Citizen Science and Pangolins: iNaturalist as a Survey Tool

Pangolins are among the hardest mammals to survey. Nocturnal, secretive, and rapidly declining across their range, they resist most standard monitoring approaches. Camera-trap coverage is impossible at continental scale. This is where citizen science platforms, particularly iNaturalist, play an increasingly important complementary role.

Why Pangolin Surveys Are Difficult

For most wildlife species, camera trapping, direct observation, acoustic monitoring, and track surveys can generate reliable population data. Pangolins resist most of these approaches.

Direct observation is rare even for experienced researchers working in known habitat. Camera traps detect pangolins but at low rates — studies across Central Africa have found pangolin detections per camera-trap night typically below 0.01, meaning hundreds of camera nights generate only a handful of images. Acoustic monitoring does not apply; pangolins are largely silent. Track surveys work in soft substrates but require trained trackers and consistent effort.

The result is that across most of the eight species’ ranges, large geographic areas have essentially unknown population status. National population estimates for any pangolin species remain provisional or absent. The IUCN Red List assessments note this data gap explicitly.

What iNaturalist Is and How It Works

iNaturalist is a collaborative platform, jointly operated by the California Academy of Sciences and the National Geographic Society, where anyone with a smartphone can record and photograph wildlife. Each observation includes a photograph, GPS coordinates, and date. A community identification process follows, where expert identifiers confirm, revise, or flag records. Observations reaching consensus are marked “Research Grade” and shared automatically with the Global Biodiversity Information Facility (GBIF).

As of 2026, iNaturalist hosts over 200 million observations across all taxa. Pangolin observations — across genera Manis, Smutsia, and Phataginus — number in the thousands, concentrated in South Africa, Zimbabwe, and India, reflecting both species distribution and smartphone-equipped observer density.

What Citizen Science Data Reveals

Range confirmation and extension. Historical range maps based on museum specimens can be over a century old. iNaturalist observations provide current, geolocated presence data. For the Temminck’s ground pangolin (Smutsia temminckii), observations from livestock farmers in southern Africa’s semi-arid zones have confirmed persistence in areas where the species was assumed locally extinct. A single photograph with GPS coordinates submitted by a farm manager constitutes genuine scientific data.

Phenological patterns. When observations accumulate over years, patterns in encounter timing emerge. Are pangolins seen more after rain? Are certain seasons associated with road crossings — which both increase encounter probability and vehicle strike risk? Citizen science data has begun revealing these patterns for some species.

Habitat association documentation. Geolocated observations overlaid with land cover data and elevation models reveal habitat associations. Analysis of iNaturalist Temminck’s ground pangolin records shows clear associations with intermediate tree cover and proximity to water sources, consistent with termite colony distribution patterns. These associations help refine species distribution models used for protected area planning.

Detection of trafficking incidents. Several iNaturalist submissions have been photographs of live pangolins held in captivity or market stalls. Researchers monitoring the platform have, in documented cases, passed coordinates and imagery to law enforcement — an unintended but real conservation benefit.

Data Quality Challenges

Misidentification. Not all observers know pangolins. In sub-Saharan Africa, armadillos — which do not naturally occur there — are occasionally submitted. Scaled animals in poor light generate ambiguous photographs. Different pangolin species are confused with each other. The community ID process catches many but not all errors.

Observer bias. Observations cluster where smartphone-equipped observers go — roads, trails, tourist areas, farmland — not where pangolins go. This geographic bias can create misleading impressions about species distribution. Statistical methods exist to model observation effort and adjust for bias, but require information iNaturalist does not always capture.

Temporal and geographic gaps. In many range countries — including Democratic Republic of Congo, Central African Republic, and Myanmar — iNaturalist uptake is low. This reflects absence of observers, not absence of animals. Absence of data and absence of animals must not be conflated.

Duplicate records. A single well-photographed pangolin encountered by a hiking group may generate multiple separately submitted observations. Without de-duplication, abundance estimates from citizen science data can be inflated.

Integration with Professional Survey Methods

The strongest use of iNaturalist data is as a complement to professional field work, not a standalone method.

Camera-trap studies can use iNaturalist observations to prioritise placement. If citizen reports cluster around a particular valley or farm boundary, that locality is a logical first target for systematic detection effort. Conversely, camera-trap data validates whether citizen-reported presence translates into detectability under systematic conditions.

Distribution models integrating iNaturalist occurrences with remotely sensed environmental data — vegetation indices, precipitation, land use classification — have been used to predict pangolin habitat suitability across large extents. MaxEnt and ensemble models use occurrence records as presence points. More records, even from citizen science, improve model robustness.

In South Africa, the African Pangolin Working Group (APWG) operates one of the most organised pangolin monitoring networks globally. iNaturalist observations from farm workers, game rangers, and the public contribute to a consolidated national database. The APWG reviews all records, standardises coordinates, removes duplicates, and integrates confirmed observations into population trend assessments — a pipeline from smartphone photograph to peer-reviewed dataset that is a model for formalising citizen science.

How to Contribute a Quality Pangolin Observation

1. Photograph from multiple angles — dorsal (back), lateral (side), and face if safe. Scale pattern and scale count on the head help distinguish species.
2. Enable GPS — location is the most scientifically valuable metadata.
3. Record date and time — most smartphones embed this in EXIF data.
4. Note the habitat — a brief description (riverine woodland, maize field margin, rocky hillslope) adds context that GPS alone cannot capture.
5. Do not handle the animal — a curled pangolin should be left to uncurl and move away. Handling causes stress and can lead to abandonment of young.
6. Submit to iNaturalist and tag appropriately to reach identifiers familiar with the genus. In South Africa, also contact the APWG directly. Do not post exact coordinates publicly on social media — this can alert poachers.

The Future of Citizen Science for Pangolins

Machine learning is changing what is possible with citizen science imagery. Automated identification algorithms trained on verified pangolin photographs can now assign preliminary species identifications to new submissions with reasonable accuracy for well-represented species. As these models improve, turnaround between submission and Research Grade status will shorten, increasing the speed at which new observations enter the scientific record.

For one of the world’s most-trafficked and least-studied mammals, every verified pangolin record entered into GBIF is a permanent contribution to scientific knowledge that will still be queried decades from now. Smartphone-equipped observers — farmers, trackers, travellers, rangers — represent a distributed detection network no funded survey programme can replicate in scale.

Frequently Asked Questions