can understand these parameters necessary for the conservation of these jaguars.
The results presented here would seem surprising given the limited size of the study area. The 2017 abundance estimate is the second highest observed within Bolivia (based on 14 reviewed non-SECR studies) using the same methodologies. Precision (SE/Estimate) is inversely related to estimated abundance (𝑟 2=.576, p=.008), and yet in spite of the 2017 estimates being the second highest recorded in Bolivia, they are also among the most precise. All of this can be found within a study area of just 5% that of the average study area in Bolivia. Only a single survey in Bolivia has recorded a higher estimated abundance (n=13), however the study area in question was 51.28x the size of the study area within WCC Ambue Ari. To further support these estimates, our raw data, number of jaguars observed, is 8.57-8.92x times than that predicted by our highly significant models.
While standard capture-recapture methods used in this study are robust, given the limited objectives of the study (i.e. no estimate of jaguar density over a larger scale), the methods themselves have now been superseded. New methodologies exist, incorporating the spatial movement of individuals within the landscape, modelling detection probabilities (𝑝 ) as a function of capture distances from a hypothesised home-range centre, in order to provide robust and standardised estimates of density across a wider area. These methods however require much larger study areas than was possible within this study, with estimates becoming unbiased at spatial scales of >150km2 . At present these methods are simply beyond the scope of our current equipment, manpower, and funding. However, until the time when we are able to fully commit both physically and financially, we may never be able to give this potentially highly unique population of jaguars the focus, and conservation attention, that they might deserve.
To summarize, within WCC Ambue Ari, a small, privately protected reserve on the frontier of agricultural expansion, there exits a resident, breeding population of jaguars, highly unusual in their abundance. While it is clear (given the large home ranges normally associated with the species) that WCC Ambue Ari makes up just a small part of their territories, the fact that within such a relatively small area so many individuals might be present in the same space and time, consistently throughout the years, is something that should be considered unique. This, coupled with an unusually high ocelot population, as well as a diverse, stable prey population, 150m from a potential jaguar dispersal corridor and 45km away from one of the eight most important JCU’s throughout their range, highlights the importance of the surrounding area towards the future of jaguar conservation, not only in Bolivia, but across all of south and central America.
Within this image we can see the two strands of the planned jaguar dispersal corridor, almost completely fractured by the expansion of intensive mechanised agriculture. In light red is displayed the boundaries of several community agricultural lands, a proposed mixture of intensive and small-scale community agriculture, cattle ranching, and selective logging. If some form of conservation action is not implemented, the proposed jaguar corridor will be rendered unusable within the coming years.
Intensive mechanised agriculture can be observed the entire distance along the road, fractuing the proposed corridor. Many of these agricultural lands span distances of up to 5km from forest edge to forest edge. Assuming jaguars have circular home ranches, this equates to multiple individual areas of ~20km.Under the lowest estimates of jaguar individual territories, this line of agriculture represents whole home ranges being converted. The reserve of WCC Ambue Ari and the neighbouring hills may be the last surviving parts of the proposed corridor situated along the road, the last linkages between populations north and south. As such, their continued preservation may be vital for jaguars within this area.
However, the area itself may in future years come under increased threat. By the year 2025, Bolivia plans to significantly increase its agricultural output, becoming a net exporter rather than importer. Under these plans, within the next eight years agricultural lands will be increased from 3m hectares to 13m hectares, livestock production will be tripled, 50% of the public will have some form of participation within community or large-scale agricultural schemes, and much of this land used for this is proposed to come from the areas of Santa Cruz and the Beni.
Under current modelled scenarios, much of the forest surrounding WCC Ambue Ari, and both within, and outside the corridor itself, will be converted. By the year 2025 as much as 40-50% may be converted. By the year 2050, possibly as high as 70-80%
In order to protect these jaguars, as well as the entire community of species living within their vast home-ranges, we need to understand their populations better. We require large-scale standardised density estimates of both this jaguar population and their prey species, so that their relative importance, and the importance of this area to long-term jaguar conservation, may be understood and presented to conservation planners. We require an understanding of how different types of agriculture, both community based and intensive, may impact rates of jaguar occupancy, so that we can develop tools to mitigate these impacts.
“We must pool our resources and join forces – not only to save the jaguar, but the ecosystems in which they are part.”
These jaguars have been observed in a small area of land, surrounded on three sides by intensive agriculture, yet only 150m outside of a proposed jaguar corridor and 45km outside of JCU 77, of the eight most important locations for jaguar conservation across their entire range. I believe that these results suggest that there likely exists a large, ecologically functional, yet unstudied population within the surrounding area. We need to reach out, to work with local communities, governments, and conservation NGO’s, to find ways of achieving community goals, whilst at the same time protecting these jaguars and the ecosystems of which they are a part.
For more information, please send us a message to email@example.com
By: Ollie Bartlett - Coordinator of Environmental Research
New Howler Monkey Enclosure, Thanks to IPPL! The International Primate Protection League (IPPL) funded the construction of a new Howler Monkey Enclosure at Ambue Ari Sanctuary.
Current results and discussion from the research underway at Ambue Ari - VIII Bolivian Congress of Mastozoology and IV Latin American Congress of Mastozoology - La Paz 11.07.2018 Potentially unusual jaguar abundance within a 3.9km2 section of WCC Ambue Ari and the potential importance of surrounding landscapes to future jaguar conservation. Ollie Bartlett - Coordinator of Environmental Research
Socialization Programme for our Capuchin Monkeys News about our socialization programme at Machia
CIWY research projects An update on the progress of our research proyect in Ambue Ari
Working as a vet with CIWY Read more about Marta's experience as a vet at CIWY
My experience at CIWY Karen shares with us what she learned and lived while being at CIWY
Tejon Garden Update Check out how the construction of our Tejon garden goes
A new enclosure for Ivan Our tayra at Machia gets a new space
Back to Freedom The history of 7 boas
Artist in Residence 2017/18 Meet Silvia and her art
Calling all Artists interested in #conservation and #wildlife! Applications are now open to join our Artist in Residence programme in 2018.
News from Machia - November 2017 Snakes, Caimans and insects at WCC Machia
News from Jacj Cuisi - November 2017 Capitan's enclosure and our vegetable garden
News from Ambue Ari - October 2017 Information about our new squad member, the construction of the Tejon Garden and Maggie
News from the research project at WCC Jacj Cuisi Habitat preferences of the endangered lowland tapir (Tapirus terrestris)at WCC Jacj Cuisi.
CIWY is celebrating 25 years as an organization, with already 21 years in Machia. In Ambue Ari we have been working for the past 15 years and also in Jacj Cuisi it has been already 8 years now.
Research Project - Jacj Cuisi An update from our Research Project in Jacj Cuisi
News from Ambue Ari - Mayo 2017 News from our WCC Ambue Ari including saying goodbye to Sama, constructions on Leo's management cage and new members.
Come and join us at Pete the Monkey Festival! We are looking for 4 volunteers who have visited CIWY in the past to help us run a fun, awareness-raising, craft activity
Friends of Inti Wara Yassi (FIWY) is our sister organization in the UK. They have been a major source of support since their founding in 2008. Find out more about FIWY here.
Quest Overseas organizes gap year trips for British and international students. Since 2001, Quest has worked with CIWY to bring much needed volunteers and funds. If you are interested in the programs they have with us, find out more here.