Interplay of biogas gains and methane losses
AD is widely adopted for sludge stabilization in large municipal WRRFs because of economic and environmental benefits23. In the USA, ~64% of treated wastewater is processed at facilities with AD (Supplementary Fig. 1). However, the net emissions and resulting climate benefits from biogas recovery depend on effective control of methane leakage. Historically, regulatory oversight and emission inventories have largely overlooked fugitive emissions from the biogas supply chain, due in part to a lack of representative emission factors and accurate facility-specific data24. As biogas projects gain popularity, understanding leakage thresholds and their implications for net climate outcomes has become increasingly important. To address this, we developed a net GHG emission model for the US wastewater sector to quantify emissions of two biogas utilization pathways (CHP combustion or upgrading to RNG) and to identify leakage thresholds for net climate benefits (Fig. 1a). The model incorporates energy inputs, avoided emissions from renewable energy recovery and fugitive methane leakage, enabling consistent comparisons across operational and grid decarbonization scenarios (Fig. 1b).
Fig. 1: System boundary and model overview.
The alternative text for this image may have been generated using AI.
a, Wastewater sludge treatment in WRRFs. The control baseline is facilities with sludge treatment but without biogas use. Treat offsite (also referred to as offsite biogas use) refers to facilities that produce biogas and transport it to external plants for subsequent use. Treat onsite refers to facilities that produce and use (CHP or RNG) biogas onsite. For facilities with AD biogas generation, two utilization pathways including CHP and RNG were considered and compared. b, Simplified schematic of the evolving net climate benefits model framework with scenario illustrations. Credit: EMM region shapefiles from the US EIA (January, 2024).
Figure 2 depicts how the interplay between biogas energy recovery and methane leakage rates (0–20%, reflecting the range observed in most empirical data), lead to the changes of net GHG emissions. The net emissions increase with higher methane leakage, with increasing speed varying depending on sludge generation level and technology types; each panel shows a net-zero leakage threshold, defined as leakage rate at which the net emissions curve crosses zero. Below this threshold, biogas recovery yields climate benefits. Once leakage reaches or exceeds the threshold, total emissions turn net-positive and the WRRF no longer achieves climate benefits. This net-zero emission threshold varies by facility, technology type and operating scenario.
Fig. 2: National-scale methane leakage thresholds for achieving net-zero emissions from biogas use at US WRRFs.
The alternative text for this image may have been generated using AI.
a–d, Net GHG emissions from CHP systems under BAU (a) and CET (c) and from RNG systems under BAU (b) and CET (d), shown as functions of CH4 leakage rate, overlaid with the compiled leakage observations. Shaded regions indicate the upper- and lower-bounds of simulation results. Horizontal red lines and shaded regions mark CH4 leakage thresholds for achieving net-zero emissions; exceeding these thresholds results in net-positive emissions and adverse climate impacts. Vertical grey lines indicate net-zero emission point. Dots represent observed leakage rates, independent of the model. ADR, AD heat demand only, where recovered heat is limited to onsite digester heating; BAU electricity scenario (2022 grid emissions); CET scenario aligned with the carbon neutrality goals of the states under 2050 grid projections; and FHR, where all CHP-generated heat is fully used.
Under business-as-usual (BAU) electricity conditions, a 7–10% CH4 leak will completely offset CHP energy-related emission benefits, while this threshold drops to 5–6% for RNG systems (Fig. 2a,b). This difference reflects the dual displacement of electricity and thermal energy with CHP, while RNG requires additional energy inputs for upgrading and compression. Heat use strongly modulates CHP performance. When recovered heat is used only to meet digester demand rather than fully recovered, net-zero thresholds decrease substantially (Supplementary Note 3).
Grid decarbonization reshapes these relationships. Under the clean electricity transition (CET) scenario, where the average power grid EI drops by over 50% by 2050, CHP threshold tightens to about 4–6% with full heat recovery, as the climate value of displacing grid electricity diminishes (Fig. 2c). In contrast, RNG threshold loosens to about 6–7% because the upstream electricity used for upgrading becomes cleaner (Fig. 2d). This highlights the strategic potential of RNG deployment in a clean energy future particularly for hard-to-abate sectors such as heavy-duty transport where direct electrification remains challenging.
When observed methane leakage rates from actual WRRF operations are plotted against the thresholds, the challenges become apparent. Over 30% of observed CHP leakage rates exceed the net-zero threshold under current condition (BAU), increasing to more than 60% in a cleaner future (CET). For RNG, over 80% of observed leakage rates exceed their respective thresholds, although sample sizes are limited. This discrepancy highlights a critical gap between theoretical climate targets and operational realities, underscoring the urgent need for better leak detection, quantification and mitigation to ensure biogas recovery truly advances emissions-reduction goals.
While national grid decarbonization is expected to change the dynamics and make the threshold stricter over time, locational differences also warrant attention. Specifically, regions with higher grid EIs show higher net-zero thresholds for CHP systems (Fig. 3a). In these regions, permissible leakage rates could reach up to 14%, nearly double the national average. Conversely, regions such as Northwest and Upstate New York already have a lower carbon footprint as a result of reliance on renewable sources. Even minor methane leaks can quickly negate or even surpass the environmental benefit of displacing grid electricity. Therefore, biogas deployment strategies should also be tailored to regional grid characteristics and decarbonization trajectories to ensure environmental integrity.
Fig. 3: Regional-scale methane leakage thresholds for net-zero emission biogas use at US WRRFs under different scenarios.
The alternative text for this image may have been generated using AI.
a, Net-zero emission methane leakage thresholds for CHP systems by subregion in the BAU-FHR scenario. b, Net-zero emission methane leakage thresholds for RNG systems by subregion in the BAU scenario. c, Comparative threshold ranges for CHP and RNG systems across subregions. Markers denote median methane leakage thresholds across facilities within each eGRID subregion. Error bars indicate the minimum and maximum region-level thresholds. Light blue bars show regional grid emission factors. Subregion abbreviations are provided in Supplementary Fig. 2. Credit: EMM region shapefiles from the US EIA (January, 2024).
Overall, CHP systems maintain consistently higher net-zero thresholds than RNG systems (Fig. 3c). Although regulatory and financial support for RNG has grown in recent years, adoption remains limited in many regions, reflecting barriers related to infrastructure requirements, pipeline interconnection and market access. In fact, regions without plotted RNG thresholds indicate no current RNG adoption on the basis of available data (Fig. 3b). Moreover, the climate benefits of RNG can be further eroded by methane emissions from the natural-gas distribution grid after injection. Studies have shown that methane emissions from urban distribution systems are substantially higher than inventory estimates, with ageing infrastructure in regions such as northeastern USA being a major contributor18,25. Delivered RNG inherits the leakage rate of the local distribution network, introducing a substantial, location-dependent emissions burden that is largely unaccounted for in life-cycle assessments. Addressing this gap represents an important direction for future research, particularly given the strong spatial heterogeneity and current lack of facility-level data.
Despite these pressing needs, clear regulations mandating methane leakage control in the biogas sector are lacking, particularly for wastewater applications. For comparison, Danish Biogas Association established a national goal of reducing total methane loss to less than 1% by 2020. When quantification indicates that plant methane loss exceeds 2% of annual production or 50 t yr−1, external consultants must suggest mitigation actions for each leakage source26. Similarly, the Recast EU Renewable Energy Directive (RED II) ties biogas eligibility to demonstrated life-cycle GHG reductions. The criteria inherently favour the use of biomethane in the transport sector over the heat sector as a result of a combination of a higher Fossil Fuel Comparator with a lower emissions savings requirement27. This policy-driven prioritization is strategically sound as electricity grids decarbonize, thereby reinforcing the strategic value of RNG in the future. These international precedents highlight the importance of facility-wide methane accounting as a foundation for credible climate performance not only for wastewater system but also across the broader biogas sector.
Distinct nature of methane losses from WRRFs biogas recovery
Our findings show that most observed leakage rates exceed methane leakage thresholds, indicating an urgent need for a deeper investigation. There is substantial variation in methane loss percentages, ranging from 0.40% to 65% of total methane produced based on the literature and field data collected (Supplementary Table 2). Most measured leakage exceeded the 5% threshold recommended by IPCC guidelines for AD of organic waste. It is worth noting that technical standards ensuring that unintentional CH4 emissions are flared should result in near-zero methane emissions28. In practice, however, methane slippage from open flares can be substantial. This discrepancy suggests systematic underestimation of emissions during biogas use, yet IPCC recommendations remained unchanged in the 2019 refinement.
After excluding statistical outliers (interquartile range (IQR) method), smaller facilities exhibited higher leakage rates (Fig. 4a), aligning with previous findings that plant size strongly influences methane emissions17,22,29. At the plant level (Fig. 4c), offsite biogas use demonstrated the lowest methane leakage rate, indicating substantial methane leakage primarily occurring during onsite gas handling and use30,31. Therefore, centralized biogas processing might offer environmental advantages, particularly for smaller facilities without adequate resources. Among onsite pathways, RNG systems exhibit higher leakage rates than CHP systems, probably introduced by the complexity of multistage upgrading processes (Supplementary Note 4). This finding contrasts with previous research suggesting minimal differences between RNG and CHP system32. The observed difference probably arises from our exclusive focus on WRRF-specific biogas technologies rather than general biogas facilities using diverse feedstocks, reinforcing the importance of sector-specific analysis. WRRFs present unique emission profiles that warrant dedicated study rather than reliance on broader biogas industry benchmarks that may underestimate actual performance variability.
Fig. 4: Methane leakage rate from biogas recovery in wastewater facilities.
The alternative text for this image may have been generated using AI.
a, Scale-dependent methane leakage across biogas utilization pathways. b, Facility-level methane leakage rate (%) from biogas recovery in wastewater facilities by digestate storage type. c, Facility-level methane leakage rate (%) by biogas utilization pathways, including utilities that used both pathways and offsite use. d, Unit-level methane leakage rate (%) across key process units, including feedstock handling, biomethane production (AD), biogas handling (CHP, upgrading to RNG), digestate management and miscellaneous sources. Boxplots show 25th, 50th and 75th percentiles and outlier-bounds are based on 1.5× IQR (equalling 75th percentile minus 25th percentile). Numbers represent the arithmetic mean. Violins represent probability distributions using kernel density estimation on either side. n, number of monitoring data.
At the unit process level (Fig. 4d), methane leakage was unevenly distributed across stages. Digestate handling (post-digestion treatment and storage) and biogas use are identified as primary leakage sources31,33,34,35,36,37,38. During biogas use, RNG units exhibited higher leakage compared with CHP units, corroborating plant-level findings. One should note that the sources of leakage determine where LDAR can deliver improvements and where technology-based recovery thresholds impose limits (Supplementary Note 4). In CHP systems, incomplete combustion leads to unavoidable emissions that can only be mitigated through post-treatment. By contrast, RNG systems experience losses primarily from historically lower biomethane recovery rates and venting of tail gas during upgrading. These can be reduced with advanced upgrading technologies and tail-gas oxidation as recommended by the US Environmental Protection Agency (EPA)39.
Digestate emissions occur as a result of continued microbial activities and release of residual methane gas bubbles from digestates. Across storage configurations (Fig. 4b) at facility level, WRRFs with open or not gas-tight digestate storage consistently exhibited the higher losses. While most WRRFs store digestate in closed tanks, these tanks are generally not integrated with the biogas handling system. Gas-tight digestate storage has proven highly effective in reducing emissions and is widely applied in agricultural digestion, yet remains underused in wastewater systems33,40. Recent comparative studies indicate that WRRFs exhibit systematically higher methane loss than agricultural biogas plants as a result of infrastructure and operational design choices17,41, rather than an inherent limitation of AD. While indoor gas-captured storage achieved the lowest losses, its adoption is constrained by higher infrastructure costs and operational complexity.
In summary, methane leakage is not an inherent flaw of wastewater biogas technology but a consequence of identifiable and controllable technical and operational issues (Supplementary Table 3). In particular, future work is needed to assess whether facilities with more advanced sludge configurations, such as full nutrient removal, exhibit systematically lower methane leakage. Such work would help to clarify the extent to which methane control is linked to overall management quality, rather than being driven solely by gas-handling infrastructure. However, detailed understanding of emissions from these sources remains limited, highlighting the need to improve data availability and quality to better assess emission impacts on WRRF carbon footprints and identify appropriate emission reduction measures.
Benefits of leakage minimization in WRRF biogas recovery
To quantify net GHG emissions holistically from biogas generation and utilization systems, a comprehensive assessment including both scope 1 and scope 2 emissions was conducted under three representative methane leakage scenarios using Monte Carlo simulation: (1) leakage observed (LO) subscenario, based on collected field data reflecting real-world conditions; (2) leakage controlled (LC) subscenarios representing potential reductions; and (3) leakage free (LF) subscenario, assuming zero methane leakage. Notably, incorporating observed methane leakage rates materially altered the results (Fig. 5), challenging the long and default assumptions that biogas systems have net climate benefits due to renewable energy production.
Fig. 5: Simulation of net emissions from biogas energy recovery at US WRRFs under BAU-FHR scenarios.
The alternative text for this image may have been generated using AI.
a–c, Net GHG EI per MGD treated under LO (a), LC (constrained to net-zero thresholds) (b) and LF (c) scenarios, respectively. Panels share a common legend. d, Distribution of net GHG emissions by biogas utilization pathway under the LO scenario. e, Energy generation across methane leakage scenarios. f, Comparison of average energy recovery intensity and associated economic value intensity across leakage scenarios. The red line denotes net-zero emissions. Marker size is proportional to s.d. derived from 1,000 Monte Carlo simulations per facility. Values shown next to bars (with ±) indicate the s.d. of intensity uncertainty derived from 1,000 Monte Carlo simulations.
Under leakage without control (BAU with full heat recovery and leakage observed, BAU-FHR-LO) scenario, CHP systems generally achieved lower net GHG emissions compared with RNG systems (P < 0.05). Approximately half of the CHP facilities achieved net-negative emissions (Fig. 5a), indicating measurable climate benefits. Emission distributions indicate that while both pathways exhibit variability, CHP generally shows a broader distribution of emissions, including both strongly negative and positive net emissions outcomes (Fig. 5d). Moreover, CHP systems remain climate viable for smaller WRRFs although the offset of electricity costs remains crucial for economic viability42. Notably, facility size did not substantially influence net emissions outcomes or uncertainties in emissions (P < 0.05). However, geographic region emerged as a strong determinant of emissions variability (P < 0.05), driven primarily by regional differences in grid EI as discussed in Fig. 3. For instance, WRRFs in regions with grid EI exceeding 300 kgCO2 MWh−1 generally achieved climate benefits using CHP. RNG systems generally concentrated more around moderate positive emissions outcomes, driven by higher methane leakage rate observed in RNG systems. Their narrower distribution also reflects smaller market share of RNG in WRRFs biogas recovery in the USA. When methane leakage was controlled at net-zero emission thresholds (as identified in ‘Interplay of biogas gains and methane losses’), theoretically all WRRFs achieved emission intensities near or below the climate-net-zero line (Fig. 5b), and emissions variability was substantially reduced. In the no-leakage case (Fig. 5c), nearly all facilities achieve negative net emission intensities, clearly demonstrating substantial climate benefits.
Beyond climate impacts, methane leakage also undermines energy efficiency and poses health and safety risks. Every unit of lost methane represents a missed opportunity to generate renewable energy while also contributing to potential fire, explosion and occupational exposure hazards. For example, in 2024 an internal gas leak at the Dubuque Water & Resource Recovery Center (Iowa, USA) triggered an explosion in a sludge-processing building43,44. Under no-leakage (BAU-FHR-LF) scenario, WRRFs achieved the highest energy outputs with some facilities surpassing 300 GWh annually (Fig. 5e). When leakage is introduced (BAU-FHR-LO), most WRRFs showed reduced energy production together with substantial increase in net emissions. This indicates that unit energy production remains relatively stable across leakage scenarios, whereas emission performance is highly sensitive to methane losses because of the high global warming potential of methane.
Economically, energy sales contribute substantially to operational cost recovery, meaning even modest improvements in leak reduction can yield important economic impacts. For instance, in a 500 million gallons per day (MGD) WRRF upgrading biogas to RNG, a 7% increase in energy recovery (by reducing leakage from 12% in BAU-FHR-LO to 5% in BAU-FHR-LC scenarios) represents approximately US$1–2 million in value at current natural gas prices (US$2–4 million MBtu−1)45. At the national level, implementing leakage controls in RNG systems increased average recoverable energy by ~52 MWh MGD−1. Importantly, energy carriers are not economically equivalent, even when derived from the same biogas stream. Although the energy intensity of electricity recovery is lower than that of RNG, its economic value is higher under current market conditions46 (Fig. 5f). In this analysis, the economic value of recovered heat was conservatively estimated using the industrial natural gas price (as a proxy for displaced thermal demand), while RNG was valued using the city-gate natural gas price, consistent with typical US WRRF biogas end-use pathways. In practice, however, RNG may displace diesel fuel in heavy-duty transport rather than pipeline natural gas and earn renewable fuel credits, in which case its economic value could be substantially higher than our conservative estimates. Overall, electricity and RNG generally access broader and higher-value markets, whereas recovered heat is largely confined to onsite or local applications with more limited revenue potential. Further research is needed to incorporate sector-specific fuel displacement and emerging carbon credits to more fully capture the evolving economic value of biomethane.
Methane leakage strongly affects financial viability and returns on investment for biogas energy recovery projects over long-term. For CHP systems, LDAR payback periods demonstrate dramatic sensitivity to leakage rates, increasing from less than 1 year at 20–30% methane loss to over 20 years when losses fall below 2% (ref. 32). RNG projects are more sensitive and in practice municipal facilities typically achieve paybacks on the order of several years. For example, South Platte Renew in Colorado recently reported a 5.5-yr payback under renewable fuel standard and low-carbon fuel standard (LCFS) incentives. These figures underscore that methane leakage rate is not only a climate imperative but also a direct driver of project economics. Municipal WRRFs generally face higher capital and operational costs due to stringent design requirements than do agricultural biogas systems, and as a result the economic benefits may take longer to realize32. However, systematic leakage detection and reporting can still meaningfully improve project performance, even if these benefits are not immediately apparent without consistent monitoring.