Most mammals of our forests are almost never seen. Lynx, wildcat, marten and wolf are active at night and avoid people. A deer or a wild boar notices a surveyor long before the surveyor notices it. A camera trap solves this simply: it sits in the forest for weeks, does not smell of people, does not get tired and does not sleep.

What a camera trap gives

• Evidence, not impressions. Every detection is a photo with a date, time and coordinates. Any specialist can check it, today or twenty years from now.
• Nocturnal and elusive animals. A camera with an infrared flash sees what happens when nobody is in the forest.
• Comparability. If cameras follow the same rules, data from different parks and years can be put side by side. This is how trends appear: whether encounters with a species become more or less frequent.
• Daily activity and seasonality. When a species uses the trail, and how this changes with the season and under human pressure.
• Human pressure. The same cameras record people, vehicles and dogs, showing how "protected" an area really is.

What it has already shown

In the Roztochya Nature Reserve, large mammals have been counted for more than 40 years, and since 2020 a camera-trap network complements these counts. The cameras confirm the continued increase of roe deer, the recovery of the wild boar population after its decline, and the regular presence of the wildcat.

As of October 2026, biomon.app holds more than 818,000 photos from 912 locations of 23 institutions, from the Carpathians to Polissia and the steppe. 57 animal species have been identified.

Why it matters right now

Nature reserves and national parks are short of staff. Because of the war, many experienced employees and young specialists have left. Classical counts depend on a person walking through the forest, and without people they stop. A camera trap keeps working while nobody is there. Data accumulate and can be checked later.

There is a wider reason too. Ukraine is moving towards European reporting on the state of species and habitats. That reporting needs numbers collected by a transparent method, not descriptions such as "the species is common". A camera-trap network with an open database provides exactly such numbers.

Where AI fits and where people do

One camera can take tens of thousands of frames in a season, most of them empty: grass in the wind, sunlight, snow. Nobody can look through all of them by hand. So on biomon.app AI works first: it groups frames into series and suggests a species. But people have the final say. A series counts as verified when two independent people agree on the species. This combines the speed of the machine with the reliability of a specialist.

What a camera trap cannot replace

• It sees only a strip of a few metres in front of it. Whatever does not pass in front of the camera does not exist for it.
• No detection does not mean no species. For "there is no lynx here" to mean anything, hundreds of camera-days are needed.
• Small mammals, birds in the canopy, bats and amphibians need other methods, such as acoustic monitoring.

How to take part

• Deploy cameras by the rules: Camera-trap field methods.
• Help check photos: register on biomon.app and start with Verifying series. Fifteen minutes a day already makes a difference.
• All guides in one place: Monitoring standards.