Comparative Carbon Intensity Assessment for Renewable Natural Gas Transportation Options

Abstract

Renewable natural gas (RNG) is associated with reduced emissions but a comparison of its transport modes is needed. For this study, carbon intensity (CI) values were calculated using the Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (GREET-2024 version) life cycle analysis model and its California version, CA-GREET 4.0. RNG transport by pipeline was found to have the lowest CI score over medium to long distances due to the high energy efficiency of the transmission network. Compressed natural gas (CNG) tube trailers had the lowest carbon intensity (3.2 gCO₂e/MJ) over shorter distances (up to 250 miles) due to minimal static emissions, but their limited payload capacity leads to higher delivery emissions (1.56 gCO₂e/MJ per 100 miles) over distances above 250 miles. As a result, liquid natural gas (LNG) trailer transport, despite having higher static emissions (10.2 gCO₂e/MJ), becomes more favorable over longer distances (above 650 miles) due to its superior delivery efficiency. These findings suggest that while CNG transport is advantageous for short hauls, pipeline transmission remains the most efficient option, with LNG trailers only becoming competitive over long distances (over 900 miles). The study provides utilities and RNG producers data to consider transport options with the lowest carbon footprint.

1. Introduction

The global effort to transition away from fossil fuels in response to rising atmospheric carbon dioxide (CO₂) levels, now also includes renewable natural gas (RNG) as a potential lower-carbon alternative to traditional natural gas. Pipeline quality RNG, with a methane (CH₄) concentration of over 95% and devoid of impurities, qualifies as a “drop-in fuel” for exhibiting the same properties as fossil natural gas. RNG is commonly made by upgrading biogas, a product of anaerobic decomposition of organic waste with a methane concentration of 45–65%, depending on the feedstock. This is significant since CH₄, considered a short-lived climate pollutant (SLCP), has a much higher (as high as 81.2 over a 20 year period) global warming potential (GWP) than CO₂. RNG is produced from a variety of sources based on agriculture/livestock residue, wastewater treatment, and landfills. Other sources and pathways to produce RNG include thermal gasification (TG), and power-to-gas (P2G). ,

Biogas collection and its utilization as RNG avoids methane emissions that would have occurred if the organic waste were managed via conventional practices or left unmanaged. RNG production is, therefore, incentivized at both federal and state levels, with producers benefiting from both renewable identification number (RIN) credits under the renewable fuel standard (RFS), and state program credits such as those under California’s low carbon fuel standard (LCFS). However, the sustainability of RNG depends not only on its production but also on its transportation, as different transport methods can vary in efficiency, cost, and emissions. Understanding the environmental impact of RNG transport is, therefore, critical to evaluating its role in decarbonizing hard-to-abate sectors.

RNG can be transported from the point of production to intended end use in the same modes as traditional natural gas (NG). Due to the relatively high energy efficiencies and resulting low costs, the predominant method of transporting RNG for large distances over land within North America is via existing gas pipeline networks, where it can be blended with conventional natural gas. This method involves the use of “book and claim” accounting, which has been a driver in clean and renewable energy development and continues to drive the expansion of solar and wind-generated electricity. There are concerns, however, that high rates of fugitive methane emission throughout the natural gas network could contribute significantly to greenhouse gas emissions, reducing the climate benefits of RNG. − Injecting into the grid also requires additional upgrading to meet gas quality specifications of the main pipeline system, increasing energy requirements.

Connecting a new source of natural gas to the existing network requires logistical changes to the current flow of gas, however, the details of which must be worked out ahead of time by all parties affected by these projects. Another option for RNG transport is a dedicated pipeline connecting an RNG producer directly to an end user, which offers the advantages of greater flexibility in both flow rates and gas quality. Such pipeline infrastructure by nature requires that few entities take on the financial burden, making them a practical option in instances where supply is located close to demand (<20 miles) and underpinned by long-term contracts. Alternatively, RNG can be transported via truck in the form of compressed natural gas (CNG) or liquefied natural gas (LNG). These modes, commonly referred to as the on-road “virtual pipeline”, are used to effectively transport gas when no pipeline network is available; the latter having the advantage of higher density at the cost of additional energy consumption to achieve liquefaction than for compression. A fundamental understanding of carbon intensity (CI) scores associated with available RNG transport modes could also help meet the challenge of partially feeding planned large data centers that are likely to put strain on the existing electric grid.

2. Methods

2.1. Greenhouse Gas Accounting

The most recognizable greenhouse gas is CO_2, and it is by far the main contributor to rising atmospheric CO₂ levels. There are other greenhouse gases, such as CH₄, which can have much higher radiative-forcing effects per molecule than CO₂, contributing to heating the earth higher despite existing in the atmosphere at much lower concentrations. For these gases, a GWP is used to describe their contributions to total atmospheric heating, typically using CO₂ as a reference. Unlike CO₂, other major greenhouse gases break down in the atmosphere over time. For example, over 20 years, 1 gCH₄ in the atmosphere will have the equivalent effect of 81.2 gCO₂, while over 100 years, it will have the effect of 27.9 gCO₂. For methane therefore, the 20 year (GWP-20) and 100 year (GWP-100) global warming potentials are 81.2 and 27.9 g of CO₂ equivalent (gCO₂e), respectively.

Using GWPs, it becomes possible to quantitatively describe the contributions of different molecules to atmospheric warming. The lifecycle carbon intensity (CI) of a fuel refers to the total GWP of all greenhouse gases emitted per unit of energy delivered and aims to include all aspects of the fuel lifecycle, from production and in-between conversions/transportation to end-uses, as well as how it may indirectly affect other products and markets. A megajoule (MJ) of delivered fuel was used as it is a commonly used functional unit that allows comparison between fuels of different forms. For this study, greenhouse gas emissions and resulting CIs were calculated utilizing a modified version of the Research& Development Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (R&D GREET) life cycle model developed by the U.S. Department of Energy’s Argonne National Laboratory (2024). GREET, updated yearly, contains a robust database of relevant emission factors and established product pathways from which upstream emissions, such as those for electricity, were accounted for. The greenhouse gases tracked by GREET pathways include CO₂, CH₄, N₂O, CO, NO _x , volatile organic carbons (VOC), black carbon (BC), and organic carbon (OC). All pathways were also modeled in CA-GREET 4.0 modela variant of GREET (2024), used by the California Air Resources Board for the Low Carbon Fuel Standard regulatory process.

When calculating CIs, an important distinction must be made between biogenic carbon and fossil carbon. The Intergovernmental Panel on Climate Change (IPCC) defines biogenic carbon as “carbon released as carbon dioxide or methane from combustion or decomposition of biomass or biobased products”. When a fossil fuel is burned, previously sequestered carbon in the earth is released into the atmosphere in the form of CO₂, increasing global atmospheric concentrations. This is different from when a renewable fuel, like wood, is burned. The carbon that makes up the wood originated from existing atmospheric CO₂, so burning it returns only that same carbon to the atmosphere and results in no net-increase in atmospheric concentration.

2.2. Modeling: RNG Production

Biogas collection and utilization as RNG avoids emissions that would have occurred if the organic waste was managed via conventional practices or left unmanaged. To quantify avoided emissions, a business as usual (BAU) scenario must be defined for each waste feedstock, since each waste is unique in the emissions it generates during its management. For example, wastewater treatment plants generate methane from the anaerobic digestion of sludge waste, which is most often flared in the case that it is not upgraded for use as RNG; therefore, gas flaring and its associated emissions define the BAU scenario. The total CI of RNG is calculated by subtracting the avoided emissions (BAU) from the emissions associated with the new process. It is important to note that the BAU scenario has great influence over the total CI of RNG, and that in practice, there is significant variability in BAU scenarios resulting in noteworthy variation in CI even within a particular feedstock. −

The source of RNG is considered in accounting for losses at different stages, which incur additional input requirements per unit of RNG throughput. GREET’s default wastewater treatment plant (WWTP) to RNG pathway was used as the RNG source, with a CI of 42.5 gCO₂e/MJ. This includes credit for avoided waste management (BAU) emissions, as well emissions from biogas production and upgrading to pipeline-quality RNG via pressure swing absorption (PSA). It should be noted, however, that this is a reference point, as CIs for various RNG production options vary considerably (Figure ), ranging from 5 to 50 gCO₂e/MJ. For this study, it was assumed that both the RNG production facility and end use facilities were located within the continental United States (U.S.), removing ocean freight as a viable mode of transport.

1.

Carbon intensity (CI) of various fuels and RNG from various waste sources. Reprinted with permission from Cyrs, T., Feldmann, J., & Gasper, R. (2020). Renewable Natural Gas as a Climate Strategy: Guidance for State Policymakers. Working Paper. Washington, DC: World Resources Institute. Available at https://www.wri.org/research/renewable-natural-gas-climate-strategy-guidance-state-policymakers. Copyright 2020 World Resources Institute. Licensed under https://creativecommons.org/licenses/by/4.0/.

2.3. Modeling: Transport

Below, key RNG transport methods and assumptions are outlined.

2.3.1. Common-Carrier Pipeline

Transport of RNG using the existing natural gas pipeline network consists of long-distance transmission pipelines followed by a local distribution system. GREET accounts for transmission leakage and venting in a pipeline scenario with an emission factor of 50.5 mgCH₄/MJ per 680 miles, and for leakage and venting in the distribution network with a flat emission factor of 16.8 mgCH₄/MJ. Critically, GREET RNG for off-site refueling pathways only include transmission emissions for pipeline transport to a refueling station in its calculations. Therefore, an alternate pathway was modeled to include the distribution pipeline connection to the end user using the aforementioned emission factor. Embedded emissions for natural gas delivered through the existing pipeline network are taken to be negligible on a per MJ basis based on prior estimates of steel and diesel requirements for pipeline construction and installation. Following delivery, RNG is compressed from delivery pressures (50 psia) to 4800 psia for refueling. This compression was modeled as an adiabatic process using the GREET compression energy calculator, which assumes the use of multistage electrical compressors with a compression ratio of 2.1 per stage and an adiabatic efficiency of 65%. , This translates to an electricity requirement of 0.021 MJ per MJ of RNG.

2.3.2. Modeling: Virtual Pipeline (Trucking)

The CI of vehicle-cycle (manufacture and disposal) emissions of a class 8 diesel truck is estimated at 65 gCO₂e/mile: however, this report will not attempt to address embedded emissions under the assumption that existing trucks are likely to be used.

Transportation of natural gas in the form of LNG begins with on-site liquefaction. The cryogenic liquid is then transported by truck or train to its destination, where it is unloaded and stored for an amount of time dependent on the station’s capacity and throughput. Liquefaction was modeled in GREET assuming a single mixed refrigerant (SMR) liquefaction cycle requiring an average of 0.043 MJ of grid electricity per MJ of RNG. This assumption is based on a review of liquefaction technologies conducted for the addition of landfill gas to the official GREET model, which concluded that small-scale RNG projects were more likely to deploy simpler SMR cycles over marginally more efficient, large-scale liquefaction technologies. GREET calculates emissions from LNG storage using a default boil-off rate of 0.1%/day, a boil-off recovery rate of 80%, and an average storage time of 3 days. Transportation in the form of CNG begins with on-site compression for injection into a tube trailer. This compression was modeled with an adiabatic efficiency of 65% using grid average electricity and assumes compression to 3600 psia for tube trailer transport from upgrade outlet pressure of 70 psia. This translates to an electricity requirement of 0.016 MJ per MJ of RNG.

On-road LNG trailer transport was modeled using emission factors for a class 8B truck with a standard diesel engine and a payload of 15 tons of LNG. CNG tube trailer transport was modeled using GREET’s on-road LNG trailer transport pathway as a template; Class 8 diesel trucking emission factors were used, with an adjusted payload of 5 tons to simulate the lower RNG capacity of CNG tube trailers. , Both boil-off emissions and associated loss factors were omitted from CNG transport calculations.

2.3.3. Modeling: Purpose-Built, Dedicated Pipeline

A dedicated pipeline connection between production and end use facilities was considered as an alternative to the use of existing NG transmission and distribution pipelines. If a new pipeline is to be constructed, the emissions related to both pipeline materials (typically steel or plastic), and construction must be considered. Figure shows the CIs for the material production and pipe manufacturing for various pipeline sizes and materials including carbon steel (CS), high density polyethylene (HDPE), and concrete. Chohan et al. (2023) also evaluated transportation and installation related emissions which combined account for less than 5% of total embedded emissions.

2.

CI of the material production and pipe manufacturing phase of different pipeline materials. Reprinted with permission from Chohan, I. M.; Ahmad, A.; Sallih, N.; Bheel, N.; Ali, M.; Deifalla, A. F. A Review on Life Cycle Assessment of Different Pipeline Materials. Results in Engineering 2023, 19, 101325. Copyright 2023 Results in Engineering (Elsevier). Licensed under https://creativecommons.org/licenses/by/4.0/.

Assuming a project lifetime of 20 years, which is typically used when evaluating the impacts of potential RNG projects, , CI per MJ of RNG can be calculated as a function of project throughput. Most existing RNG projects produce between 30,000 and 500,000 ft³ of biogas/day but there are a few newer projects with production rates much greater, skewing the U.S. national average to 3 million ft³/day.

$$\mathrm{E}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{d}\mathrm{d}\mathrm{e}\mathrm{d}\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{s}=\frac{\left(\frac{\mathrm{e}\mathrm{m}\mathrm{i}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{s}}{\mathrm{m}\mathrm{i}\mathrm{l}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{p}\mathrm{i}\mathrm{p}\mathrm{e}}\right)\times \mathrm{d}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}(\mathrm{m}\mathrm{i})}{\mathrm{l}\mathrm{i}\mathrm{f}\mathrm{e}\mathrm{t}\mathrm{i}\mathrm{m}\mathrm{e}\times \mathrm{t}\mathrm{h}\mathrm{r}\mathrm{o}\mathrm{u}\mathrm{g}\mathrm{h}\mathrm{p}\mathrm{u}\mathrm{t}}$$

The operation of a dedicated connection was modeled as a simplified local distribution pathway, preceded by on-site compression to desired distribution pressures. It is likely that the meters and other fixtures will be the primary sources of leakage, as is the case with pipelines generally. CA-GREET (GREET Customized for California) uses an emission rate of 16.8 mgCH₄/MJ (17.7 gCH₄/MMBtu). The age of a new dedicated connection and its relative simplicity suggest, however, that these emissions are likely an overestimation. Therefore, this represents a conservatively high estimate of emissions. Similar to transmission pipelines, a dedicated pipeline would require additional compression for longer distances, however, due to the lower efficiencies associated with smaller diameter pipe, a dedicated pipeline but would likely require recompression more often. A distance of 50 miles was chosen to represent the maximum delivery distance before repressurization would be required though it should be noted that this is highly speculative, as no existing projects could be identified that employed such a long low-volume pipelines.

2.4. Losses and Fugitive Emissions

Other factors to consider in calculating CI of RNG are methane loss and fugitive emissions. Losses refer to methane emissions inherent to the transport of RNG, such as the boil-off loss from LNG, while fugitive emissions refer to potentially mitigatable sources of emissions, such as leaks or spills. Methane can leak during production of RNG directly from the anaerobic digester, during upgrade, and during transport and storage. Additional leaks can also occur at the refueling station if used as transportation fuel. Refueling station “site-specific” leaks typically result from faulty connections, damaged/worn infrastructure, or during refueling events. These emissions are highly variable and are typically small; however, the distribution of these emissions is believed to be highly skewed with a small share of stations representing most fugitive emissions, and most fugitive emissions occurring during relatively rare, unintended high-loss events.

It should be noted that refueling station leakage and losses are not included in the GREET due to uncertainty in measurements. While this may be fair for individual stations with relatively low rates of leakage, cases of disproportionally high leakage are likely to raise the average contribution to CI from refueling stations on a larger scale. Even within one station, emissions can be highly variable. A study of six LNG refueling stations and 43 refueling events found an average methane emission rate of 17.7 gCH₄/event, however, the authors reported that over 43% of the recorded methane emissions occurred during a single event.

In the case of LNG, the boil-off must also be accounted for. Boil-off refers to the process in which the LNG stored in cryogenic tanks undergoes a phase change, from liquid to gas. Despite LNG being stored and transported in well-insulated tanks, a slight variation in temperature leads to a gradual buildup of pressure that requires venting to the atmosphere, recapture or use, or any combination of the three. , To account for the boil-off, GREET uses a boil-off rate of 0.1%/day in transport or storage, and assumes a recapture rate of 80%. The remaining 20% is assumed to be vented to the atmosphere.

3. Results and Discussion

3.1. Integrated Model

A reference RNG system was developed to conceptually capture the contributions to life cycle of various RNG pathways as shown in Figure .

3.

Reference RNG system for the scope of this studyintegrating resources, production, transportation, and use (RNG carrier and transport calculated using ANL-GREET). * Residential heating CIs from Merrin and Francisco, 2019. Vehicle and industrial heat/power emissions from CA-GREET.

3.2. Common-Carrier Pipeline

GREET modeling resulted in a CI for the transmission pipeline of 1.2 gCO₂e/MJ per 100 miles, of which fugitive emissions were found to account for approximately half of (0.6 gCO₂e/MJ per 100 miles). The remaining 50% per mile CI results from emissions associated with recompression and storage stations along the pipeline powered by NG turbines. The distribution network, modeled with a leakage rate of 16.8 mgCH₄/MJ, contributes 1.4 gCO₂e/MJ to the total CI. The frequency and intensity of leaks in the distribution network depends on both the characteristics of the material and age of the pipelines, as some of the pipelines were laid several decades ago. Considering this, the frequency of leaks for various distribution pipeline materials and ages is shown in Figure . The materials for this study were chosen to represent the entire gas pipeline infrastructure in the United States that are classified as low-pressure/small diameter and high-pressure/large diameter pipelines. The same study found that current pipelines (21% bare steel, 21% cast iron, 22% coated steel, 36% plastic) have an average emission factor of 2 gCH₄/min/leak. This leakage rate results in significantly higher distribution network emissions rates (up to 5 times higher) than currently employed by GREET. , Another study employing direct measurement and surveying of local distribution companies calculated a methane emission rate of 0.1–0.22% (17.9–39.4 mgCH₄/MJ) delivered through local distribution pipelines. Final compression at the refueling station, modeled as described, adds a CI of 2.8 gCO₂e/MJ.

4.

Estimated activity factors as a function of pipe installation age (in years) and material. The data shown is for four pipeline materials (bare steel, cast iron, coated steel and plastics). The data points in the plots show circlesthe size of each circle is related to increasing distance the gas travels in the pipeline The plotted data in Figure is scattered but some segments through which the lines are drawn do show trends. Generally, the leakage rate generally increases with (1) pipeline age and (2) distance. Reprinted with permission from Weller, Z. D.; Hamburg, S. P.; von Fischer, J. C. A National Estimate of Methane Leakage from Pipeline Mains in Natural Gas Local Distribution Systems. Environ. Sci. Technol. 2020, 54 (14), 8958–8967. Copyright 2020 American Chemical Society.

3.3. Dedicated Pipeline

Using GREET, compression from upgrade outlet pressures (about 70 psia) to distribution pressures (200 psia) was found to have a process CI of 0.9 gCO₂e/MJ when powered by US average electricity, and compression at a refueling station a CI of 2.8 gCO2e/MJ. Including pipeline leakage (1.4 gCO₂e/MJ), this results in a CI value of approximately 5.1 gCO₂e/MJ for dedicated pipelines over short distances prior to the consideration of embedded emissions, with an additional 0.9 gCO₂e/MJ for repressurization every 50 miles transported. These results are summarized in Table . It should be noted that the contributions to CI from embedded emissions were found to be very small in comparison and do not significantly affect total CI calculations at scales normally employed. Dedicated pipelines are typically only a few miles long and have not been included in the following comparative discussion.

1. CI Values for Each Step in the Transport and Use of RNG via Dedicated Pipeline to a CNG Refueling Station for Use as Transportation Fuel.

pathway steps	ANL-GREET CI [gCO2e/MJ]
compression to dist. pressure	0.5–1.4
dedicated pipeline leakage	1.4
off-site compression	2.8
transport and delivery total	5.1
>50 mi: route recompression	+0.9/50 mi

3.4. Virtual Pipeline

Using U.S. average grid electricity, the CI of liquefaction was calculated to be 5.5 gCO₂e/MJ, while compression for truck transport was found to have a CI of 2.5 gCO₂e/MJ. LNG storage results in a marginal CI of 1.0 gCO₂e/MJ from venting of boil-off gas before and after truck transport. Following delivery, RNG is compressed to vehicle fueling pressure. The compression process adds a CI value of 2.8 gCO₂e/MJ for LNG-trailer transport. Additional compression may also be necessary to bring CNG from delivery pressures (∼3600 psia) to fuel pressures (∼4800 psia) during CNG-trailer transport, for a compression CI value of 0.69 gCO₂e/MJ.

3.5. Transport Pathways for End Use as CNG Transportation Fuel

This section evaluates the emissions associated with transporting RNG for refueling of CNG vehicles via pipeline, on-road CNG, or LNG tube trailer, based on GREET. The 20 year carbon intensity was calculated for each step in the transport of RNG to a CNG refueling station for use as a diesel substitute. The total CI of each pathway can be broken down into static emissions, attributed as flat contributions to CI in gCO₂e/MJ, and delivery emissions, resulting from the transport distance and reported in gCO₂e/100 miles (Figure ) for comparison.

5.

Waste-to-wheel total comparative 20 year CI values for transportation of RNG via pipeline or on-road CNG/LNG tube trailers options. In each transport option, key intermediate steps that contribute to the total CI value are shown, along with associated CI values.

Static emissions for CNG tube trailer transport were found to be the lowest, at 3.2 gCO₂e/MJ, followed by pipeline transport (4.1 gCO₂e/MJ). This results in a range of 0–250 miles, in which CNG transport has the lowest CI. The relatively low payload capacities of CNG trailers and subsequent lower energy efficiencies, however, result in higher delivery emissions (1.56 gCO₂e/MJ per 100 miles) than either pipeline (1.21 gCO₂e/MJ per 100 miles) or LNG trailer transport (0.54 gCO₂e/MJ per 100 miles). The LNG trailer transport option was found to have a much higher static contribution to CI than either other transport mode, at 10.2 gCO₂e/MJ. As a result, the low delivery emissions associated with LNG transport do not become apparent until 650 miles, at which the CNG trailer transport CI surpasses RNG transport as LNG. Similarly, the LNG trailer transport CI becomes lower than that of the pipeline transport at distances over 900 miles. Figure summarizes these trends and Table presents resulting waste-to-wheels CI’s for each transport mode over 100, 500, and 1000 miles.

6.

A plot of 20 year CI values for the transport of RNG over varying distances for CNG refueling station (data from Figure ).

2. Total Waste-To-Wheels CI for Wastewater Sludge-Based RNG Used as a Transportation Fuel in CNG Heavy Duty Vehicles.

		20 year carbon intensity (gCO2e/MJ)
RNG transportation mode	transportation distance (mi)	ANL-GREET	CA-GREET
pipeline transmission & distribution (book & claim)	100	59.7	61.4
	500	64.5	67.0
	1000	70.5	73.9
CNG tanker truck	100	59.1	60.2
	500	65.4	65.6
	1000	73.1	72.6
LNG tanker truck	100	65.1	67.0
	500	67.3	68.9
	1000	70	71.3

In Table , all values shown include a contribution of 42.2 gCO₂e/MJ from production. The total also includes end-use emissions; 12.2 gCO₂e/MJ from CA-GREET, and 13.6 gCO₂e/MJ from ANL-GREET. The total does not include station leakages, which as noted before in the text, are highly variable.

4. Conclusions

To put the variability between procurement methodologies into perspective, this study calculated the comparative emissions associated with available RNG transport pathways. Specifically, three distinct pathways considered were: (1) common carrier pipeline, (2) virtual pipeline connection between producer and end-user, and (3) new direct pipeline connection between the producer and the end-user. The study then evaluated emissions from leakage as a function of distance, boil-off, emissions associated with compression and trucking, emissions embedded in equipment, and other relevant factors that contribute to the total emissions in each of the three pathways using ANL-GREET.

Our analysis found that the RNG transport by pipeline has the lowest CI over the distance range of 250–900 miles due to the high energy efficiency of the transmission network. However, static emissions from pipeline use (4.1 gCO₂e/MJ) and moderate delivery emissions (1.21 gCO₂e/MJ per 100 miles) mean that the pipeline CI is not the lowest over short distances. The CNG tube trailer transport exhibited the lowest static emissions of all three modes (3.2 gCO₂e/MJ), making it the lowest-CI option up to 250 miles transport. However, its relatively low payload results in higher delivery emissions (1.56 gCO₂e/MJ per 100 miles), causing its CI to exceed that of the pipeline beyond 250 miles, and that of the LNG transport beyond 650 miles. Although the LNG trailers have the highest static emissions (10.2 gCO₂e/MJ), their low delivery emissions (0.54 gCO₂e/MJ per 100 miles) lead to a crossover point where LNG becomes the lower-CI option compared to the CNG at 650 miles and the pipeline transport at distances over 900 miles. These results suggest that while the CNG transport method may be preferable for short distances, pipeline transmission remains the best option over most realistic distances, and LNG trailers only offer an emissions advantage for very long hauls, unlikely to occur in practice. These results offer a quantitative analysis for utilities and RNG producers to adopt a low-carbon pathway to deliver RNG to customers.

Original spreadsheets available upon request.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c09259.

• Explanation of the GREET model, description of modifications to GREET model, Figures S1 and S2 (PDF)

The authors declare no competing financial interest.