Community science · Biodiversity

Help monitor moths near your place

Set up a simple night-time light trap, photograph what lands, and add habitat notes. Moths feed birds and signal whether a place is healthy.

Upload a trap photo →My contributions School inquiry pathway →

How it works

Set trap → Upload → Identify → Contribute

Set trap
White sheet + UV torch after dark, away from competing lights.
Upload photo
Moth on plain white — fill the frame, no jar needed.
AI identification
Order/family level with honest confidence when unsure.
Contribute
Habitat notes turn a pretty photo into place-based evidence.

Read the full setup guide →

Show, don't just tell

Example good photo

Place: School garden edge
Land use: Urban green space
Result: Lepidoptera — medium confidence

Moth fills the frame on a plain white sheet. Habitat notes make it science, not just a gallery.

Borrowed from MaramaTrap's example-photo pattern — works for every platform.

Honest AI identification

MaramaTrap uses the same AI-assisted approach tested for a UCOL light-trap research study in Whanganui. It identifies moths at order level (for example, Lepidoptera) — not species — and tells you when it is not sure rather than guessing. That is a feature, not a flaw: you stay the human check on what the model suggests.

The model is still learning NZ-specific patterns. Family-level accuracy will improve as more trap photos are contributed — another reason to participate.

Read about the science & data →

Learning & identification

Moths are ecological indicators. These resources support curious nights and careful science.

Host plants
Many NZ moths are tied to specific native plants — note what's growing nearby.
Seasonal activity
Warm still nights bring more visitors — log moon phase and weather.
Place notes
Forest edge, farm paddock, and urban garden host different communities.

17 moth families we are learning

The classifier is moving toward family-level IDs. When you correct a suggestion, pick from this list (spec 05).

Family-level only for now — not species claims (spec 03 §7). Corrections on your contributions help future training.

Good to know

Why place notes matter
Forest, farm, and garden host different moths. Your notes turn a photo into evidence.
AI honesty
The classifier says when it is not sure — that is a feature, not a failure.

At school?

Kōkiri Lab also runs a full invertebrate biodiversity inquiry world with VOICE investigation tools, datasets, and teacher guides. MaramaTrap community upload is open to everyone — no class code needed.

Also try community science

The same fair-testing habits work across platforms — honest notes, repeat visits, no street addresses.

All four platforms →