Analyzing the Carbon Footprint of an LNG Tanker Using Real Operational Data: Quantifying Methane Slip Effects

Abstract

This paper presents an exploratory operational assessment of the carbon footprint of an LNG tanker using real operational data collected by a continuous emission monitoring system over a ten-month period of vessel operation. The analysis included carbon dioxide (CO₂) and methane (CH₄) emissions from the main engines and diesel generators, the calculation of CO₂-equivalent using the GWP₁₀₀ and GWP₂₀ global warming potential factors, and a comparison with a hypothetical heavy fuel oil (HFO) operating scenario. The methodology is based on a Tier III approach, that is, on real operational data, which allows a more realistic assessment of emissions than approaches based on standard emission factors. The results show that CO₂ emissions make up the largest share of total emissions, but including methane emissions significantly increases the ship’s overall climate impact. Total methane slip was 3.62%, with diesel generators exhibiting higher slip than the main engines. When GWP₂₀ was applied, total emissions expressed as CO₂-equivalent were, in some periods, comparable to or higher than those estimated for the HFO scenario, despite lower direct CO₂ emissions. The emission distribution indicated that the main engines dominated CO₂ emissions, while methane emissions were more evenly distributed between the main engines and the auxiliary generators, with generators making a significant contribution to total CO₂-equivalent emissions due to their higher methane slip. The results confirm that any assessment of the climate performance of LNG-fueled operation must include methane emissions and should be based on real operational data; otherwise, the overall climate impact may be underestimated.

1. Introduction

International shipping plays a key role in global trade, but at the same time it is a significant source of greenhouse gas emissions [1,2]. It is estimated that the maritime sector accounts for about 2–3% of total global CO₂ emissions, with a rising trend linked to the growth of world trade [2]. In line with this, the decarbonization of the maritime sector has become one of the main goals of international regulatory policies, including the International Maritime Organization (IMO) strategy for reducing greenhouse gas emissions [3].

Current emission-reduction policies in shipping increasingly require broader greenhouse-gas accounting rather than focusing only on CO₂. The 2023 IMO Strategy aims to reduce the carbon intensity of international shipping by at least 40% by 2030 and to reach net-zero GHG emissions by or around 2050 [3]. At the European Union level, CO₂ emissions from large ships have been included in the EU Emissions Trading System (EU ETS) since 2024, while CH₄ and N₂O will be included from 2026 [4]. In parallel, the EU MRV Maritime framework has covered CO₂, CH₄ and N₂O since 2024 [5]. Recent decarbonization research also reflects this broader focus, including studies on alternative marine fuels [6], wind-assisted propulsion [7], and operational-data-based fuel-consumption prediction [8]. These developments make methane emissions increasingly relevant for both scientific assessment and regulatory compliance in LNG-fueled shipping.

For the quantitative assessment of the environmental impact of ships, an increasingly used concept is that of the carbon footprint. It is commonly expressed as carbon-dioxide-equivalent (CO₂-eq) [9,10,11], which includes CO₂ and other greenhouse gases weighted by their global warming potential (GWP). This approach allows a more realistic assessment of the climate impact of different fuels and engine systems [12,13,14].

In the analysis of shipping emissions, tank-to-wake (TtW), well-to-tank (WtT), and well-to-wake (WtW) approaches are commonly distinguished. These cover different stages of the fuel life cycle: TtW includes emissions during ship operation, WtT includes emissions related to fuel production and supply, while WtW represents their sum [15,16,17]. Although the WtW approach is necessary for a full assessment, TtW analysis based on real operational data remains essential for understanding actual ship emissions, especially methane emissions from LNG-fueled operation [18,19].

In the last decade, liquefied natural gas (LNG) has often been seen as a transition fuel that can reduce CO₂ emissions compared with conventional marine fuels such as heavy fuel oil (HFO) [20,21,22]. This is mainly due to its more favorable chemical composition, which leads to lower CO₂ emissions per unit of produced energy, as well as lower emissions of sulfur compounds and some air pollutants [23,24,25]. Considering the expected growth of the LNG fleet and the volume of LNG maritime transport [26], understanding actual emissions during ship operation is becoming increasingly important for assessing its long-term sustainability [18,19].

However, the use of LNG is also linked to emissions of unburned methane, known as methane slip, which can significantly affect the overall climate impact of a ship [18,19,27]. Methane has a much higher global warming potential than CO₂, especially over shorter time horizons, so even relatively small CH₄ emissions can increase the total CO₂-equivalent of emissions [12,13,14]. Recent studies show that the climate benefit of LNG is not always achieved and that, under certain conditions, methane emissions can partly or even fully offset the reduction in CO₂ emissions [16,25,28,29,30].

In addition, recent measurement-based studies confirm that methane slip strongly depends on engine type and operating conditions, especially engine load and operating mode [18,27,31,32,33]. Studies based on real onboard measurements show that CH₄ emissions can differ considerably from standard emission factors and that auxiliary engines and generators may make an important contribution to total methane emissions [19,31,32,33].

In this context, the level of detail of the data used in emission analysis is very important. The Tier I approach is based on generic, globally applicable emission factors, while Tier II uses more specific emission factors adjusted to certain fuel types, technologies, or operating conditions. In contrast, Tier III includes more detailed approaches, which may involve advanced models or real operational data [34,35,36]. For operation on LNG, where methane emissions strongly depend on actual operating conditions, approaches based on real operational data allow a more realistic assessment of the true climate impact than methods based on standard emission factors [18,19,32,33].

Despite the growing interest in LNG as a marine fuel, available review papers show that published data on methane slip from LNG marine engines are still limited, especially for newer engines and real operating conditions [18,37]. A similar lack of data was pointed out in the first primary emission measurements from an LNG carrier, where the need for additional measurement studies was also emphasized [19]. At the same time, assessments of the environmental impact of alternative marine fuels are often based on LCA/LCCA models and WtW analyses [15,16], scenario-based or simulation approaches [30,38] and average emission factors with assumed methane slip values [25], while real operational variations of a specific ship are not always well captured [38]. For this reason, studies that simultaneously include both CO₂ and CH₄ emissions, cover both main and auxiliary engines, express the results as CO₂-equivalent, and allow comparison with conventional fuels are especially important for a reliable assessment of the climate impact of LNG-fueled operation [19,25,30].

This paper analyzes the carbon footprint of an LNG tanker using real operational data collected by a continuous emission monitoring system during a ten-month period of regular vessel operation. The analysis includes CO₂ and CH₄ emissions from the main engines and diesel generators, their conversion into CO₂-equivalent using GWP₁₀₀ and GWP₂₀ factors, and the estimation of methane slip for different engine groups as well as for the ship as a whole. Special attention is given to the comparison between actual LNG emissions and a hypothetical HFO operating scenario. The aim is to assess to what extent the inclusion of methane emissions and the choice of GWP time horizon change the evaluation of the climate impact of operation on LNG compared with conventional marine fuels.

The novelty of this study is therefore the use of real operational emission data from an LNG tanker to quantify the contribution of methane slip to the ship’s carbon footprint, separately considering main engines and diesel generators and comparing the results under both GWP₁₀₀ and GWP₂₀ time horizons. Because the analysis is based on one LNG tanker and a ten-month observation period, the study is framed as an exploratory operational case study rather than as a fleet-wide assessment. Its purpose is to show how measured CO₂ and CH₄ emissions affect the carbon-footprint assessment of a specific ship under real operating conditions.

The research question guiding this study is: How does the inclusion of measured methane emissions from both main engines and auxiliary generators change the assessed carbon footprint of an LNG tanker compared with an assessment based on CO₂ emissions alone and with a hypothetical HFO reference scenario?

The remainder of the paper is organized as follows. Section 2 describes the case vessel, monitoring system, operational data, and calculation methods. Section 3 presents the emission results, methane slip variability, LNG–HFO comparison, and uncertainty assessment. Section 4 discusses the implications, operational relevance, and limitations of the results, while Section 5 summarizes the main conclusions and recommendations.

2. Materials and Methods

2.1. Case Vessel and Data Collection

The carbon footprint analysis was carried out using real operational data from an LNG tanker during regular vessel service. The ship observed in this study is a modern membrane-type LNG tanker equipped with two slow-speed two-stroke dual-fuel main engines (ME1 and ME2) and four diesel generator sets (DG1–DG4), which provide the electrical power needed for onboard systems. Dual-fuel engines can operate on liquefied natural gas (LNG) or on conventional liquid fuels, which is a common technical solution on modern LNG tankers [18,39]. During the observed period, the ship mainly operated on LNG, with pilot fuel used in the combustion process.

Emission data were collected from the Emission Monitoring System (EMSYS, Emsys Maritime Ltd., Altrincham, UK; Emsys-iS-OMU-02, version 2.1), installed on the observed LNG tanker [40]. EMSYS enables continuous monitoring, recording, and analysis of exhaust gas emissions under real ship operating conditions. The system uses an extractive measurement method with heated filter probes and heated sampling lines, while gas analysis is performed using a multi-channel quantum cascade laser (QCL) based on infrared absorption spectroscopy. The QCL sensor can be configured for multi-gas monitoring, including NO, NO₂, SO₂, CO₂, CO, CH₄, NH₃ and N₂O, and the reported repeatability and accuracy are ±2%. The system requires an annual calibration check and mandatory class/flag zero/span checks [40]. In this study, daily records of carbon dioxide (CO₂) and methane (CH₄) emissions were used for each main engine and diesel generator. This allowed emissions to be separated by individual source and their contribution to total CO₂-equivalent emissions to be analyzed, which is particularly important for the assessment of methane slip. Monitoring and reporting of ship emissions are also part of the wider regulatory framework in maritime transport, including the EU MRV and IMO DCS systems [5,41,42].

The analysis covered the period from September 2024 to June 2025. Daily CO₂ and CH₄ emission data were used for the calculations, while in some parts of the analysis they were summed to the monthly level in order to show trends and emission variability. Operational data, including engine running hours, cargo/maneuvering modes, and port hours, were taken from the ship’s noon reports and are presented in the Supplementary Materials (Table S1).

2.2. Calculation Procedure and Data Analysis

In this study, an approach corresponding to Tier III methodological detail was applied, because the analysis is based on real operational data and daily emission records obtained from a continuous emission monitoring system, rather than on standardized emission factors [34,35,36]. This approach allows a more realistic assessment of emissions under actual ship operating conditions, especially for LNG-fueled operation and dual-fuel engines, where methane emissions can differ significantly from standard values because they depend on engine load and operating conditions [18,19,32,33].

The total carbon footprint was expressed as CO₂-equivalent (CO₂-eq), which includes CO₂ and CH₄ emissions weighted by their global warming potential (GWP). For methane, the values GWP₁₀₀ = 28 and GWP₂₀ = 84 were used, according to the IPCC AR5 tabulated values for CH₄ without climate-carbon feedbacks [13]. The use of both time horizons allows the assessment of both the shorter-term and longer-term climate impact of methane emissions [14].

The total CO₂-equivalent was calculated as:

$$C{O}_{2}eq=C{O}_2+GWP·C{H}_4$$

(1)

where all quantities are expressed in the same mass units, that is, in tonnes in this study.

Methane slip was defined as the share of unburned methane relative to the total amount of methane entering the combustion process [18,19]. In this study, methane slip was not measured directly but was calculated from the measured CO₂ and CH₄ emissions using a stoichiometric approach. It was assumed that CO₂ emissions mainly result from the complete combustion of methane according to the following reaction:

$$C{H}_4+2O_2\to C{O}_2+2H_{2}O$$

(2)

Based on molar masses (16 g/mol for CH₄ and 44 g/mol for CO₂), the mass of burned methane was determined from the measured CO₂ emissions as:

$$m_{C{H}_{4}burned}={\displaystyle \frac{16}{44}}\cdot m_{C{O}_2}$$

(3)

The total amount of methane entering the combustion process was then determined as the sum of burned methane and emitted methane:

$$m_{C{H}_{4}total}=m_{C{H}_{4}burned}+m_{C{H}_{4}emitted}$$

(4)

where $m_{C{H}_{4}emitted}$ is the directly measured CH₄ emission. Methane slip was then calculated as:

$$Sli{p}_{C{H}_4}={\displaystyle \frac{m_{C{H}_{4}emitted}}{m_{C{H}_{4}total}}}\cdot 100\%$$

(5)

This calculation is based on the carbon mass balance and gives the methane slip percentage in conditions where direct data on LNG fuel mass flow to individual engines are not available. The methane slip values calculated in this way should therefore be interpreted as operational estimates based on measured CO₂ and CH₄ mass emissions, rather than as direct measurements of the fraction of methane entering each engine. The approach assumes that the measured CO₂ originates predominantly from methane combustion and that LNG can be represented by its dominant component, methane, for the purpose of the stoichiometric calculation. Since dual-fuel engines also use pilot fuel, a small part of the measured CO₂ may originate from liquid fuel combustion, which can affect the estimated amount of burned methane and, consequently, the calculated slip percentage. These assumptions should be considered when interpreting the exact methane slip values.

To compare the climate impact of operation on LNG, a hypothetical reference scenario of ship operation on heavy fuel oil (HFO) was defined. HFO was selected because it represents a conventional fossil marine fuel commonly used as a reference in comparisons with LNG operation. The scenario was based on the same observed operational period and the same reported fuel-use basis, so that the comparison reflects the actual activity level of the vessel rather than an arbitrary operating profile. In this scenario, CO₂ emissions were calculated using the standard HFO emission factor, $E{F}_{HFO}=3.114\mathrm{t}\mathrm{C}{\mathrm{O}}_2/\mathrm{t}\mathrm{f}\mathrm{u}\mathrm{e}\mathrm{l}$ [34,43]. The total fuel amount in the HFO scenario was taken from the ship’s operational data as the sum of reported fuel consumption and the fuel-oil-equivalent (FOE) amount corresponding to the consumed boil-off gas (BOG), as presented in the Supplementary Materials (Table S2). Since the purpose of this scenario was to compare the direct climate impact of LNG and HFO operation in terms of CO₂-equivalent, CH₄ emissions in the HFO scenario were considered negligible, and therefore: $C{O}_{2eqHFO}=C{O}_{2HFO}$. The HFO scenario was not intended to reconstruct an actual alternative voyage of the same vessel under HFO operation or to represent a full engine-performance simulation; rather, it provides a transparent conventional-fuel reference for assessing whether the apparent CO₂ benefit of LNG operation remains after measured methane emissions are included in the CO₂-equivalent assessment.

The data analysis included the calculation of total CO₂, CH₄, and CO₂-equivalent emissions, the distribution of emissions by engine, the calculation of specific emission rates (t/h), the calculation of methane slip for the main engines, diesel generators, and the ship as a whole, as well as the analysis of monthly trends. To assess the relationship between methane slip and operational parameters, Pearson’s correlation coefficient was used. In particular, the relationships between methane slip and cargo hours, maneuvering hours, and port hours were analyzed at the monthly level (Supplementary Materials, Tables S3 and S4). All calculations and graphical presentations were prepared in Microsoft Excel (Version 2605, Microsoft Corpora-tion, Redmond, WA, USA).

3. Results

Emission data obtained from the EMSYS [40] were analyzed for the period from September 2024 to June 2025, covering ten months of continuous operation of the LNG tanker. During the observed period, the main engines recorded 6576 operating hours (ME1) and 6572 operating hours (ME2), while the diesel generators recorded between 2985 and 5215 operating hours. The ship spent most of the observed period at sea, with 272 sea days, 11 port days and 8 anchored days, while the total time in cargo/maneuvering mode was 152 h. The highest number of cargo/maneuvering hours was recorded in September and October 2024, while no such activities were recorded in February and June 2025. Monthly operational data are presented in Supplementary Materials Table S1.

3.1. Total Emissions and Monthly Distribution

During the observed period, total emissions of 47,851.19 t CO₂ and 654.24 t CH₄ were recorded. When methane emissions are included in the carbon footprint calculation, the total emissions amount to 66,170.04 t CO₂-eq using GWP₁₀₀ and 102,807.74 t CO₂-eq using GWP₂₀. Monthly values of CO₂, CH₄, and the corresponding CO₂-equivalent are shown in

Table 1. Monthly CO₂ and CH₄ emissions and corresponding CO₂-equivalent emissions for the observed LNG operation.

Month	CO2 [t]	CH4 [t]	CO2eq (GWP100) [t]	CO2eq (GWP20) [t]
Sep 2024	5532.91	106.62	8518.27	14,488.99
Oct 2024	5617.80	144.59	9666.31	17,763.33
Nov 2024	5877.98	146.69	9985.22	18,199.71
Dec 2024	4234.38	52.03	5691.09	8604.52
Jan 2025	3232.11	27.02	3988.62	5501.64
Feb 2025	5771.04	40.33	6900.40	9159.11
Mar 2025	3461.91	33.69	4405.31	6292.12
Apr 2025	4542.68	35.51	5536.97	7525.54
May 2025	4058.63	35.34	5048.24	7027.46
Jun 2025	5521.74	32.42	6429.61	8245.33
Total	47,851.19	654.24	66,170.04	102,807.74

.

Methane contributed 27.7% of the total CO₂-equivalent when GWP₁₀₀ was used, and 53.5% when GWP₂₀ was used. In other words, although the mass of emitted CH₄ was much lower than the mass of emitted CO₂, its contribution to the total climate impact became very large because of its high global warming potential. The shares of CO₂ emissions and methane-derived contribution in total CO₂-equivalent are shown in Figure 1.

The monthly distribution of emissions shows that CO₂ emissions alone do not describe the total climate impact of operation on LNG well enough. In months with higher CH₄ emissions, the difference between direct CO₂ emissions and the corresponding CO₂-equivalent becomes much more pronounced. This is especially visible in September, October, and November 2024, when both the absolute CH₄ emissions and their share in total CO₂-equivalent were among the highest, while in February and June 2025 the relative contribution of methane was clearly lower.

3.2. Distribution of Emissions by Engine

The distribution of emissions by engine shows a clear difference between the contribution of the main engines and that of the auxiliary units. The main engines dominate total CO₂ emissions, with ME1 and ME2 together accounting for about 59.8% of the total CO₂-equivalent under GWP₁₀₀ and 55.6% under GWP₂₀. Diesel generators account for a smaller, but still significant, share of total emissions. When emissions are expressed as CO₂-equivalent, their relative contribution increases, especially under GWP₂₀, because of relatively high methane emissions. A detailed breakdown of emissions by individual engine is given in

Table 2. CO₂, CH₄ and CO₂-equivalent emissions by individual engines during the observed period.

Engine	CO2 [t]	CH4 [t]	CO2eq (GWP100) [t]	CO2eq (GWP20) [t]	Share in CO2eq (GWP100)	Share in CO2eq (GWP20)
ME1	15,104.17	162.79	19,662.33	28,778.65	29.7%	28.0%
ME2	15,640.64	151.68	19,887.68	28,381.76	30.1%	27.6%
DG1	2620.86	55.25	4167.93	7262.06	6.3%	7.1%
DG2	6119.47	115.43	9351.46	15,815.42	14.1%	15.4%
DG3	5719.94	96.27	8415.63	13,807.00	12.7%	13.4%
DG4	2646.10	72.82	4685.02	8762.85	7.1%	8.5%
Total	47,851.19	654.24	66,170.04	102,807.74	100.0%	100.0%

.

Total CO₂ and CH₄ emissions by individual engine are shown in Figure 2. The main engines had the highest absolute CO₂ emissions, which is expected given their role in ship propulsion. In contrast, the contribution of the diesel generators to CO₂ emissions was much lower. For CH₄ emissions, however, a different pattern can be seen. Although the main engines still had considerable absolute methane emissions, the share of the generators in total CH₄ emissions was much higher than in the case of CO₂. This is also visible in the relative shares shown in the figure, where the generators accounted for about half of total methane emissions. This distribution indicates that methane emissions were not linked only to the main propulsion system and that a substantial part also came from the auxiliary units.

When emissions are expressed as CO₂-equivalent, the relationship between the main engines and the generators changes. Although the main engines remain the dominant source of total emissions, the share of the auxiliary units increases because of their relatively larger contribution to CH₄ emissions. This effect is even stronger under GWP₂₀, where the generators account for a larger share of total CO₂-equivalent than under GWP₁₀₀. The distribution of CO₂-equivalent by engine type, with separate contributions from CO₂ and CH₄, is shown in Figure 3.

To compare the contribution of the main engines and the auxiliary units more clearly, emissions were also analyzed as specific emission rates per operating hour. The results show that the main engines had about twice the CO₂ emissions per operating hour (2.338 t/h) compared with the diesel generators (1.067 t/h), or about 2.2 times higher. Specific methane emissions per operating hour, however, were very similar for the two engine groups (0.024 t/h for the main engines and 0.021 t/h for the generators). When emissions are expressed as CO₂-equivalent, the difference between the main engines and the generators becomes much smaller. The ratio between emissions from the main engines and the generators was about 1.81 for GWP₁₀₀ and 1.53 for GWP₂₀, which is much lower than the ratio based on CO₂ emissions alone. Specific CO₂-equivalent emissions were 3.008 t/h for the main engines and 1.660 t/h for the generators under GWP₁₀₀ and 4.347 t/h and 2.846 t/h, respectively, under GWP₂₀. These results show that, although generators have much lower CO₂ emissions per operating hour, their relative contribution to total CO₂-equivalent becomes more important because of methane emissions. Detailed emission rate values are given in the Supplementary Materials (Table S5).

3.3. Methane Slip and Its Variability

To quantify the effect of unburned methane on the total carbon footprint of the ship, methane slip was calculated for the main engines, the auxiliary generators, and the ship as a whole. Total methane slip during the analyzed period was 3.62%, with 2.74% for the main engines and 5.18% for the diesel generators. The higher slip in the generators agrees with the distribution of CH₄ emissions by engine and with their relatively larger contribution to total CO₂-equivalent. The resulting values are shown in

Table 3. Monthly and total methane slip values for main engines, diesel generators, and the entire ship.

Month	ME Slip [%]	DG Slip [%]	Total Slip [%]
Sep 2024	4.89%	5.36%	5.03%
Oct 2024	6.44%	6.92%	6.61%
Nov 2024	6.28%	6.81%	6.42%
Dec 2024	1.30%	5.83%	3.27%
Jan 2025	0.39%	5.06%	2.25%
Feb 2025	0.68%	5.06%	1.89%
Mar 2025	0.59%	4.66%	2.61%
Apr 2025	0.36%	4.41%	2.10%
May 2025	0.39%	4.41%	2.34%
Jun 2025	0.79%	3.24%	1.59%
Total	2.74%	5.18%	3.62%

.

Monthly methane slip values showed strong variability. The highest total values were recorded in September, October, and November 2024 (5.03%, 6.61% and 6.42%), while the lowest values were recorded in June and February 2025 (1.59% and 1.89%). Methane slip from the main engines was particularly high during the first three observed months, after which it decreased sharply and remained below 1% during the rest of the period. In contrast, generator slip remained elevated throughout the whole period, mostly in the range from 3.24% to 6.92%, which points to a consistently higher relative loss of unburned methane in the auxiliary power system than in the main engines.

The analysis of methane emission shares showed that its contribution to total CO₂-equivalent changed considerably over the observed period. Relatively small changes in CO₂ emissions from one month to another were not necessarily followed by proportional changes in total CO₂-equivalent. For example, between October and November 2024, CO₂ emissions increased moderately, while a stronger increase in total CO₂-equivalent was recorded, indicating a significant influence of CH₄ emissions. The monthly share of methane in total CO₂-equivalent is shown in Figure 4.

Operational data for cargo/maneuvering modes suggest that months with more cargo/maneuvering hours were also among the months with higher methane slip and a larger relative contribution of methane to total CO₂-equivalent. For example, September and October 2024 had 54.0 h and 39.5 h of cargo/maneuvering activity, with total slip values of 5.03% and 6.61%, while February and June 2025 had 0.0 h of such activity and lower total slip values of 1.89% and 1.59% (Supplementary Materials Table S3). Correlation analysis showed a positive relationship between methane slip and cargo hours (r = 0.692, p < 0.05) and between methane slip and total port hours (r = 0.691, p < 0.05), while the relationship with maneuvering hours was weaker (r = 0.553, p ≈ 0.10) (Supplementary Materials Table S4). These results suggest that the ship’s operating mode probably contributes to methane slip variability, although it does not explain it fully.

3.4. Comparison of LNG and HFO Scenarios

A comparison between actual LNG emissions and a hypothetical HFO operating scenario was made on the basis of LNG CO₂-equivalent emissions shown in Table 1 and the HFO scenario defined in the Materials and Methods section. Monthly values for the HFO scenario and the relative differences compared with LNG CO₂-equivalent emissions are shown in

Table 4. Comparison of the hypothetical HFO scenario with LNG CO₂-equivalent emissions. Percentage differences were calculated relative to the HFO scenario using LNG CO₂-eq values under GWP₁₀₀ and GWP₂₀.

Month	HFO CO2eq [t]	ΔLNG GWP100 vs. HFO [%]	ΔLNG GWP20 vs. HFO [%]
Sep 2024	6486.46	+31.3	+123.4
Oct 2024	6916.19	+39.8	+156.8
Nov 2024	7212.02	+38.5	+152.4
Dec 2024	4926.35	+15.5	+74.7
Jan 2025	4496.62	−11.3	+22.4
Feb 2025	6729.35	+2.5	+36.1
Mar 2025	4331.57	+1.7	+45.3
Apr 2025	5184.81	+6.8	+45.1
May 2025	4699.03	+7.4	+49.6
Jun 2025	6480.23	−0.8	+27.2
Total	57,462.64	+15.2	+78.9

.

For the whole observed period, direct CO₂ emissions under operation on LNG were 16.7% lower than the emissions estimated for the HFO scenario. However, after including CH₄ and converting it into CO₂-equivalent, total emissions under LNG operation were 15.2% higher than in the HFO scenario when GWP₁₀₀ was used and 78.9% higher when GWP₂₀ was used. The monthly comparison shows that LNG CO₂-equivalent emissions under GWP₁₀₀ were lower than the HFO scenario only in January and June 2025, while under GWP₂₀ they were higher in all observed months. This indicates that the apparent CO₂ advantage of LNG operation is strongly affected by methane slip, especially when the shorter GWP₂₀ time horizon is applied. The monthly comparison of LNG and HFO scenarios, together with total values for the whole observed period, is shown in Figure 5.

These results show that including CH₄ significantly changes the comparison between LNG and HFO scenarios, especially when GWP₂₀ is applied.

To evaluate the influence of data uncertainty and methodological assumptions on the main conclusions, a sensitivity-based uncertainty assessment was performed (Supplementary Materials, Table S6). The tested cases included variations in measured CH₄ emissions, measured CO₂ emissions, and HFO reference emissions. The results showed that the exact numerical values of methane slip and CO₂-equivalent emissions vary with the assumptions, but the main conclusion remained unchanged within the tested ranges: methane emissions substantially increase the assessed climate impact of LNG operation, especially when GWP₂₀ is applied.

4. Discussion

4.1. Methane Slip and GWP Time Horizon

The results show that the climate impact of an LNG tanker cannot be assessed only from direct CO₂ emissions. For the observed vessel, direct CO₂ emissions were lower than those estimated for the hypothetical HFO scenario, but this advantage was not retained after measured CH₄ emissions were included in the CO₂-equivalent assessment. Total LNG emissions were 15.2% higher than the HFO scenario under GWP₁₀₀ and 78.9% higher under GWP₂₀. Therefore, the main interpretation is not that LNG is necessarily worse than HFO in all cases but that its climate performance depends strongly on methane slip and on the selected GWP time horizon.

This interpretation is consistent with recent literature, which questions the earlier and simpler view of LNG as automatically a “cleaner” fuel. Earlier studies on LNG as a marine fuel often emphasized its potential to reduce CO₂ emissions, sulfur compounds, and some other air pollutants and presented it as a promising transition solution in shipping [20,23,29,44]. However, more recent studies show that the overall climate benefit of LNG can be greatly reduced or even completely lost when unburned methane emissions are taken into account [25,28,30]. The results obtained in this study support these observations using real operational data.

In this context, the difference between the GWP₁₀₀ and GWP₂₀ time horizons is particularly important. Methane has a relatively short atmospheric lifetime but a very high global warming potential over a shorter time horizon, so its contribution to total CO₂-equivalent becomes much larger when GWP₂₀ is used. In this study, the contribution of CH₄ to total CO₂-equivalent was 27.7% under GWP₁₀₀, but as high as 53.5% under GWP₂₀. This shows that the choice of GWP horizon is not only a technical detail of the calculation but also a methodological decision that can significantly change the conclusion about the overall climate impact of operation on LNG [13,14].

4.2. Role of Auxiliary Generators and Operating Conditions

Besides the choice of GWP horizon, the distribution of emissions among the individual engines on board is also important for a proper assessment of the climate impact of LNG-fueled operation. The results show that the main engines and the diesel generators have different roles in the ship’s total climate impact. The main engines dominate CO₂ emissions, while CH₄ emissions are more evenly distributed between the main engines and the diesel generators. As a result, the generators, despite their smaller contribution to CO₂ emissions, still make a significant contribution to total CO₂-equivalent because of their relatively higher methane emissions.

This is also reflected in the methane slip values. Total methane slip in this study was 3.62%, with 2.74% for the main engines and 5.18% for the diesel generators. This total value is close to the average value of 3.8% reported in another LNG carrier case [19], where generator engines were also identified as an important source of methane emissions. Additional confirmation of the importance of the generators is given by the specific emission rates per operating hour. The main engines had about 2.2 times higher CO₂ emissions per operating hour than the generators, but CH₄ emissions per operating hour were very similar for both engine groups. Therefore, the relative importance of the generators increased when emissions were expressed as CO₂-equivalent. This confirms that auxiliary engines and generators should not be neglected in the climate assessment of LNG-fueled ships [19,37].

The monthly results indicate that operating conditions also affected methane slip variability. Months with more cargo/maneuvering activity and port-related operation were among the months with higher methane slip and higher methane contribution to CO₂-equivalent emissions. Correlation analysis showed a positive relationship between methane slip and cargo hours, while the relationship with maneuvering hours was somewhat weaker. Because the number of monthly observations was limited, these results should not be interpreted as proof of direct causality but rather as an indication that operating modes related to cargo operations and varying engine loads may contribute to methane slip variability. These interpretations are in line with studies showing that methane slip depends on engine type, engine load and actual operating conditions [18,19,31,32,37].

4.3. Value of Onboard Monitoring

The variability of methane slip described above confirms the importance of using a Tier III approach based on real operational emission data rather than only on standardized emission factors. Standardized emission factors, although useful for wider inventories and preliminary assessments, cannot always capture the variability of the real operation of a specific ship [34,35,36]. The results of this study clearly show that, without real CH₄ emission data, the total climate impact of the observed ship would not have been properly assessed. In this sense, the findings support the conclusions of studies that stress the need for real onboard measurements and more detailed operational data when assessing emissions from LNG ships [18,19,27,33,37]. This is especially important in the context of current regulatory changes, since EU ETS and MRV Maritime include CH₄ from 2026 and 2024, respectively. This makes methane emissions increasingly relevant not only for scientific assessment but also for the regulatory framework of maritime transport [4,5]. From an operational perspective, such monitoring can help identify engine groups and operating periods associated with higher methane contribution to CO₂-equivalent emissions and can support engine-load management, maintenance planning, and future methane slip mitigation measures.

4.4. Study Limitations

Several limitations of this study should be noted. Since the analysis was carried out on one ship and over one ten-month period, the study should be interpreted as an exploratory operational assessment rather than as a fleet-wide evaluation of LNG-fueled vessels. Therefore, the results cannot be directly extended to all LNG tankers, all types of LNG-fueled systems, or all operational profiles. Methane slip was not measured directly but was calculated from measured CO₂ and CH₄ emissions using a stoichiometric approach. In addition, daily data were used, which means that it was not possible to analyze short transient operating modes and instantaneous engine loads in detail. Even so, the results clearly show that methane slip has a strong effect on the assessment of the climate impact of operation on LNG and that ignoring it may lead to incorrect conclusions about the environmental performance of LNG.

5. Conclusions

This paper analyzed the carbon footprint of an LNG tanker using EMSYS data collected over a ten-month period, with special attention given to methane emissions and their contribution to total CO₂-equivalent. The results show that operation on LNG reduces direct CO₂ emissions but does not necessarily lead to lower total greenhouse gas emissions once methane slip is included in the analysis. In the observed case, total methane slip was 3.62%, with higher slip recorded for the diesel generators than for the main engines. This confirms the importance of including all onboard emission sources and not only the main engines. Methane emissions had a major effect on the total climate impact, especially when the GWP₂₀ factor was applied.

When CH₄ emissions were included and converted into CO₂-equivalent, the overall climate impact of operation on LNG in the observed case became comparable to, or even worse than, the HFO scenario, especially when GWP₂₀ was used. The results also show that an approach based on real operational data is of key importance, because standardized emission factors cannot always capture the variability of methane slip and the distribution of emissions among engines. Therefore, LNG may be considered a possible transition solution, but its climate advantage depends strongly on the level of methane slip, the ship’s operational profile, and the emission assessment methodology. Future studies should include a larger number of ships, longer observation periods, and more direct methane slip measurements.

From a practical and regulatory perspective, the results support the wider use of onboard monitoring systems that record both CO₂ and CH₄ emissions, especially for LNG-fueled ships. Future regulatory and technical assessments should account for methane slip explicitly and should distinguish between main and auxiliary engines where possible. For ship operators, particular attention should be given to operating conditions and engine groups associated with higher methane slip, since these can substantially affect the overall CO₂-equivalent performance of LNG operation.