Camera traps for Natura 2000 monitoring

Under the Habitats Directive (92/43/EEC), Member States report every six years — Article 17 — on the conservation status of the Annex II, IV, and V species a site is designated for: presence, distribution, population trend, and the pressures acting on them. Camera traps have become an established, non-invasive way to gather that evidence for mammals, and the Untamed Data Platform structures those captures into records a monitoring team can stand behind.

What the directive asks for

Article 17 rates each species as favourable, unfavourable-inadequate, or unfavourable-bad, built from data on population and distribution and their trends. Article 11 requires ongoing surveillance of conservation status. And under Article 6(3), any plan or project that could affect a site needs an appropriate assessment — which rests on knowing what is actually there. All three are data problems before they are anything else.

What camera traps and the platform provide

Many Annex species — wolf, Eurasian lynx, brown bear, wildcat, otter — are exactly what camera traps record without disturbing them. The platform structures that into:

• Presence and distribution — every detection carries the camera's GPS, so you can show which species occur where across a site or a multi-camera grid.
• Relative abundance (RAI) — detections normalized to independent events per 100 trap-nights, so cameras that ran for different lengths compare on equal footing and a trend can be tracked between reporting periods. It's valid for the same species over time, not as a ranking between species.
• Activity and phenology — when each species is active through the day and across the season.
• Survey effort — trap-nights per camera give a defensible measure of effort, and a species accumulation curve shows how completely a site has been surveyed.
• An auditable record — every detection is stamped with the model name and version that produced it, uncertain identifications pass through a human review queue, and corrections are logged.
• Export — the underlying data as CSV, or a report as PDF, into your own reporting and GIS. The data stays yours, hosted in Europe.

What it does not do

Camera traps are strong for mammals and some ground-dwelling birds; they are not a full biodiversity census, and they do not replace genetic sampling or dedicated bird and bat survey methods. RAI is a relative index, not a head-count — on its own it does not correct for how detectable a species is, so it is best read as a comparable trend rather than an absolute population size. For individually-marked species such as lynx, camera-trap capture–recapture can go further toward density estimates — a direction we are building toward, not a figure the platform reports today. And the platform produces the evidence layer; it does not write the Article 17 report or determine conservation status. That judgment stays with the ecologists and the competent authority.

Adjacent obligations

The same evidence base feeds the 2024 Nature Restoration Regulation, which asks Member States to monitor and show improvement over time. Wherever the requirement is “show, with data, what is on the ground and how it changes,” non-invasive camera monitoring with an auditable trail is a practical way to build it.

Analytics & ecology → Biodiversity & reserves → ← All knowledge topics