How many pangolins are left in the wild?

No reliable global population estimate exists for any of the eight pangolin species. This is not evasion — it reflects a genuine gap in scientific data. Pangolins are nocturnal, largely solitary, and live at low densities across vast, often remote habitats. Their burrowing habits and overlapping ranges with other species make standard wildlife census methods unreliable. The IUCN Red List assessments for pangolins are largely based on trade data and suspected population declines rather than direct counts. Scientists know pangolins are declining across their range based on reduced encounter rates and seizure trends, but precise numbers remain elusive.

Do camera traps work for counting pangolins?

Camera traps are the most widely used method for detecting pangolin presence and estimating relative abundance, but they have significant limitations for producing precise population counts. Pangolins trigger cameras at low rates compared to larger mammals due to their small body size, slow movement speed, and the tendency to hug terrain rather than walk upright through camera frames. Occupancy modelling — which estimates the probability that a species is present in a given area based on detection/non-detection across camera stations — has produced useful relative density estimates for Temminck's ground pangolins in South Africa and some Asian species, but converting occupancy to actual animal numbers requires assumptions that are difficult to validate in the field.

What is the most reliable method for surveying pangolins?

No single method is considered definitively reliable for pangolin population surveys. A multi-method approach combining GPS telemetry for individual home range data, camera trap occupancy modelling for relative density, and community-based interview surveys for historical and current distribution information currently produces the most useful estimates. For species where individuals can be identified by scale pattern or scar features in camera images, mark-recapture statistical analysis provides population size estimates at the site level. Scat DNA sampling, which identifies individuals from faecal samples, is a promising newer method that is being validated for several pangolin species.

Why are pangolin populations so hard to count?

Several biological characteristics make pangolins among the most survey-resistant mammals in their ranges. They are primarily nocturnal, active for only four to six hours per night. They are naturally solitary, existing at low densities of typically less than one individual per square kilometre. Their foraging areas are large and overlapping, making territory-based counting approaches inapplicable. They burrow and spend daylight hours underground or in dense vegetation, making direct observation surveys ineffective. They do not vocalise reliably, removing acoustic survey options. And their small body size makes detection by camera trap, aerial survey, or visual transect disproportionately difficult compared to larger mammals.

How do researchers track individual pangolins?

Individual pangolins are most commonly tracked using GPS collars or harnesses fitted during capture events. VHF (Very High Frequency) radio telemetry was the standard for many years, requiring researchers to manually locate animals with directional antennae — a time-intensive approach. Modern GPS units with satellite or GSM network upload capability allow remote tracking over months or years without repeated capture. Microchip implantation (PIT tags) is used to uniquely identify individuals at close range during veterinary assessments or recapture events. In South Africa, a national database maintained by the African Pangolin Working Group records every confiscated, rehabilitated, and GPS-tagged pangolin, providing longitudinal individual life history data across the population.

Research Methods 11 June 2026 9 min read

Pangolin Population Surveys: How Scientists Try to Count the Uncountable

Ask a conservation biologist how many African elephants remain in the wild, and they will give you a number with a confidence interval: approximately 415,000, according to the 2016 African Elephant Status Report, derived from aerial surveys of most of the species' known range. Ask the same question about pangolins — the world's most trafficked wild mammals, listed across all eight species as Threatened to Critically Endangered — and the honest answer is: we do not know, and we cannot easily find out.

This is not for lack of effort. The IUCN SSC Pangolin Specialist Group, the African Pangolin Working Group, and dozens of research teams across Africa and Asia have invested millions of field hours trying to generate reliable population data for pangolin species. The difficulty is biological: pangolins are built, almost as if by design, to be as hard to count as possible.

Why Pangolins Defeat Standard Survey Methods

The toolkit of wildlife population science was largely developed for species that can be seen. Aerial count methods work for elephants, zebra, and wildebeest — large animals that aggregate in open habitats and can be distinguished from aircraft. Point count and transect methods work for birds and diurnal primates. Even elusive forest carnivores are detectable through track stations, scent lures, and hair snares.

Pangolins resist all of these approaches. They are active for only four to six hours per night. They exist at naturally low densities — field studies in South Africa document Temminck's ground pangolins at densities of less than one individual per square kilometre in good habitat. They spend daylight hours underground in burrows or curled in dense vegetation, completely invisible to aerial or visual survey. They do not vocalise reliably, which removes acoustic monitoring as an option. They are small enough that heat signature from aerial thermal imaging is ambiguous at typical survey altitudes. And they move slowly and close to the ground, which means they trigger camera traps at rates far lower than larger mammals on the same landscape.

The result is that even in areas with active research programs, detecting a pangolin in the field is a genuinely rare event. A team deploying 100 camera trap stations for 30 days in good pangolin habitat might accumulate 3,000 camera-trap nights and obtain a handful of pangolin detections — a dataset from which inference about population size requires assumptions that can significantly affect the answer.

Camera Trap Occupancy Modelling

Camera trap surveys are the most widely used method for monitoring pangolin populations, particularly for Asian species in dense forest habitats and for ground-dwelling African species in savanna and bushveld. The standard analytical framework is occupancy modelling: a statistical approach that uses the pattern of detections and non-detections across a grid of camera stations to estimate the probability that the target species occupies any given unit of habitat.

Occupancy models account for imperfect detection — the fact that an animal may be present but not photographed during a given survey period. By modelling detection probability as a function of covariates (camera placement, habitat type, season, weather), researchers can separate genuine absence from missed presence, producing more reliable estimates of distribution than raw detection/non-detection data would suggest.

What occupancy models do not directly produce is a count of individual animals. Converting occupancy probability to animal density requires additional assumptions about home range size, which must come from telemetry data on the same population. For species with well-studied telemetry datasets — like Temminck's ground pangolins in parts of South Africa — density estimates derived from occupancy and home range data can be produced with reasonable confidence intervals. For Asian species where telemetry data are sparse, the conversion is far less certain.

Camera trap identification of individual pangolins has been achieved in some studies using scale pattern variation — the arrangement of rows and individual scales on an animal's flanks is unique to each individual, analogous to a fingerprint. Where individual identification is possible, mark-recapture statistical analysis — a well-validated method for estimating closed-population sizes — can produce population estimates with quantified uncertainty. This approach has been applied to Sunda pangolins in Borneo and Temminck's ground pangolins in South Africa, producing site-level population estimates that are among the most robust data available for the species.

Line Transect and Sign Surveys

Line transect surveys — walking predefined paths and recording direct sightings of target animals — are impractical for pangolins because nocturnal direct sightings are too rare to generate usable data at typical transect sampling intensity. Modified sign-based transects, however, have proved more useful. Trained field teams walking set transects record indirect evidence of pangolin presence: characteristic digging scrapes at ant and termite mounds, claw marks on mound surfaces, faecal deposits, burrow entrances with distinctive scratch patterns, and scale fragments at feeding sites.

Sign survey data are used to produce relative abundance indices — measures of whether pangolin activity is increasing or decreasing over time in a given area — rather than absolute population counts. Over a time series of standardised sign surveys, trends in sign encounter rates can indicate population trajectory with reasonable confidence even when absolute numbers are unknown. This approach is widely used in South Africa's private game reserve sector, where regular sign surveys by rangers contribute to monitoring databases maintained by conservation organisations including the APWG.

GPS Telemetry and Individual Tracking

GPS telemetry — fitting individual pangolins with tracking collars or harnesses equipped with GPS units — provides the most detailed individual-level data available for the species. Telemetry studies have documented home range sizes, activity budgets, habitat selection, seasonal movement patterns, and the effects of rescue and rehabilitation on post-release survival.

In South Africa, telemetry studies have produced published home range estimates for Temminck's ground pangolins ranging from 54 to 484 hectares, with considerable individual variation by sex, season, and habitat quality. These home range data are essential inputs for converting occupancy estimates to density values, and they are the most direct evidence available for understanding how pangolins use landscapes at the individual level.

Telemetry does not directly estimate population size, and the logistical challenge of capturing and fitting enough individuals to represent population-level variation is significant. Capture events require specialist skills to minimise stress and injury, veterinary oversight, and regulatory permits. The number of simultaneously tracked individuals in most studies is measured in tens, not hundreds — a sample that may or may not represent the broader population distribution.

Community-Based Interview Surveys

For many pangolin species, particularly in regions where scientific research capacity is limited, community-based interview surveys provide the most practical means of assessing distribution and abundance trends. Structured questionnaires administered to hunters, rangers, farmers, and traditional knowledge holders ask about encounter frequency, habitat associations, and perceived population change over defined time periods.

Interview data carry methodological limitations: recall bias, social desirability effects (interviewees may underreport encounters if they perceive stigma around pangolin interactions), and varying consistency in descriptions of the target species. However, in areas where camera trap deployment is logistically impossible and telemetry studies have never been conducted, interview data may be the only evidence available for distribution modelling.

A study by Willcox et al. published in the journal Oryx used structured interviews across multiple West African countries to document pangolin distribution, habitat associations, and perceived population trends across a landscape where scientific survey coverage was minimal. The resulting distribution maps, while carrying wider uncertainty than camera trap data would provide, filled spatial gaps in the IUCN range map that would otherwise have been left as data-deficient.

Genetic Sampling: eDNA and Scat Analysis

Environmental DNA (eDNA) — genetic material shed by organisms into their environment through skin cells, faeces, mucus, and other biological material — has transformed survey methodology for aquatic species and is beginning to show promise for terrestrial mammals. For pangolins, eDNA from soil samples collected at known burrow sites or foraging areas can confirm species presence and, with sufficient sample size, estimate the number of distinct individuals contributing to the genetic pool.

Scat DNA sampling is a more established approach: faecal samples collected along survey transects are genotyped to identify species, sex, and individual identity. Where individual identification is achieved across multiple samples, mark-recapture analysis of the genetic data produces population estimates. The method has been validated for several secretive mammal species, including leopards and wolves, and is being adapted for pangolin species. The challenge is that pangolin scat is difficult to locate and identify in the field, and DNA degradation in tropical climates reduces genotyping success rates.

What the Data Gaps Mean for Conservation

The absence of reliable population data for pangolins has direct conservation consequences. Without a population baseline, it is impossible to measure the impact of conservation interventions — anti-poaching programmes, habitat restoration, or translocation projects — with statistical rigour. Detecting a 20 percent population decline from a baseline of unknown size is simply impossible. This makes adaptive conservation management extremely difficult and limits the ability of conservation organisations to demonstrate impact to donors and policymakers.

The IUCN Red List assessments for pangolin species rely heavily on inferred population declines derived from trade data and expert opinion rather than direct counts. This approach has scientific validity when applied with appropriate uncertainty ranges, but it means that the conservation status of pangolins rests on an evidential foundation that is weaker than most stakeholders realise.

Efforts to address this gap are ongoing. The IUCN SSC Pangolin Specialist Group's range-wide status assessment, published in 2019 and updated subsequently, represents the most comprehensive synthesis of available data across all eight species. Regional programs in South Africa, Zimbabwe, and several Southeast Asian countries are building multi-year camera trap and telemetry datasets that will eventually support more robust trend analysis. Citizen science platforms that allow members of the public to record pangolin sightings are filling distribution data gaps in regions where professional survey coverage is sparse.

The fundamental challenge remains: pangolins are among the hardest wild animals on earth to count, and they face among the most intense commercial pressures. Closing that gap — between what we know and what conservation management requires us to know — is one of the central unsolved problems in pangolin biology.