Shared Action Plan Year 1 Report

Utilities Enacting the Leak Extent Method (2019/2020)

Executive Summary

Massachusetts has the second oldest natural gas infrastructure in the country. Old pipes leak methane, a greenhouse gas 86 times more potent than carbon dioxide over the first 20 years in the atmosphere.1

The amount of gas leaked annually from the Commonwealth’s aging gas distribution system is equivalent to 40% of the emissions of all of the state’s stores and businesses combined.2 The cost of this wasted gas is passed on to the customers, estimated to be over $11 million per year.3 In addition to polluting the air, methane suffocates street trees as it seeps into their root zones, depriving them of oxygen.

Research by Boston University and Gas Safety Inc. in 2016 showed that just 7% of the greater Boston distribution system leaks emit half of all the gas by volume, creating a clear policy opportunity.4 Later the same year, the Massachusetts Legislature enacted a law requiring that these leaks of significant environmental impact (SEIs) be repaired, since doing so would cut methane emissions in half for the least cost to the utilities and the least disruption to cities and towns.

However, given that gas companies had always been mandated to focus on the explosive potential of a leak and not emissions, they had no reliable and accurate method to identify these largest leaks that have a significant environmental impact. In 2017, HEET coordinated a large collaborative study working with Columbia Gas MA (now Eversource Gas of MA), Eversource Gas, and National Grid Gas, together with Gas Safety Inc., Mothers Out Front and other stakeholders. This research team field tested multiple methods and found the leak extent method5 was a quick, effective and low cost solution.6

This report documents the progress of this first-in-the-nation program and reports on the use of this new identification protocol – the leak extent method7– to identify and repair SEIs in the 2019 dig season in Columbia Gas, Eversource Gas, and National Grid Gas territories. HEET independently verified the results, with Gas Safety Inc. and provided analysis.

In 2019, HEET was also awarded a Pipeline and Hazardous Materials Safety Administration (PHMSA) grant to perform an extensive 4-seasons study to investigate how 100 SEIs evolved over time. This report includes a summary of this study’s findings and recommendations.

Massachusetts is the first to enact legislation to identify environmentally significant leaks, the first to determine an SEI protocol, and the first to test it widely in the field across multiple gas companies. We hope to report in coming years that we are also first in the nation to cut in half our methane emissions from the gas distribution system.

Results from Year 1 of the Shared Action Plan indicate that gas distribution companies can successfully identify and repair gas leaks that emit the most methane in order to cut emissions most effectively. 

Shared Action Plan Year 1 Findings at a Glance

  • Gas company’s reported use of the leak extent method significantly increased.
  • Whilst gas companies continued to use the leak extent method to identify SEIs in the field, they appeared to be over-reporting the actual number of SEIs, and reporting larger leak extents than the real extents, potentially because the protocol is still new.
  • Columbia Gas and Eversource found lower rates of SEIs than in previous years, possibly due to these utilities having begun to identify and repair their SEIs in 2018, before the regulation was formally enacted. Should this prove true, we recommend utilities continue using the leak extent method to measure leaks, whilst adjusting the SEI definition threshold to continue to capture the largest 7% of leaks.
  • SEI repairs continue to appear to be successful only 1/3rd of the time. Improving this repair rate will save money, emissions and work hours.
  • We didn’t see the correlation we expected between FluxBar8 steady state flux and leak extent and will continue to study this relationship.
  • The cavity ringdown spectrometer continues to be an excellent tool for finding new potential gas leaks.

Pipeline and Hazardous Materials Safety Administration (PHMSA) TAG Project Findings at a Glance

  • SEI leak extent size was not a predictor of SEI leak extent growth rate.
  • SEI grade 3 leaks evolved in terms of hazard potential at a similar rate to non-SEI Grade 3 leaks, with about 13% of both categories becoming grades 2 or 1.
  • SEIs were found predominantly on low pressure 6″ cast iron pipes installed before 1930 and intermediate pressure 2″ coated steel pipes installed after 1930. Given that the same size hole on an intermediate pressure pipe will leak gas at a faster rate than one on a low pressure pipe, prioritizing leaks on intermediate pressure 2” coated steel pipes may be an effective method of cutting emissions rapidly.


  • Utilities work with HEET to continue improving the adoption and use of the leak extent method to identify super-emitting leaks
  • Utilities continue using the leak extent method to measure leaks, whilst adjusting the SEI definition threshold to continue to capture the largest 7% of leaks
  • Explore using a cavity ringdown spectrometer as a tool for finding new potential leaks
  • Our finding that 13% of Grade 3 leaks across both regular and G3SEI categories did evolve grade in the course of a year calls into question the definition of Grade 3 leaks being ‘non-hazardous’ and raises questions regarding leak classification. To ensure such grade evolutions are caught promptly, we recommend assessing the need for standardizing Grade 3 resurvey rates and potentially enhancing 49 CFR § 192.173 – Distribution systems: Leakage surveys.
  • Our finding that repairs were only fully successful 36% of the time was concerning. This low rate may be due to multiple leaks existing at those leak locations, with only a subset being identified and repaired. Improving surveying protocols to better find these multiple constituent leaks may improve repair and safety outcomes and we recommend considering enhancing 49 CFR § 192.173 – Distribution systems: Leakage surveys. Additionally, further study of leak repair failures to increase success rate may be useful, given the cost efficacy of repairs vs. pipe replacement.
  • Our finding of G3SEIs occurring most commonly on two categories of leak prone pipe can be used by gas utilities to more efficiently identify and prioritize G3SEI leaks for repair. Given an initial Massachusetts gas utility study9 indicated that leaks with higher leak flux were on pipes with higher gas pressure, which seems intuitive, we further recommend the prioritization of the intermediate steel category over the low pressure cast iron category. In order for PHMSA to act on this, we suggest further study and data collection of G3SEI flux (see Recommended further study and research).
  • Finally, HEET urgently recommends that all gas utilities around the country follow the lead of Eversource Gas, National Grid Gas, the former Columbia Gas MA, and the Commonwealth of Massachusetts in prioritizing the identification and repair of G3SEIs. This is an effective method for cities and municipalities to accelerate decarbonization safely and cost-effectively, while moving towards their greenhouse gas reduction goals as quickly as possible.

Download the full report here.

  1. IPCC Climate Change Report, “Climate Change 2013: The Physical Science Basis,” Table 8.7
  2. See page 5
  3. See Appendix 4, Annual Total Cost from the Distribution System for more information. Reference is calculated using findings from McKain et al, Methane emissions from natural gas infrastructure and use in the urban region of Boston, Massachusetts
  4. Hendrick et al, “Fugitive methane emissions from leak-prone natural gas distribution infrastructure in urban environments”
  5. Originally suggested by Bob Ackley of Gas Safety Inc. – see Appendix 2
  6. Magavi, Z., Ackley, R., Hendrick, M., Salgado, E., Schulman, A., Phillips, N., “A Method of Identifying Large Volume Leaks in Natural Gas Distribution Systems”, publication pending.
  7. See Appendix 2
  8. See FluxBar Adoption and Use
  9. Magavi, Z., Ackley, R., Hendrick, M., Salgado, E., Schulman, A., Phillips, N., “A Method of Identifying Large Volume Leaks in Natural Gas Distribution Systems”