Challenges
Historically, Hetch Hetchy relied on periodic LiDAR surveys and manual field work for inspections. One forester was responsible for inspecting all 45 miles of distribution line and 324 miles of transmission line at least once a year, using LiDAR for an overhead view, navigating with a handheld device, and performing manual distance calculations with tools like range finders.
This approach created several issues:
- It was time-consuming and difficult to scale for a single forester.
- Manual distance calculations in the field increased the risk of errors.
- It was hard to document decisions in a way non-forestry stakeholders could easily understand.
- A LiDAR campaign had taken more than 18 months to deliver usable insights, by which time vegetation conditions had already changed.
Desired State
Hetch Hetchy wanted to move from a reactive, manual process to a proactive, data-driven vegetation management program that:
- Provides an up-to-date, system-wide view of vegetation risk
- Helps the team prioritize work by risk and plan resources ahead of time
- Supports regulatory compliance while minimizing unnecessary tree removal
- Makes it easy to demonstrate progress and inspection coverage to managers and other stakeholders
Solution: AI and Satellite-Powered Vegetation Management
To this end, Hetch Hetchy partnered with LiveEO to transform its vegetation management program. It chose LiveEO’s satellite-powered and AI-enabled vegetation management platform, Treeline, primarily for its scalability, cost benefits, and speed improvements compared with LiDAR. Treeline delivers the following:
- Identifies vegetation encroachment on conductors
- Uses predictive analytics to highlight vegetation likely to grow into the regulated clearance space in the next 1–2 years
- Flags fall-in trees that could strike lines if they failed
- Presents risk levels visually using intuitive color codes based on distance to the conductor
This enables Hetch Hetchy’s vegetation management team and tree crew to:
- Prioritize the highest-risk spans
- Plan crews and budgets more efficiently
- Stay ahead of growth and maintain regulatory compliance
- Focus the forester’s time on tree health and risk factors rather than manual distance calculations
"Treeline is much more visual, it's quick, and it has a different risk color depending on the distance away from the conductor. It’s visually more appealing and informational than LiDAR points with classifications.” — Joy Mellera
Outcomes
In the three years since deploying Treeline, Hetch Hetchy has seen significant improvements:
- 50% reduction in vegetation-related outages annually
- Significant reduction in vegetation maintenance effort (e.g. 260 hours of work saved every year for the forester)
- Reduced margin of error associated with manual calculations
Operationally, Treeline has significantly transformed the way Hetch Hetchy manages vegetation risk. It allows them to:
- Demonstrate full inspection coverage of the network
- Improve visibility into contractor work
- Visually track progress of vegetation work over years, span by span
- Help stakeholders without a forestry background understand risk levels and work completed
"Our management team can use Treeline to understand how the vegetation management program is working and how it’s improving year to year without having to dive into spreadsheets and numbers. They can visually see that there’s less risk on the system and we’re doing what we should do.” — Joy Mellera
Using Treeline Beyond Vegetation Management
Hetch Hetchy has also found additional use cases for Treeline data:
Targeted de-energization during wildfires:
Instead of conservatively shutting off power over a broad area, Hetch Hetchy can pinpoint segments with higher vegetation risk and take more precise actions.
Strategic undergrounding:
Treeline helps identify areas with consistently high vegetation pressure, where moving lines underground can reduce future maintenance. This supports Hetch Hetchy’s preference to prune rather than remove trees whenever possible, while still managing long-term risk and cost.
Next Steps: Expanding AI-Driven Situational Awareness
Looking ahead, Hetch Hetchy is exploring ways to deepen its use of AI and satellite data:
- Increasing the frequency of satellite analyses to twice per year
- Extending risk assessments to access roads and critical facility routes
- Incorporating tree species behavior into risk models (e.g. black oaks with tendencies for sudden limb failure and fast-growing eucalyptus trees)
- Targeting maintenance even more precisely by combining species-specific risk with clearance and fall-in analyses





