By the PICA Corp Engineering Team  |  Updated July 2026  |  Est. reading time: 9 min

Two numbers define the aging pipeline problem in North America, and they point in opposite directions.

The first: 452,000 miles of water main in the United States and Canada have outlived their useful life and gone unreplaced for lack of funds. That is 20 percent of the network, up from 8 percent in 2012, and the Utah Water Research Laboratory prices the shortfall at $452 billion.

The second: overall water main failure rates fell about 20 percent between 2018 and 2023. Both figures come from the same study.

An older network is breaking less often. Why that happened, and why it will not keep happening on its own, is the most useful thing an asset manager can understand about aging infrastructure right now.

---

How old is the buried network, actually?

ASCE counts roughly 2.2 million miles of drinking water pipe in the United States. Add Canada and the working figure is about 2.3 million miles. Of that, 33 percent is more than 50 years old, which is around 770,000 miles of pipe installed before the first moon landing.

The age is not spread evenly across materials. Eighty-six percent of cast iron pipe has passed 50 years, along with 41 percent of asbestos cement pipe. Those two materials also carry the highest break rates of anything in common use, so the oldest inventory and the most failure-prone inventory are largely the same pipe.

The reported average failure age is 53 years, a number quoted often and misread almost as often. It marks the centre of a very wide distribution, not an expiry date. Cast iron from the 1890s is still in service in several eastern cities while ductile iron installed in the 1980s has already been dug up in corrosive ground.

---

The funding gap stopped being cyclical

For most of the past two decades the water sector described its capital problem as a backlog: a large but finite pile of deferred replacement that sufficient investment would eventually clear. AWWA’s 2026 analysis argues that framing no longer holds.

What AWWA projects through 2050

Total United States drinking water infrastructure needs are now projected at $2.1 to $2.4 trillion between 2026 and 2050 in 2025 dollars. Utilities currently invest about $33.6 billion a year in capital improvements. Meeting the projection would require roughly $90.2 billion a year, an annual gap near $56.6 billion and a 168 percent increase over current spending.

The number climbed past earlier estimates because replacement is no longer the only cost driver. PFAS treatment, lead service line replacement, climate resilience and cybersecurity all compete for the same capital. Federal support covers about 3.9 percent of public spending on water utilities, the lowest share of any major infrastructure class, and the IIJA infusion that softened the last five years expires after FY2026.

What the EPA needs survey shows about buried pipe

EPA’s most recent Drinking Water Infrastructure Needs Survey and Assessment put 20-year national needs at $625 billion. Distribution and transmission accounted for $421 billion of that, more than two thirds of the total, and rose 36 percent from the previous survey’s $310 billion. Buried pipe is where the money has to go, and buried pipe is the part of the system nobody can see.

ASCE’s 2025 Infrastructure Report Card graded drinking water C-, wastewater D+ and stormwater D, all unchanged from 2021. Four years of elevated federal spending moved none of the three.

---

What the failure data shows, and where it misleads

Break rates fell while the network aged

The 20 percent decline in failure rates between 2018 and 2023 did not happen because pipes got younger. Researchers link it to the shrinking share of cast iron and asbestos cement, which together fell from 41 percent of installed mains to 33 percent over the same window as utilities replaced them with PVC and ductile iron.

That is a specific and repeatable lesson. The improvement came from removing the materials with the worst performance, not from replacing pipe in order of installation date. Material and environment drove the outcome. Age was a proxy that happened to correlate.

Soil does more damage than the calendar

Seventy-five percent of utilities report corrosive soil conditions, consistent across three survey cycles going back to 2012. In highly corrosive ground, ductile iron fails more than six times as often as the same pipe in low-corrosivity soil. Two mains of the same material, same vintage, same diameter, a few blocks apart, can sit on opposite ends of the risk distribution because of what surrounds them.

Diameter shifts the picture again. Pipes 12 inches and under fail roughly five times more often than mains of 14 inches and above, and 86 percent of the network sits under 12 inches. Frequency and consequence are different things, though. A 6-inch break floods a street; a 60-inch transmission break takes out a district, and the failure mechanisms in large-diameter concrete pressure pipe give far less warning.

Age flags a third of the network, which is not a priority list

Here is the practical trap. If age is the screening criterion, 770,000 miles qualify as at-risk. No utility can inspect or replace a third of its system, and the ranking within that third is exactly what age data cannot supply.

Install date, material, soil map and break history narrow a network to candidate segments cheaply, and most utilities already hold all four. What they cannot do is tell you which of two 1968 cast iron mains has 40 percent wall loss and which has 5 percent. That requires measurement.

---

Most utilities are still working without condition data

About 44 percent of utilities conduct some form of regular condition assessment of water mains. ASCE found roughly one third operating a robust asset management plan, up from 20 percent in the prior report card, so the trend is moving the right way from a low base.

The majority, then, allocate capital using install date and break history. That produces two predictable errors, and utilities usually make both at once. Sound pipe gets replaced because it is old, spending capital that bought nothing. Deteriorated pipe stays in the ground because it has never broken, until it does.

The inspect before you replace argument is not that replacement is wrong. It is that with a $56.6 billion annual gap, no volume of spending closes the problem, so the only remaining lever is aim.

---

Why one inspection method leaves gaps

Aging pipe does not deteriorate through a single mechanism, which is the flaw in any program built around one technology.

Cast iron fails through graphitic corrosion, where the iron leaches out and leaves a graphite skeleton holding the original shape. A camera sees an intact pipe wall. The pipe can be soft enough to penetrate by hand.

Prestressed concrete cylinder pipe has at least two routes to failure. The familiar one runs through prestressing wire breaks reducing pre-load until the concrete core cracks. The second starts at the liner: if the lining cracks, fluid reaches the steel cylinder and corrodes it while the wires stay intact. A wire-break inspection returns a clean result, because nothing is wrong with the wires.

This is why PICA sequences methods rather than selling a tool. Remote Field Testing measures cylinder wall thickness continuously along the run, plus broken wires or bars and loss of pre-load on pipe segments, and reads through internal liners up to 25-30 mm thick. Near Field Testing quantifies 5 or more adjacent broken prestressing wires or bars in concrete pressure pipe but does not measure cylinder wall thickness. CCTV, laser and lidar document visible surface condition, liner loss and ovality but see nothing inside the wall. Each blind spot is a different shape, and only the combination covers the pipe.

---

What condition assessment costs against replacement

A combined NDT program on a large-diameter PCCP transmission main runs $100,000 to $200,000 or more per mile, the range reflecting diameter, linear footage, access complexity, number of tiers and mobilisation distance. Small and medium metallic distribution mains from 4 to 24 inches are a different order: several thousand dollars up to roughly $40,000 per mile.

Set that against the failure it prevents. Inspection typically costs 10 to 50 times less than an unplanned failure on the same run, counting only direct response, and not the water lost, the road rebuilt or the businesses closed.

At network level the arithmetic is simpler. Every dollar spent replacing a main that had 30 years of service left is a dollar unavailable for the main that had two.

---

How aging pipelines get inspected

PICA runs five NDT service tiers and selects among them by pipe material, diameter, access and suspected failure mode. A full pipeline condition assessment works through them in order of cost and disruption.

Screening comes first. The Navigator multi-sensor acoustic sphere travels a live pressurised line from 6 to 78 inches at up to 300 psi, locating leaks, gas or air pockets and pressure anomolies, with data analysed within 72 hours for leak identification. It is most useful on smaller-diameter pipe, where acoustic signatures are distinguishable.

Structural assessment comes from electromagnetic inspection. In-service RFT tools cover 2 to 36 inches free-swimming, with HydraSnake deploying through a fire hydrant on 6-inch and 8-inch cast and ductile iron mains with limited service interruption. Above 36 inches the line comes out of service: RAFT covers 36 to 48 inches through manway-sized access, EMIT covers 48 to 96 inches assembled inside the pipe, and NFT extends reach to 136 inches for PCCP and CCP pipe inspection. Handheld Bracelet Probe and ultrasonic measurements at excavated locations then confirm what the in-line run reported.

The output is a condition score by pipe segment, the format a capital plan can actually use. It converts 770,000 miles of merely old pipe into a ranked list of measurably compromised pipe. For what those tools are looking for, see the major threats to pipeline integrity and the water main inspection service overview.

---

Frequently asked questions

How old is North America’s water pipe network?

Roughly 33 percent of water mains in the United States and Canada are more than 50 years old, which works out to about 770,000 miles of the estimated 2.3 million miles in service. The concentration is worse in the oldest materials: 86 percent of cast iron pipe and 41 percent of asbestos cement pipe have passed the 50-year mark. Those figures come from the Utah Water Research Laboratory’s 2023 study of more than 800 utilities covering nearly 400,000 miles of main.

How long does a water main actually last?

Utilities report an average failure age of 53 years across all pipe materials, but the spread around that average is enormous. Two mains installed the same year in the same city can differ by decades in remaining life depending on soil corrosivity, pressure history, water chemistry and manufacturing batch. Design life is a planning assumption, not a measurement of the pipe in the ground.

Does replacing old pipe reduce water main breaks?

It does when the right pipe is replaced. Overall failure rates in the United States and Canada fell about 20 percent between 2018 and 2023, and the researchers attribute the drop to shrinking inventories of cast iron and asbestos cement, the two materials with the highest break rates. Combined, those two dropped from 41 percent of installed mains to 33 percent over the same period. The improvement came from targeting the worst-performing materials rather than from replacing pipe by age.

How much will it cost to fix aging water infrastructure?

AWWA projects total United States drinking water infrastructure needs of $2.1 to $2.4 trillion between 2026 and 2050 in 2025 dollars. Utilities currently invest about $33.6 billion a year in capital improvements against an estimated $90.2 billion needed, leaving an annual gap of roughly $56.6 billion. Federal funding covers only about 3.9 percent of total public spending on water utilities, so the shortfall lands on local ratepayers.

What percentage of utilities inspect their water mains?

About 44 percent of utilities conduct some form of regular condition assessment on their water mains, according to the 2023 Utah Water Research Laboratory survey. ASCE’s 2025 Infrastructure Report Card found roughly one third of utilities operating a robust asset management plan, up from 20 percent in the previous cycle. Both figures mean the majority of the network is being managed on install date and break history rather than measured wall condition.

Which pipe materials fail most in aging systems?

Cast iron and asbestos cement carry the highest break rates, which is why utilities have prioritised removing them. Soil drives the ranking as much as material does: 75 percent of utilities report corrosive soil, and ductile iron in highly corrosive ground fails more than six times as often as in benign ground. PVC records the lowest break rate of the common materials.

How do you decide which aging pipes to replace first?

Narrow with the data you already hold, then measure the shortlist. Install date, material, soil corrosivity and break history identify candidate segments cheaply but cannot rank them, because two mains with identical records can have very different remaining wall. Electromagnetic inspection supplies the missing variable, measuring actual wall thickness and defect location along the run and returning condition scores by pipe segment.

How is an aging pipeline inspected?

Through a sequence of methods matched to material, diameter and access. Acoustic pre-screening runs a live pressurised line to locate leaks and gas or air pockets. CCTV, laser and lidar document visible internal condition once the line is dewatered. Remote Field Testing measures wall thickness continuously and detects broken wires or bars and loss of pre-load, while Near Field Testing quantifies broken prestressing wires or bars in concrete pressure pipe up to 136 inches. Handheld probes then validate findings at excavated locations.

---