Updated: 2 days ago
EnviroEyes: From Observation to Action
EnviroEyes is Zealandia Consulting's camera monitoring platform for environmental sites, and it is built and operational. It turns a stream of site photos into timely, plain-language alerts, so teams see what changed and can act before a short-lived problem becomes a lasting one.
Watch: EnviroEyes, how it works
The video walks through the problem EnviroEyes solves, how the platform works, and what we learned on a live state highway project. This article covers the same ground for readers who prefer to skim.
The challenge: the moments that matter are short
On construction and earthworks sites, the events that harm waterways often last minutes, not days. A sediment plume can enter a pond and pass through between site visits. A weir can overtop during heavy rain overnight, when no one is there to see it.
Sediment is one of the most widespread contaminants in Auckland's freshwater, estuaries and coastal waters. Councils and contractors need evidence of what happens on site between inspections, not just a snapshot on the day someone visits.
Key benefits
EnviroEyes brings every camera, image and alert into one secure platform. Here is what that means in practice.
Feature | What it does | Why it matters |
Works with any camera | Accepts purpose-built solar cameras, Android phones, trail cams, partner feeds and public or IP camera streams | Use the hardware that suits the site and budget, without being locked to one supplier |
Up in minutes | Add a phone by scanning a code; cameras can be fixed or vehicle-mounted and shown on a live map | Fast deployment for floods, emergencies and short-term works |
One live dashboard | Shows every site on one screen, with alerts alongside and saved layouts | A common picture for the office wall or the field team |
Time Machine | Scrubs through days of images, plays any period as a time-lapse, and exports clips or frames | Find exactly when something happened and share the evidence |
Built-in AI checks | Detects objects on site, watches set zones for change, and describes each scene in plain words | Hours of footage reviewed automatically, with people focused on what changed |
Alerts that explain themselves | Sends the photo with a plain-language summary to the right people, tracked through to close | Faster, clearer responses with a record of what was done |
Images meet sensor data | Shows sensor readings beside each camera; click a spike to see the matching image | Visual context for every reading, and a growing record for future learning |
Secure by design | Role-based access, Microsoft single sign-on, an audit log of changes and a separate deployment for each client | Client data stays private and every action is traceable |
Alerts you can trust
Every alert is checked before it is sent, so the people who receive it can act with confidence rather than chase false alarms.
1. Classify the frame against a 16-label library of pond and site conditions.
2. Allow for sun glare and low water, which can fool a camera.
3. Confirm the condition persists across more than one frame.
4. Check the context, such as rainfall and known site activity.
5. A person reviews the frame before anything is sent.
Confirmed alerts are then tiered so the response matches the risk:
Tier | Typical response |
Urgent | Inspect the same day |
Action | Review within one working day |
Watch | Included in a weekly summary and tracked |
Alerts describe what the camera saw and what is recommended. They never state non-compliance; that finding is for an authorised person to make.


Comments