Summary
A gas transmission company operating a 340 km pipeline across the Rwenzori mountain range required a geohazard monitoring system to manage landslide risk at 23 identified high-risk crossings. Hypotenuse deployed a satellite-ground hybrid monitoring approach with AI-powered slope stability modelling, providing 48-hour predictive warnings that enabled pre-emptive pipeline isolation.
Background & Context
Annual rainfall exceeding 2,400 mm on the western slopes of the Rwenzori range creates persistent landslide hazard that threatens buried pipeline crossings. Historical records show 7 pipeline exposures and 2 ruptures due to slope movements in the preceding 15 years. Repair operations in remote terrain typically require 3–6 weeks and cost $2–8M each.
Sensor Deployment
Real-Time Risk Assessment
Hypotenuse's Slope Risk Engine combines traditional infinite slope stability analysis with a machine learning rainfall-response model tuned to each crossing's specific hydrogeological characteristics. The engine generates a Landslide Probability Index (LPI) on a 0–1 scale for each of the 23 crossings, updated hourly.
Four cumulative rainfall thresholds — calibrated against documented instability events — define the four alert levels. When an alert is triggered, the system automatically notifies pipeline controllers to initiate isolation valve procedures.
Key Outcomes & Results
Slope Failures Predicted
3 damaging slope movements accurately predicted — pipelines isolated pre-failure
Pipeline Isolations
8 pre-emptive isolations executed — 3 subsequently confirmed as necessary
Cost Avoided
Estimated $14M in repair and remediation costs avoided
Warning Lead Time
Average 52 hours before slope movement threshold breach
Environmental Impact
Zero hydrocarbon releases from geohazard events since deployment
23
High-risk crossings
52h
Landslide warning lead
$14M
Costs avoided
340km
Corridor monitored
Deployment Snapshot
Location
Rwenzori Mountains, Uganda
Annual Rainfall
> 2,400 mm
Risk Crossings
23 locations
Sensor Count
217 instruments