New Mexico's Environment Department paired satellite radar analytics with AI to find hidden water leaks in five pilot utilities, locating 82 leaks and cutting water loss by 240 gallons per minute before the state awarded ASTERRA a four-year contract to expand statewide.
82
Leaks identified (Phase 1, by May 2025)
240 gallons/minute
Water loss reduction (Phase 1)
31
Leaks found in Truth or Consequences (2 days)
Details
Maturity
Scaling
Promoter
New Mexico Environment Department (NMED), with ASTERRA
Period
2024-2026
Keywords
satellite radar analytics, AI leak detection, water infrastructure, public utilities
Context
New Mexico's Environment Department (NMED) worked with satellite-analytics company ASTERRA to find water leaks that traditional listening and inspection methods miss, particularly non-surfacing leaks in the state's geothermally complicated terrain.
Activities
Under the 'LeakTracer' programme, ASTERRA's L-band satellite radar and AI algorithms scanned water networks in five utilities — Truth or Consequences, Bernalillo, Cloudcroft, Timberon and Tranquillo Pines — distinguishing potable-water leaks from geothermal groundwater signals, with field crews from McKim & Creed verifying flagged sites on the ground.
Results
Phase 1, completed May 2025, identified 82 leaks and reduced water loss by an estimated 240 gallons per minute; in Truth or Consequences alone, satellite analytics pinpointed 31 leaks in just two days.
Conclusions
Based on these results the state awarded ASTERRA a four-year contract to extend the service statewide, aligned with New Mexico's 50-Year Water Action Plan; the figures come from state and vendor press materials rather than an independently audited evaluation, and no cost-per-leak or false-positive rate has been published.
Implementation
Indicative cost
Medium (€50k–€500k)
Time to results
Medium (1–3 years)
Staffing & skills
ASTERRA satellite-analytics team, McKim & Creed field verification crews
Conditions for success
Satellite radar capable of distinguishing potable leaks from geothermal signals
Ground-truthing field crews
State contract to fund statewide expansion
Common failure modes
No cost-per-leak or false-positive rate published; efficiency vs. conventional surveys not independently verified
WFP's AI-assisted HungerMap Live platform, built with Google.org and covering 50+ countries, helped identify worsening hunger in Somalia's Buur Hakaba …