A 2025 study of thick-walled pipe girth welds found that phased array ultrasonic testing reached a 90% probability of detection at a flaw size of 6.87 mm, while film radiography needed 12.72 mm to reach the same rate (Journal of Materials Research and Technology 2025). That gap matters on a US network of 3,300,566 miles of pipeline, which logged 235 significant incidents in 2024 (Congressional Research Service, R44201 2025). The same study found phased array generating false positives on 13.59% of calls, so better detection comes with more digging and re-checking. The cost of a miss is concrete: the girth weld behind the 2022 Keystone rupture in Kansas had passed radiography in 2010, and PHMSA put the total cost of that failure at $600,793,659 (PHMSA Failure Investigation Report 2024). We aggregated data from PHMSA, the US Government Accountability Office, the Congressional Research Service, the European Gas pipeline Incident data Group (EGIG), Concawe, UKOPA, and peer-reviewed NDT studies; for the occupational side of the same high-frequency sound, see our industrial ultrasonic noise exposure statistics.
A few plain-language definitions used throughout. Ultrasonic testing (UT) sends sound far above human hearing into steel and listens for echoes from flaws. Phased array UT (PAUT) uses a probe with many small elements that steer and focus the beam electronically, building an image instead of a single trace. TOFD (time-of-flight diffraction) times the faint signals that bounce off the tips of a crack to estimate its height. A girth weld is the circular weld that joins two lengths of pipe end to end in the field. POD (probability of detection) is the chance an inspection finds a flaw of a given size, and in-line inspection (ILI) means a sensor-carrying device, nicknamed a “smart pig”, that travels inside the pipe with the product flow.
TL;DR
- Phased array UT reached 90% POD at 6.87 mm in thick-walled girth welds, against 12.72 mm for radiography (Journal of Materials Research and Technology 2025).
- Phased array false positives ran at 13.59%, against 0.58% for radiography in the same welds (Journal of Materials Research and Technology 2025).
- Six automated ultrasonic testing (AUT) systems found flaw heights of 2 to 4 mm at 90% POD with 95% confidence (Edison Welding Institute for PHMSA 2011).
- The US recorded 235 significant pipeline incidents in 2024, below the 2014 to 2024 average of 265 a year (CRS R44201 2025).
- Corrosion caused more than 25% of reported US pipeline incidents in 2024, up from a historical share of roughly 18% (AMPP US Pipeline Industry 2026).
- Equipment or material failure caused 43% of significant onshore gas transmission incidents from 2010 to 2022 (GAO-24-106690, 2024).
- Girth weld defects caused 39 of 207 UK onshore product-loss incidents since 1962, or 18.8% (UKOPA Product Loss Report 2024).
- European gas transmission failure frequency fell to 0.101 incidents per 1,000 km-years in 2018 to 2022, from 0.277 over 1970 to 2022 (12th EGIG Report 2024).
- In-line inspection tools covered 26% of Europe’s cross-country oil pipeline length in 2023 (Concawe Report 8/25, 2025).
- PHMSA’s pipeline safety research program counts 35 commercialized technologies since 2002 (PHMSA FY 2021-2022 R&D Report to Congress, 2024).
1. US Pipeline Network Size and Incident Counts
The headline count has drifted down, but the denominator keeps growing. The US recorded 235 significant pipeline incidents in 2024, against a 2014 to 2024 average of 265 a year (CRS R44201 2025), while federally regulated mileage rose about 8% over roughly the same period. Fewer incidents on more pipe is real progress, yet the safety advocates’ count below shows why nobody in the integrity business treats the trend as settled.
“Significant” is a PHMSA category, not a casual adjective: it covers incidents with a death or in-patient injury, large releases, or total costs of $50,000 or more in 1984 dollars, among other triggers. “Serious” is the narrower subset with a fatality or in-patient hospitalization. Mixing the two definitions is an easy way to misread pipeline safety headlines. The full background sits in the Congressional Research Service report on the federal pipeline safety program.
| Metric | Value | Source |
|---|---|---|
| Total US pipeline mileage, all regulated systems (2023 data) | 3,300,566 miles, of which 298,851 gas transmission, 225,637 hazardous liquid transmission and 2,350,258 gas distribution mains and services | CRS R44201, 2025 (PHMSA data) |
| Significant incidents in 2024 vs yearly average 2014-2024; serious incidents (death or in-patient injury) in 2024 | 235 vs 265 per year on average; 25 serious incidents in 2024 (2014-2024 average: 26 incidents, 11 deaths and 48 injuries per year) | CRS R44201, 2025 |
| Reportable incidents between the January 18, 2024 and July 22, 2025 House hearings | 881, of which over 400 significant (one significant incident every 1.32 days on average) | Pipeline Safety Trust testimony to House Energy and Commerce, 2025 |
| People killed and injured to the point of hospital admission in the same window | 18 killed, 45 injured | Pipeline Safety Trust testimony, 2025 |
| Cost of pipeline incidents reported by operators over the last 10 years | Over $6 billion | Pipeline Safety Trust testimony, 2025 |
| Incidents of all kinds reported to PHMSA per year | Between 600 and 700 | AMPP US Pipeline Industry 2026 report |
| Significant onshore gas transmission incidents, yearly average 2010-2022 | 61 per year | GAO-24-106690, 2024 |
Context note: the Pipeline Safety Trust is an advocacy group, and its 881-incident figure is a count of PHMSA reports over an 18-month window between two hearings, not a calendar-year total. Its July 2025 testimony also calls 2023 and 2024 the deadliest two-year period for pipelines in over a decade.
2. What Breaks Pipelines: Corrosion, Material and Weld Failure
These shares answer the question that matters for inspection planning: which threats can a sound wave actually find? Equipment or material failure caused 43% of significant onshore gas transmission incidents from 2010 to 2022, against 16% for corrosion (GAO-24-106690, 2024). Corrosion, which ultrasonic wall-thickness tools measure directly, is now climbing: AMPP reports it rose to more than 25% of US incidents in 2024. Weld and material flaws, the target of phased array and crack-detection tools, stay a stubborn slice on every continent.
The UK numbers are the clearest signal for girth weld inspection. Girth weld defects caused nearly as many UK product losses as external corrosion over six decades, which is why construction-stage weld inspection carries so much weight in the pipeline codes.
| Metric | Value | Source |
|---|---|---|
| Corrosion share of reported US pipeline incidents, 2024 vs historical share | More than 25% in 2024 vs roughly 18% historically | AMPP US Pipeline Industry 2026 report (PHMSA data) |
| Share of significant onshore gas transmission incidents, 2010-2022: equipment or material failure vs corrosion failure | 43% vs 16% | GAO-24-106690, 2024 |
| Leading causes of significant gas transmission incidents inside high consequence areas (zones where a failure would hurt people or property most) | Equipment or material failure about 37%; excavation damage about 19% | GAO-24-106690, 2024 |
| Girth weld defect share of UK onshore product-loss incidents, 1962-2022 | 39 of 207 incidents (18.8%) | UKOPA Product Loss Report 1962-2022, 2024 |
| External interference and external corrosion shares of the same UK incidents | 44 incidents (21.3%) and 42 incidents (20.3%) | UKOPA Product Loss Report 1962-2022, 2024 |
| Mechanical failure share of European oil pipeline spillages since 1971 (all spills / excluding theft) | 142 spills: 18% / 28%; of which 54 construction faults and 88 design or materials defects | Concawe Report 8/25, 2025 |
| Cause shares of European gas transmission incidents, 2013-2022 | Corrosion 25.7%, external interference 22.8%, ground movement 19.3%, construction defects 17.5% | 12th EGIG Report, 2024 |
Context note: the AMPP figures come from the association’s March 2026 report “US Pipeline Industry 2026: Leading Failures, New Technologies, and Future Outlook”, as summarized by World Pipelines; the full report sits behind AMPP’s site verification. GAO’s cause categories group equipment and material failure together, so its 43% is not a weld-only number.
3. European Benchmarks: Failure Frequency per 1,000 km-years
Europe publishes the cleanest long-run failure rates, measured in km-years: one kilometre of pipeline in service for one year. European gas transmission failure frequency dropped to 0.101 incidents per 1,000 km-years in 2018 to 2022, about 36% of the 0.277 average since 1970 (12th EGIG Report, 2024). EGIG credits in-line inspection among the technologies behind the decline, and the construction-defect and material failure rate fell by a similar proportion, from 0.046 to 0.016.
Corrosion is the exception. Its five-year frequency has stopped falling and, in EGIG’s words, has levelled off between 0.030 and 0.040 per 1,000 km-years, which mirrors AMPP’s US warning in section 2.
| Metric | Value | Source |
|---|---|---|
| European gas transmission incidents and exposure, 1970-2022 | 1,463 incidents over 5.29 million km-years | 12th EGIG Report, 2024 |
| EGIG failure frequency: whole period vs last 10 years vs last 5 years | 0.277 (1970-2022) vs 0.118 (2013-2022) vs 0.101 (2018-2022) per 1,000 km-years | 12th EGIG Report, 2024 |
| EGIG failure frequency by cause, 1970-2022 vs 2018-2022: construction defect and material failure; corrosion | 0.046 vs 0.016; 0.049 vs 0.035 per 1,000 km-years | 12th EGIG Report, 2024 |
| UK onshore pipeline failure frequency: 1962-2022 vs last 20 years vs last 5 years | 0.194 vs 0.074 vs 0.042 incidents per 1,000 km-years (network of 23,577 km at end of 2022) | UKOPA Product Loss Report 1962-2022, 2024 |
| European oil pipeline spill rate: long-term running average vs mid-1970s; spillages in 2023 | 0.41 vs about 1.1 spillages per 1,000 km per year; 8 spills in 2023, 6 of them theft and the 2 others corrosion (0.06 per 1,000 km excluding theft) | Concawe Report 8/25, 2025 |
Context note: EGIG covers gas transmission lines run by its member operators (about 150,000 km of pipeline collected each year), UKOPA covers UK onshore transmission pipelines, and Concawe covers cross-country oil pipelines. The three are not directly comparable with each other or with PHMSA’s “significant” category in section 1. Most recent available data: EGIG and UKOPA through 2022, Concawe through 2023.
4. In-Line Inspection Coverage and the Limits of the Tools
In-line inspection is how buried pipe gets examined at scale. Ultrasonic ILI tools either measure remaining wall thickness or fire angled sound pulses to find cracks; magnetic flux leakage (MFL) tools sense metal loss through disturbances in a magnetic field. Across Europe’s cross-country oil pipelines, ILI tools covered a unique length of 8,260 km in 2023, or 26% of the network, against about 12% a year in the mid-1990s (Concawe Report 8/25, 2025). Crack tools remain a minority of runs, a fraction of the mileage covered by metal loss tools.
Coverage is not the same as detection. A tool is built for one flaw orientation and size range, and the Keystone case shows the gap: two different ILI technologies passed over the weld that later ruptured. Depth sizing is a known weak spot for ultrasonic crack tools, because the echo amplitude used to estimate depth saturates above a probe-dependent crack depth, as Willems and co-authors explained at the 2017 Pipeline Technology Conference (most recent available data: 2017).
| Metric | Value | Source |
|---|---|---|
| ILI length run on European cross-country oil pipelines in 2023, all tool types | 14,204 km in 113 sections, including 6,363 km with metal loss tools and 3,231 km with crack-detection tools (31 sections) | Concawe Report 8/25, 2025 |
| Share of network length inspected at least once in 2023 vs 10-year average | 26% (8,260 km unique length) vs 22% | Concawe Report 8/25, 2025 |
| European oil pipeline sections inspected at least once over the last 10 years | 442 of 615 active sections (72%), covering 86% of surveyed length | Concawe Report 8/25, 2025 |
| Keystone girth weld, November 2018 MFL ILI run with girth-weld capability | Did not detect the crack, which likely existed but was below tool capability | PHMSA Failure Investigation Report, TC Oil Washington KS, 2024 |
| Keystone girth weld, September 2020 ultrasonic axial crack ILI run | Did not detect it; tool not designed for cracks in girth welds | PHMSA Failure Investigation Report, 2024 |
| Ultrasonic crack tool run in two consecutive years under API 1163 (the industry standard for qualifying ILI systems) Level 2 field validation | POD consistently reliable, depth accuracy questionable (features underestimated); tool did not meet its published specification | Guajardo and Urrea (NDT Global), IPC2026 paper abstract |
| Circumferential crack tool, 814 km of 12-inch pipe (vendor’s own claim) | POD above 95%; 99.5% of identified flaws within plus or minus 1 mm depth tolerance at 90% confidence | NDT Global case study (vendor claim) |
Context note: the last two rows come from an ILI vendor’s own staff and marketing, and are shown to illustrate how tool performance is specified, not as industry statistics. The IPC2026 abstract is notable precisely because a vendor reports its own tool missing specification. For a parallel look at sound-based sensing in another medium, see our underwater acoustics and ocean sonar statistics.
5. Girth Weld Inspection: Phased Array UT vs Radiography
The weld-level studies explain why pipeline builders are moving from film to sound, and why the move is not finished. In thick-walled 9% Ni steel pipe girth welds, phased array UT reached 90% POD at 6.87 mm against 12.72 mm for radiography, but its false-positive rate was 13.59% against 0.58% (Journal of Materials Research and Technology 2025). Phased array was better on planar flaws such as lack of fusion (where weld metal fails to bond with the pipe wall), while radiography did better on volumetric flaws such as incomplete penetration, and the authors conclude that replacing radiography in critical thick-section joints remains a challenge.
Two notation points for the table. “a90/95” is the flaw size found 90% of the time with 95% statistical confidence; “a80” is the size found 80% of the time. RMSE (root mean square error) is the typical sizing error, in millimetres, between what the inspector reported and the true flaw size. The Keystone row shows the field consequence: radiography passed a weld whose lack-of-fusion defects lay parallel to the pipe wall, and phased array found them only after the rupture (PHMSA Failure Investigation Report 2024).
| Metric | Value | Source |
|---|---|---|
| 9% Ni steel pipe girth welds, phased array UT vs radiography: flaw size at 90% POD; false-positive rate | 6.87 mm vs 12.72 mm; 13.59% vs 0.58% | Payao Filho et al., Journal of Materials Research and Technology 2025 |
| Flaw height detected at a90/95 by six automated ultrasonic testing (AUT) systems | 2 to 4 mm | Edison Welding Institute for PHMSA, 2011 |
| Phased array round robin (a comparison in which several teams inspect the same samples) on 30 tube weld specimens with 109 artificial flaws (6 teams): flaw size at 80% POD, all flaws vs planar flaws | 6.93 mm vs 5.15 mm | Applied Sciences, PAUT in lieu of RT for tube welds, 2022 |
| Same round robin: flaw size at 80% POD in carbon steel vs dissimilar metal welds; overall sizing RMSE | 4.53 mm vs 11.58 mm; RMSE about 5.8 mm against a 19 mm ASME Section XI criterion | Applied Sciences, 2022 |
| Phased array round robin on 12 thermal fatigue cracks in 5 carbon steel pipe welds (3 teams): a80 for length and depth; sizing RMSE | a80 4.21 mm (length) and 1.41 mm (depth); RMSE 5.65 mm (length) and 1.69 mm (depth) | Materials, round-robin PAUT reliability study, 2023 |
| Embedded defects in three welded plates found by phased array UT vs film radiography; minimum spacing for phased array to separate indications | 19 of 19 vs 19 of 19; about 15 mm | Herceg et al., Materials 2023 |
| Keystone failed girth weld: radiography in November 2010 vs phased array UT after the December 2022 rupture | Radiography passed the weld; phased array identified the lack-of-fusion defects | PHMSA Failure Investigation Report, 2024 |
Context note: the 2022 tube weld round robin (most recent available data: 2022) found phased array clearly better than radiography on planar flaws, while radiography found every volumetric flaw and phased array missed one in a dissimilar metal weld. The 2023 thermal fatigue round robin reports a80/90 for length at about 6.6 mm, just above its 6.4 mm acceptance length, so team-to-team variation still decides whether a procedure qualifies. The PHMSA-funded AUT evaluation dates from 2011 (most recent available data: 2011), and the Herceg plate study (most recent available data: 2023) used artificial defects in plates, not pipe. A US Nuclear Regulatory Commission study by PNNL (NUREG/CR-7204, 2015) reached the same qualitative split for carbon steel piping: phased array finds all but very small volumetric flaws and is much more capable on planar flaws than radiography. For the welding trade’s own exposure risks, see our shipyard welding and sandblasting noise statistics.
6. Regulation, Oversight and Research Funding
The rules set the inspection floor, and recent US rules raised it sharply. The 2019 gas transmission rule added about 24,000 miles of required integrity assessment, a 115% increase to nearly 46,000 miles (GAO-24-106690, 2024), mostly by extending assessments beyond high consequence areas into newly defined moderate consequence areas. On the liquids side, the 2019 hazardous liquid rule pushes operators toward pipe that can take ILI tools at all.
Oversight capacity is the constraint. A federal inspection staff of 113 for more than three million miles leans heavily on state inspectors and on operators’ own integrity programs, and PHMSA’s research budget, about $13.7 million in FY 2022 per the PHMSA R&D report to Congress, is small next to the cost of a single major rupture.
| Metric | Value | Source |
|---|---|---|
| Integrity assessment interval for piggable onshore hazardous liquid segments outside high consequence areas (“piggable” means an ILI tool can pass through) | At least once every 10 years | PHMSA 2019 Hazardous Liquid Rule overview |
| Deadline for hazardous liquid lines in or affecting high consequence areas to accommodate ILI tools | Within 20 years, unless construction cannot be modified; rule effective July 1, 2020 | PHMSA 2019 Hazardous Liquid Rule overview |
| Federal Office of Pipeline Safety staff (February 14, 2025) and state pipeline inspectors (2024) | 326 FTE, including 113 inspectors and 12 accident investigators; 444 state inspectors | CRS R44201, 2025 |
| PHMSA pipeline enforcement, 2020-2024 | 1,081 enforcement actions; about $44 million in proposed civil penalties | CRS R44201, 2025 |
| PHMSA pipeline safety R&D projects and PHMSA funding, FY 2021 vs FY 2022 | 17 projects ($12,169,481) vs 23 projects ($13,696,428) | PHMSA FY 2021-2022 R&D Report to Congress, 2024 |
| Technologies commercialized from the PHMSA research program since 2002 | 35 | PHMSA FY 2021-2022 R&D Report to Congress, 2024 |
| Scope of PHMSA advisory bulletin ADB-10-03 on girth weld quality in new large-diameter pipe | Pipe 20 inches or greater, grade X70 and higher, built in 2008 and 2009; failures well below 72% of specified minimum yield strength | PHMSA ADB-10-03, Federal Register 2010 |
Context note: ADB-10-03 (most recent available data: 2010) followed girth weld failures during hydrostatic tests and early service on several newly built pipelines; it gives no failure count, only “several recent projects”. PHMSA’s research report gives a four-program total of $28,010,135 that does not reconcile with its two yearly figures (see Data watch).
Summary: Ultrasonic Pipeline Defect Detection by the Numbers
| Metric | Value | Source |
|---|---|---|
| Flaw size at 90% POD in thick-walled girth welds: phased array UT vs radiography | 6.87 mm vs 12.72 mm | Journal of Materials Research and Technology 2025 |
| False-positive rate: phased array UT vs radiography | 13.59% vs 0.58% | Journal of Materials Research and Technology 2025 |
| a90/95 flaw height, six AUT systems | 2 to 4 mm | Edison Welding Institute for PHMSA, 2011 |
| Thermal fatigue cracks in pipe welds: phased array a80 for depth | 1.41 mm | Materials, round-robin PAUT study, 2023 |
| Tube weld round robin: flaw size at 80% POD, carbon steel vs dissimilar metal welds | 4.53 mm vs 11.58 mm | Applied Sciences, 2022 |
| US significant pipeline incidents, 2024 vs 2014-2024 average | 235 vs 265 per year | CRS R44201, 2025 |
| Reportable US incidents, January 2024 to July 2025 | 881 (over 400 significant) | Pipeline Safety Trust testimony, 2025 |
| People killed and hospitalized by US pipeline incidents, January 2024 to July 2025 | 18 killed, 45 injured | Pipeline Safety Trust testimony, 2025 |
| Corrosion share of US incidents, 2024 vs historical | More than 25% vs roughly 18% | AMPP US Pipeline Industry 2026 report |
| Equipment or material failure share, significant gas transmission incidents 2010-2022 | 43% | GAO-24-106690, 2024 |
| Girth weld defect share of UK product-loss incidents, 1962-2022 | 18.8% (39 of 207) | UKOPA Product Loss Report, 2024 |
| European gas transmission failure frequency, 2018-2022 vs 1970-2022 | 0.101 vs 0.277 per 1,000 km-years | 12th EGIG Report, 2024 |
| EGIG construction defect and material failure frequency, 2018-2022 | 0.016 per 1,000 km-years | 12th EGIG Report, 2024 |
| UK onshore pipeline failure frequency, last 5 years | 0.042 per 1,000 km-years | UKOPA Product Loss Report, 2024 |
| Share of European oil pipeline length inspected by ILI tools in 2023 | 26% (8,260 km) | Concawe Report 8/25, 2025 |
| Keystone 2022 rupture: estimated release and total cost | 12,937 barrels; $600,793,659 | PHMSA Failure Investigation Report, 2024 |
| Integrity assessment interval for piggable hazardous liquid lines outside high consequence areas | At least once every 10 years | PHMSA 2019 Hazardous Liquid Rule overview |
| Technologies commercialized by PHMSA research since 2002 | 35 | PHMSA FY 2021-2022 R&D Report to Congress, 2024 |
Methodology and Sources
We read each primary document directly during October 2026 and kept only figures stated in it. Percentages were checked against their numerators and denominators where the source gave both.
- Payao Filho et al., Probability of detection of discontinuities in 9% Ni steel pipe girth welded joints: comparison between phased array ultrasonic and radiographic testing, Journal of Materials Research and Technology, 2025
- Reliability Analysis of PAUT Based on the Round-Robin Test for Pipe Welds with Thermal Fatigue Cracks, Materials, 2023
- Reliability Assessment of PAUT Technique in Lieu of RT for Tube Welds in Thermal Power Plant Facilities, Applied Sciences, 2022
- Herceg et al., Comparison of Ultrasonic Phased Array and Film Radiography in Detection of Artificially Embedded Defects in Welded Plates, Materials, 2023
- Edison Welding Institute for PHMSA, Advanced Technologies and Methodology for Automated Ultrasonic Testing Systems Quantification, 2011
- PNNL for the US NRC, NUREG/CR-7204, Applying Ultrasonic Testing in Lieu of Radiography for Volumetric Examination of Carbon Steel Piping, 2015
- Congressional Research Service, DOT’s Federal Pipeline Safety Program: Background and Issues for Congress, R44201, May 2025
- US GAO, Gas Pipeline Safety: Better Data and Planning Would Improve Implementation of Regulatory Changes, GAO-24-106690, April 2024
- PHMSA Failure Investigation Report, TC Oil Pipeline Operations, Washington County, Kansas, December 7, 2022 rupture (published 2024)
- PHMSA, Safety of Hazardous Liquid Pipelines Final Rule (2019) overview
- PHMSA, FY 2021-2022 Pipeline Safety R&D Update Report to Congress, December 2024
- PHMSA Advisory Bulletin ADB-10-03, Girth Weld Quality Issues, Federal Register, March 24, 2010
- Pipeline Safety Trust, testimony of Bill Caram to the House Energy and Commerce Subcommittee on Energy, July 22, 2025
- AMPP, US Pipeline Industry 2026: Leading Failures, New Technologies, and Future Outlook, March 2026, as reported by World Pipelines
- EGIG, 12th Report of the European Gas pipeline Incident data Group, December 2024
- UKOPA, Pipeline Product Loss Incidents and Faults Report 1962-2022, UKOPA/RP/24/001, November 2024
- Concawe, Performance of European cross-country oil pipelines: statistical summary of reported spillages in 2023 and since 1971, Report 8/25, July 2025
- Willems, Guajardo and Kopp, Sizing Crack Indications from Ultrasonic ILI: Challenges and Options, Pipeline Technology Conference 2017
- Guajardo and Urrea (NDT Global), API 1163 validation case study for an ultrasonic crack tool, IPC2026 abstract
- NDT Global, case study on verifying emerging pipeline cracking threats with ultrasonic technology (vendor material)
- Data watch: PHMSA’s 20-year incident trend dashboards could not be read directly, so US counts come from CRS, GAO and the Pipeline Safety Trust, each quoting PHMSA data under different windows and definitions (serious, significant, all reportable). No national US figure for miles inspected by ILI in 2024 or 2025 was publicly available in readable form; operators report it on PHMSA annual report forms, but no aggregate was found. We found no public 2023-2026 study giving TOFD-specific POD figures for pipeline girth welds, so TOFD is defined but not quantified here. The AMPP corrosion share is taken from trade-press coverage because the report page was not readable. Source inconsistencies: Concawe gives the 2023 active network as both 32,327 km and 32,367 km and the 10-year ILI average as both 22% and 21%; UKOPA gives the 2022 network as 23,577 km in the text and 23,574 km in its table; the Keystone report gives 12,937 barrels released in one section and 12,973 barrels recovered, disposed of or evaporated in another; PHMSA’s R&D report gives a four-program total of $28,010,135 that does not match its FY 2021 and FY 2022 figures ($25,865,909 combined); the 2022 tube weld study gives volumetric 80% POD as 6.55 mm in its results and 6.57 mm in its conclusions; and the NDT Global case study writes its tolerance as 1 mm (0.4 in), where 1 mm is about 0.04 in. The JMRT 2025 weld study used artificial discontinuities in 9% Ni steel welded with a nickel alloy filler, so its numbers may not transfer to carbon steel line pipe. Newer data expected soon: the 13th EGIG report, the Concawe report on 2024 spillages, the next UKOPA product loss report covering 2023, and PHMSA’s 2025 incident and annual report data.
Last updated: October 25, 2026. We update this roundup quarterly, and the next refresh is expected when Concawe publishes its statistical summary of European oil pipeline spillages in 2024.