Six Billion Gallons Lost Through Pipes We Cannot See

The U.S. loses an estimated 6 billion gallons of treated drinking water daily through leaking distribution pipes, many of which are cast iron or ductile iron installed 50–100+ years ago. The critical maintenance decision — repair, reline, or replace — depends on knowing the remaining wall thickness, but no nondestructive evaluation method can reliably measure this parameter from the pipe exterior or interior without excavation. Utilities therefore replace pipes on age-based schedules rather than condition-based assessments, leading to both premature replacement of sound pipes and catastrophic failure of deteriorated ones still within their age threshold.

The U.S. has 2.2 million miles of water distribution pipe. AWWA estimates $1 trillion in needed investment over 25 years. Water main breaks cause 240,000+ incidents annually, disrupting service, flooding roadways, and contaminating water supplies through pressure loss. Each break costs $5,000–$500,000+ to repair depending on pipe size and location. Age-based replacement wastes resources on functional pipes while missing the most deteriorated segments.

Acoustic leak detection identifies active leaks but not pre-failure wall thinning. In-line inspection tools (smart pigs) adapted from oil/gas pipelines work for large-diameter transmission mains but cannot navigate the bends, diameter changes, and service connections of distribution networks (4"–12" pipe). Electromagnetic methods (broadband electromagnetic, remote field eddy current) show promise in clean lab conditions but are confounded in the field by external soil conditions, internal tuberculation (iron deposits), cement mortar linings, and polyethylene wraps. Pit-depth measurement by direct inspection requires excavation, defeating the nondestructive objective. Statistical failure models based on break history are retrospective and cannot identify specific vulnerable segments before they fail.

A through-soil or through-pipe sensing modality that can estimate remaining wall thickness to ±15% accuracy in situ, without excavation, in metallic pipes with internal deposits and external coatings. Candidate approaches include guided-wave ultrasonics adapted for corroded geometries, electrical resistance tomography from ground surface, or free-swimming miniature inspection robots that can navigate distribution-scale pipe networks. The adjacent success in natural gas pipeline inspection (where smart pigs are standard) suggests the core sensing physics is known — the barrier is adapting it to the geometrically complex, small-diameter, heavily fouled water distribution environment.

A team could prototype a low-cost guided-wave ultrasonic sensor and test it on sections of exhumed cast-iron pipe with known corrosion profiles. A comparison of electromagnetic vs. ultrasonic vs. acoustic methods on a test bed of pipe samples with artificially induced wall loss at varying depths would quantify the sensitivity limits of each approach. Relevant disciplines: nondestructive evaluation, signal processing, materials science, urban infrastructure engineering.