Methods for Calculating Greenhouse Gas Emissions in the Baltic Sea Ports: A Comparative Study

Abstract

Ports are vital nodes of maritime transport. To be able to decrease their GHG emissions, ports have developed various automated or semiautomated tools for emission assessment. In this study, we focus on an open-source tool called EVISA and compare how seven Baltic Sea ports are using this tool. We found that the results of these assessments are incomparable, all the ports use the tool differently, and report different numbers of emissions. We also compare how one port, the Port of Tallinn, uses two different tools and ends up with different numbers of emissions. The study offers a detailed comparison of the port-specific methods, data collection processes, and calculation principles, evaluating their effectiveness in measuring emissions from maritime transport in ports. Additionally, it highlights the pressing need for standardised greenhouse gas emission mapping methodologies in ports. The results highlight the need to create a cohesive, easy-to-use tool that complies with established standards like the GHG Protocol, IPCC guidelines, and ISO 14064.

1. Introduction

Maritime shipping is responsible for over 80% of the world merchandise trade by volume. It is considered an economically and environmentally efficient mode of transport with its 3% of global greenhouse gas (GHG) emissions [1,2]. Despite its relatively low contribution compared to other sectors, GHG emissions from global shipping, expressed in CO₂e, including carbon dioxide (CO₂), methane (CH₄), and nitrous oxide (N₂O), rose to 1076 million tonnes in 2018, marking a 9.6% increase from 977 million tonnes in 2012 [3]. CO₂ emissions from the global shipping industry were estimated to be 856 million tonnes in 2022 [1,4]. The 2023 IMO Strategy on Reduction of GHG Emissions from Ships outlines milestones for achieving net-zero emissions from international shipping. It sets targets to reduce annual GHG emissions by at least 20% (with a target of 30%) by 2030 and by at least 70% (with a target of 80%) by 2040, both compared to 2008 levels and to reach net-zero GHG emissions by or around 2050 [5]. From a broader perspective, the IMO and EU share similar overarching goals for reducing GHG emissions from shipping, but EU regulations are considerably stricter at the regional level. In 2021, the European Commission introduced the Fit for 55 package of proposals, aiming to reduce the EU’s total GHG emissions by 55% by 2030 and achieve full decarbonisation by 2050 [6].

In a transportation chain, a port is a hub that connects land and sea transportation. Ports are important gateways for international trade since they are involved in the loading and unloading of goods and provide value-added services such as storage, warehousing, packaging, and the organisation of inland transportation [2,7]. Over 70% of the total weight of the EU’s external trade is transported by sea, accounting for approximately 50% of its total value [7,8]. According to the European Sea Ports Organisation’s (ESPO) 2023 Annual Environmental Report, the top five priorities for European ports remain unchanged from the previous year, ranked in descending order: climate change, air quality, energy efficiency, noise, and water quality [9].

Port-related GHG emissions are substantial, with the Fourth IMO GHG study highlighting that chemical and oil tankers—two of the six major ship types—account for over 20% of their total emissions from activities in or near ports and terminals [3]. Similarly to shipowners, ports and terminals face challenges when considering investing in new equipment or terminals regarding the best option for decarbonisation [1].

In this paper, we focus on calculating, namely assessing, GHG emissions in ports. In ports, there are various sources of emissions, such as ocean-going vessels (OGVs), cargo handling equipment (CHE), or trucks [10]. Many ports conduct annual emissions calculations to identify the sources, composition, and levels of air pollutants, helping to monitor initiatives aimed at improving air quality, reducing GHG emissions, and minimising health risks to stakeholders [2]. Accurate and detailed emissions data per emission source provide a better way to support decision making for emission reduction plans, track the progress of such plans, and provide actual views of current emissions production [10].

Ships are significant contributors to air emissions in ports. For example, in the study by Winnes, et.al. [11], it was found that more than half (53%) of the Port of Gothenburg’s CO₂ emissions come from ships at berth, making shipping emissions a critical component of the port’s GHG mapping efforts. When assessing emissions at ports, a wide range of methodologies and data are available, yet there is considerable overlap and frequent cross-referencing among them [12].

This paper begins with a literature review of existing methods for mapping GHG emissions in ports, showing the variety of systems available for emissions calculations. The review focuses on comparing the theoretical frameworks, calculation models, and practical aspects of these methods, highlighting key differences and the reasons behind them.

Thereafter, in Section 3, we focus on the Baltic Sea. The Baltic Sea is one of the largest brackish water bodies in the world [7] and is designated as an Emission Control Area (ECA) under Regulation 14 of MARPOL Annex VI. This regulation imposes strict limits on sulphur content in fuel oil used onboard ships in the Baltic Sea, requiring it to be 0.1% or less [3,13]. Within this region, the EVISA tool has emerged as a novel open-source tool for calculating port emissions. In Section 3, we compare how the EVISA tool has been used in various selected ports in the Baltic Sea region. The analysis aims to synthesise interpretations of the EVISA tool and discuss how different approaches—even when using the standardised EVISA tool—can lead to varying outcomes. This comparison is key to understanding the strengths and limitations of each approach in managing regional port emissions.

Finally, in Section 4, the paper focuses on Tallinn Port, comparing the results of EVISA mapping with a local collaborative mapping effort. This analysis highlights the advantages and limitations of local versus regional approaches, offering insights into how harmonised digital tools can be developed for port GHG mapping.

The specific research questions addressed are as follows:

• What are the current methods used for mapping GHG emissions in ports?
• What are the differences in these methods’ calculation tools, and consequences of these differences?
• How can these methodologies and best practices be harmonised across ports within the same region?

2. Methods for Mapping GHG Emissions in Ports

For the literature search, this study used the ScienceDirect and Taylor and Francis Online databases, where the search was limited to peer-reviewed journals and search terms combinations of “air emissions”, “maritime transport”, “maritime ports”, “GHG emissions”, “environmental impact”, “maritime seaports”, “sustainability”, “decarbonisation”, and other similar terms were used. In addition, several strategy or overview reports of the subject were also analysed (e.g., IMO, DNV, UNCTAD, ESPO reports). General reports were used to understand the current situation and retrieve statistical information, while scientific articles were used to gather information from different case studies and to understand research gaps or possible research topics.

Greenhouse gas (GHG) emissions from human activities are usually measured as carbon dioxide (CO₂) or CO₂ equivalent (CO₂eq). This is performed by comparing the warming effect of different gases using their Global Warming Potential (GWP) [14]. Each greenhouse gas affects the climate differently. The GWP index measures the warming effect of each gas relative to CO₂, making it possible to combine all greenhouse gases into a single metric. For example, over a 100-year period, methane has a warming effect 28 times greater than CO₂, while nitrous oxide’s impact is 265 times greater [15]. Research on ship emissions and marine pollution has seen a significant surge since 2019, aligning with the 72nd session of the International Maritime Organization’s (IMO) Marine Environment Protection Committee (MEPC) in 2018 [16]. During this session, the IMO established ambitious goals for 2050, targeting a reduction in ship emissions by over 50% and a 70% decrease in carbon dioxide emissions.

The GHG Protocol Standard is a comprehensive global framework designed to help companies and governments measure and manage GHG emissions, fostering transparency and consistency in addressing climate change [17]. It categorises emissions into three scopes: Scope 1 includes direct emissions from a company’s own operations; Scope 2 covers indirect emissions from the production of electricity and heat consumed by the company but generated elsewhere; and Scope 3 encompasses other indirect emissions related to the company’s activities, both upstream and downstream, such as emissions from the production of fuels used by the company [18]. In the case of ports, Scope 1 refers to direct emissions from port activities controlled by the port authority, such as cargo handling equipment and the port’s vehicle fleet, Scope 2 includes indirect emissions related to energy production outside the port area, such as purchased electricity or district heating, and Scope 3 covers emissions from the activities of port tenants, visitors, and customers, such as ocean-going vessels, rail transport, construction equipment, and employee commuting [16,19]. In addition, the GHG Protocol has recently raised an idea of Scope 4, which includes emissions that are avoided when a product replaces other goods or services, performing the same functions but with less carbon intensity and it also considers emissions avoided through the recycling of a product [17].

When estimating ship emissions, researchers typically use two methods: the top-down approach (also referred to as “up-bottom” by [16]) and the bottom-up approach. The top-down approach estimates GHG emissions from shipping by using fuel sales data and consumption-based emission factors; it approximates total fuel usage in the industry by aggregating bunker fuel sales data from multiple countries globally [3]. This method is recommended for cases where traffic data are limited [20]. For example, in 2011, a fuel-based approach was used to study domestic shipping in the Greek seas [21], and a similar method was applied in 2001 to assess exhaust emissions in the Turkish Straits, considering vessel characteristics, engine types, fuel use, voyage durations, and speeds [22]. However, previous studies [20,21,23] have pointed out the limitations of this approach, highlighting inaccuracies due to the lack of detailed spatial and temporal ship data. It assumes steady engine operation in open waters and lacks sufficient activity data, making it difficult to fully capture the complexity of maritime traffic.

The bottom-up approach generates emission estimates by using AIS-transmitted data that provide information on the operational activities of individual vessels [3,16]. This method calculates hourly fuel consumption and emissions for each vessel, using the IHS database to confirm ‘in-service’ status [3]. It demands numerous input parameters, encompassing detailed ship specifications (like type, engine features, and fuel type) and operational data (including travel distances, speed, tracking, and activity duration) [20]. Previous studies [12,16] highlight the widespread use of bottom-up methodologies and stress the importance of improving this approach for increased accuracy [16,24]. Importantly, ship operations near ports are divided into three main modes: manoeuvring, hotelling, and cruising (Figure 1). Analysing emissions across these modes provides a more accurate picture of GHG outputs. Research indicates that the hotelling mode typically generates the highest carbon emissions, accounting for the largest proportion of total emissions within the harbour environment [11,20].

Numerous studies have been conducted to assess emissions in various ports globally, and this number is rapidly growing.

Table 1. Reviews of different carbon emissions calculation methods used in ports.

Source	1 [12]	2 [25]	3 [16]	4 [26]
Concept	Literature that explores port-related ship emission sources	Methodologies to calculate carbon footprint in ports	Examine the port’s carbon emissions estimation methodologies	Analysis of different reviews and methodologies and case study in the Port of Valencia (data of 2016)
Nr of papers analysed	32	19	49	28
Nr of ports	80	21	~83	~14
Methodologies	Bottom-up, combination of bottom-up and top-down	More than 20 different methodologies are considered	Up-bottom, bottom-up, IPCC methodology, IPCC and WPCI methodology, ASIF	The Port of Valencia method considers the scopes based on GHG protocols, IPCC Guidelines [14], and ISO 14064 [27]
Bottom-up/top-down/both/other *	29/0/3/0	0/6/1/14	45/2/4/32	N/A
Years	2008–2021	2002–2017	2007–2020	2011–2023
Conclusion	The data on the amount and type of emissions in all examined papers were analysed but were not comparable even for the same ports	Each port uses its own method and there is no single or unified method to calculate carbon footprint in ports	In most cases, the estimations only cover some emission sources mentioned in standard guidelines, leading to underestimations; the lack of standardisation and harmonisation in adopted methodologies hinders comparisons and calibrations between different ports and countries	Recent studies in CF methodology are based on international norms, localised with the characteristics of a region, port, or terminal, and focused on the strategies that can reduce CF in ports and the primary criteria for developing new methodologies; most of the focus has been on Scope 3

* Approximate estimation.

presents several literature overviews of the emission assessment methods covering totally over 200 analysed ports. There is a clear shift towards the bottom-up approach for GHG emission assessments, as demonstrated by a synthesis of 32 papers that analysed shipping emissions in 80 ports between 2008 and 2021. The majority of ports have adopted the bottom-up method, with only three exceptions where the top-down approach is still in use. The analysis of conclusions revealed that emissions data across studies were inconsistent, even for the same ports, due to the lack of a standardised methodology for calculating carbon footprints in ports. This lack of standardisation, along with a focus on only certain emission sources, leads to underestimations and complicates comparisons between ports and countries, despite recent efforts to align methodologies with international norms and prioritise Scope 3 emissions.

In addition to the growing body of published research, there is an increasing number of tool developers entering the market, each using diverse methodologies. This variation is often due to their different backgrounds, such as IT companies (e.g., Nortal (Tallinn, Estonia), Port of Tallinn (Tallinn, Estonia), Awake.AI (Turku, Finland), CloudyBoss UAB (Klaipeda, Lithuania), Witteveen+Bos (Deventer, The Netherlands) or academic institutions (e.g., TalTech (Tallinn, Estonia), University of Oulu (Oulu, Finland), Cyprus Marine and Maritime Institute (Larnaca, Cyprus), National University of Singapore (Singapore)). These tools help clients identify problem areas and opportunities to reduce environmental impact by combining Automatic Identification System (AIS) vessel movement data with unique vessel-specific insights.

3. GHG Tool Comparisons Among Baltic Sea Ports

3.1. Ports Under Study

In this study, we focus on seven ports in the Baltic Sea (Figure 2), each one of them using the same tool. The Port of Rauma, which is on the west coast of Finland, handled a total of 4.7 million tonnes of cargo by the end of 2023, an increase of more than 200,000 tonnes (4.7%) compared to the previous year, most commonly paper and timber. The port operates as a limited company and covers an area of 165 hectares, has two fairways, and 20 piers [28,29].

The Port of Pori, which is also located on the west coast of Finland, mainly handles dry and liquid bulk, with a total of about 3,6 million tonnes in 2021. Structurally, the Port of Pori includes three distinct areas: Mäntyluoto, Tahkoluoto deep-water port, and Tahkoluoto chemicals harbour, a total of 962 hectares. The port is situated approximately 20 km from the city and has one of the deepest draughts in Finland—15.3 m [30].

The Port of Mariehamn, on the Åland Islands, primarily operates as a passenger port, serving as the biggest passenger port on the Åland Islands, with over 1.2 million passengers travelling through annually. The port is divided into two sections: the passenger port and Klintkajen, dedicated to cargo activities, where the port handles a small part of cargo, including timber exports from 2023 [31].

The Port of Norrköping, which includes several terminal areas—Öhman Terminal, Pampus Terminal, Energy Harbour, and Heavy Crane Area—primarily handles forestry products, steel, grain, energy and petroleum products, containers, and project cargo from Swedish industries. Located about 130 km south of Stockholm, the port handles around 4.3 million tonnes of cargo annually [32].

The Port of Oxelösund, a private limited company, has a water depth of 16.5 m, an ice-free location on the east coast of Sweden, 100 km from Stockholm, and direct connections to both motorway and railway networks [33]. The port handled more than 4 million tonnes of cargo in 2023. Although the port primarily focuses on handling bulk goods for large vessels, it also offers integrated transportation solutions. Adjacent to the Port of Oxelösund is the Steel Harbour, owned by SSAB, which specialises in handling plate steel and plate coils from SSAB and the port manages the Steel Harbour under a long-term contract [34].

The Port of Tallinn operates two passenger ports (Old City Harbour and Saaremaa Harbour) and two cargo ports (Muuga Harbour and Paldiski South Harbour), covering a total area of 136 hectares with 66 berths and the port operates under the landlord business model. In terms of passenger volume, Tallinn Old City Harbour ranks third in the northern Baltic Sea region, following Stockholm and Helsinki, with 8 million passengers in 2023. Additionally, the port handled 13 million tonnes of cargo. The Port of Tallinn has set a goal to improve energy efficiency and achieve 90% renewable energy consumption, with a broader aim of reaching climate neutrality and maximising the use of renewable energy by 2050 [35].

The Freeport of Riga, with 122 berths across 19 kilometres, handles a wide range of cargo operations. It operates under the management framework established by the “Port Law” and the “Law on the Freeport of Riga” of the Republic of Latvia, with the Board of the Freeport of Riga serving as its highest decision-making body. In 2022, the port handled 23.5 million tonnes of cargo. Programmes aimed at reducing CO₂ emissions, developing green energy, introducing alternative fuels, and improving infrastructure to ensure the port’s environmental sustainability are key aspects of its strategy. Efforts are focused on achieving carbon neutrality through innovation and digitalisation [36].

Figure 2. Ports under study [37].

Figure 2. Ports under study [37].

3.2. EVISA Tool

The EVISA 4.1 Carbon_footprint tool was developed in the University of Oulu between 2020 and 2022 to calculate greenhouse gas emissions of ports. It is open-source and widely used in ports’ self-evaluation to create strategic energy management plans, evaluate their carbon footprint, and identify effective methods to reduce environmental impacts. In accordance with the GHG Protocol Standard, the EVISA tool categorises the port’s greenhouse gas inventory into three scopes, each comprising various subcategories (Figure 3). Scope 1 refers to direct emissions from all operations within the port area that are directly controlled by the port authority. This includes harbour craft, cargo handling equipment owned or leased by the port, the port’s internal vehicle fleet, and stationary energy sources that all contribute to the GHG emissions which is illustrated with arrow on Figure 3. Scope 2 covers indirect, energy-related emissions, including emissions generated from energy production outside the port, such as purchased electricity, district heating, and water consumption under the port authority’s control. Scope 3 encompasses all greenhouse gas emissions associated with the activities of port tenants, visitors, and customers within the port’s boundaries, both on land and water. This includes emissions from visiting ocean-going vessels, rail transport, heavy-duty vehicles entering the port, construction equipment, and employee commuting. EVISA tool calculates port’s carbon footprint in CO₂e by taking into account greenhouse gases CO₂, CH₄, and N₂O emissions [19].

The EVISA self-assessment tool is basically an Excel worksheet that is divided into multiple sheets, with each sheet specifically dedicated to an individual category, e.g., energy, ocean-going vessels, harbour crafts, cargo handling equipment, on-road heavy-duty vehicles and trucks, rail transport, stationary energy sources, passenger vehicles and commuting, and construction equipment [38]. In the Excel file, cells featuring a white background contain results or supplementary information and should remain unchanged, whereas data entry is only permitted in cells with a grey background [19]. The tool and instructions are publicly available to calculate the total carbon footprint for each port, whereas the data must have been collected and then entered to Excel manually [19]. The results, when all the available data by the port are provided, are provided in tonnes of CO₂e [38].

4. Results and Discussion

4.1. Comparison of the EVISA Tool Across the Studied Ports

Table 2. Emissions mapping for different emissions categories in case ports for Scope 1.

Emissions Category in Scope 1/Port		Port of Rauma	Port of Pori	Port of Mariehamn	Port of Norrköping	Port of Oxelösund	Port of Tallinn	Port of Riga
Domestic vessels/harbour crafts	Activity-based			x			x
	Based on fuel consumption	x		x
Cargo handling equipment	Based on fuel consumption	x	x		x
	Based on known emissions						x
Stationary energy sources	Based on fuel consumption			x		x
	Based on known emissions						x
Passenger vehicles and commuting	Based on travelled distance
	Based on fuel consumption	x	x	x	x	x
	Based on known emissions						x

,

Table 3. Emissions mapping for different emissions categories in case ports for Scope 2.

Emissions Category in Scope 2/Port		Port of Rauma	Port of Pori	Port of Mariehamn	Port of Norrköping	Port of Oxelösund	Port of Tallinn	Port of Riga
Energy	Electricity/district heat 100% renewable energy			x	x	−/x
	Electricity/district heat carbon-neutral electricity
	Electricity/district heat Finnish average	x/x	x/x	−/x		x		x/x
	Electricity/district heat						x/x
	Clean water	x	x	x		x
	Wastewater	x		x		x

and

Table 4. Emissions mapping for different emissions categories in case ports for Scope 3.

Emissions Category in Scope 3/Port		Port of Rauma	Port of Pori	Port of Mariehamn	Port of Norrköping	Port of Oxelösund	Port of Tallinn	Port of Riga
Energy	Electricity/district heat 100% renewable energy	x
	Electricity/district heat carbon-neutral electricity
	Electricity/district heat Finnish average	x/x
	Electricity/district heat						x/x
	Clean water	x	x
	Wastewater	x	x
Ocean-going vessels	Based on ship calls and vessel information			x	x	x		x
	Known emissions (ports’ own management system)	x	x				x
Domestic vessels/harbour crafts	Activity-based	x	x				x
	Based on fuel consumption
	Pilot boats (operated from port area)	x
Cargo handling equipment	Based on fuel consumption	x	x			x
	Based on known emissions						x
On-road heavy-duty vehicles and trucks	Activity-based—specific information available			x	x		x
	Activity-based—based on estimated shares (Year of manufacturing/emission standard (EURO))	x	x			x
Rail	Based on travelled distance (diesel train)		x		x	x
	Based on travelled distance (electric train)
	Based on known emissions						x
Stationary energy sources	Based on fuel consumption	x
	Based on known emissions						x
Passenger vehicles and commuting	Based on travelled distance	x		x
	Based on known emissions						x
Construction equipment	Based on fuel consumption	x			x

show that every one of the ports used the emission calculation tool differently. The “x” in the table shows that this cell has been filled with the data for each port. “−/x” show that there are several possibilities and the port has provided data for one of them or is using one of the options—the one that is marked with x. If there is “x/x” then port has filled table with data for both options. If the cell is left empty, then there has been no data provided. The first differences in GHG emissions mapping across the seven ports stem from varying levels of data completeness across different scopes and emission categories. For Scope 1 emissions, most ports have reasonably comprehensive mapping, especially in categories like fuel consumption and direct energy use (Table 2). In addition, not all ports reported port vehicle and vessels commuting. These discrepancies suggest that some ports may not consistently track all direct emissions from on-site operations. In addition, not all ports were able to obtain complete Scope 1 data for emissions calculations.

All ports have largely mapped most Scope 2 emission categories, including electricity and heat consumption (Table 3). Some ports provide detailed data on electricity from renewable or carbon-neutral sources, while others either lack this differentiation or have not mapped it at all. Reporting on district heating emissions also varies, with some ports offering detailed breakdowns and others omitting this category. The biggest differences arise in the mapping of clean and wastewater, as more than half of the ports did not report data for this category.

The greatest variation in emissions mapping occurs in Scope 3 (Table 4), due to indirect emissions related to activities like energy use, vessel traffic, cargo handling, on-road heavy vehicles, rail, waste management, water usage, and construction. Many ports lack comprehensive data in these areas, with significant gaps in water, waste, vessel calls, and commuting emissions. These inconsistencies highlight the challenge ports face in fully capturing the broader impact of their operations, particularly for emissions outside their direct control.

In addition to the above differences, the ports used various years for their statistics: four ports used data from 2022, while others used data from 2019, 2021, and 2023. Several ports were also unable to provide energy data about their tenants. Although all ports provided the required data to calculate emissions from ocean-going vessels, inconsistencies were identified between the number of visiting vessels reported on their homepage statistics and the data entered into the EVISA tool. Four ports calculated emissions based on ship calls and vessel information, while three used their own management systems for these calculations.

In addition, the ports applied different emission coefficients, which may not be based on the same criteria or methodologies. Moreover, while different ports apply the same emission calculation coefficients, the energy production often occurs in entirely different facilities or, in some cases, even in different countries. For instance, the Port of Oxelösund used Finland’s 2021 emission factor to calculate Scope 2 electricity consumption emissions for 2022. Similarly, the Port of Mariehamn applied Finland’s 2020 emission factor for calculating district heat emissions (Scope 2) in 2022.

Different ports employed varying levels of precision when mapping emissions across categories. For instance, the Port of Tallinn tracked each individual vehicle entering the port, applying vehicle-specific emission coefficients to calculate overall emissions. In contrast, many other ports relied on estimated shares of different vehicle types.

Emissions from visiting vessels were also calculated using significantly different methods, making comparisons difficult. The Port of Tallinn, for example, used existing data infrastructures (EMSA and AIS) to automatically compute emissions for specific ships. In contrast, other ports used their own emission calculation systems based on vessel characteristics, though many of these models relied on outdated emission coefficients.

Minor differences also arose from the inclusion of emissions from rail transport in some ports. Additionally, if a port was undergoing development, emissions from construction equipment (e.g., Port of Rauma and Port of Norrköping) could provide valuable insights but may introduce discrepancies when compared to other ports. For instance, passenger ports, such as the Port of Mariehamn, may lack cargo handling equipment, whereas cargo ports typically rely heavily on such equipment. These variations in methodology, data availability, and port-specific factors make comparing CO₂ emissions across ports highly uncertain (as shown in

Table 5. Comparison of results when using EVISA tool.

Port/Scope	Scope 1, t CO2e	Scope 2, t CO2e	Scope 3, t CO2e	Total, t CO2e	Year
Rauma	52.16	234.08	14,678.67	14,964.91	2021
Pori	233.97	510.82	5705.92	6450.71	2022
Mariehamn	13.74	78.58	14,081.43	14,173.75	2022
Norrköping	898.78	0	17,926.46	18,825.24	2022
Oxelösund	40.84	1018.29	33,559.28	34,618.41	2022
Tallinn	561.61	8650.77	100,832.11	110,044.49	2019
Riga	0	818.17	38,753.77	39,571.94	2023

).

4.2. Comparison of the GHG Mapping Methods: Port of Tallinn vs. EVISA

The final comparison focused on GHG mapping at the Port of Tallinn using the EVISA tool and the port’s own tool. The latter web application was specifically developed to systematically collect GHG emission data from the Port of Tallinn and its operators. These data are then aggregated and visualised on the Port of Tallinn website using Microsoft Power BI version 2.138.782.0 [39]. In this case, the differences are not related to scope or level of detail, as both tools use the same datasets. Instead, the variations arise from how each tool aggregates different emission sources. Since 2019, the Port of Tallinn (PoT) has been tracking GHG emissions based on the ownership or control of pollution sources, categorising them into three scopes. Scope 1 covers direct emissions from the port’s own operations, including ships, vehicles, and boiler houses. Scope 2 includes indirect emissions from purchased electricity and thermal energy used for facilities and infrastructure. Scope 3 accounts for additional indirect emissions from activities such as tenant operations, visiting vessels, transit traffic, and cargo handling [35]. In recent years, the Port of Tallinn has enhanced its GHG emissions calculations by digitising processes and refining its methodology. This includes adopting updated guidelines from the Stockholm Environment Institute (SEI), tailored specifically to Estonian conditions [40], resulting in more accurate emissions estimates. However, these improvements have introduced discrepancies in the GHG data for 2019–2021 due to changes in the CO₂ calculation methods [35].

Finally, we made a separate study of the Port of Tallinn that used two different calculation methods, their own development, which we call here the PoT tool, and the EVISA tool. These methods differ in the scope of emissions they capture. For Scope 1, the PoT method focuses on heat, machinery, and vessels, while the EVISA method covers a slightly broader range by including all stationary energy sources. Scope 2 is the same for both methods, covering electricity and heat. In Scope 3, the EVISA method extends beyond the PoT method by also accounting for emissions from construction equipment and employee commuting [19] in addition to all the sources covered by the PoT method (

Table 6. Content of Scopes 1, 2, and 3 for different methods.

Scopes/Method	PoT Method	EVISA Method
Scope 1	Heat, machinery, vessels	Own car fleet, cargo handling equipment, harbour craft, stationary energy sources
Scope 2	Electricity, heat	Energy consumption
Scope 3	Electricity, heat, machinery, vessels	Ocean-going vessels, harbour craft, cargo handling equipment, heavy road traffic, rail transport, construction equipment, port operators’ energy consumption, stationary energy sources, employees’ commuting

).

In the PoT method, data collection from various sources, such as visiting vessels and vehicles operating in the port area, has been automated and digitised. In contrast, the EVISA tool requires manual data entry into an Excel sheet, which can reduce the level of detail and lead to potential information loss due to aggregation.

The PoT and EVISA tool were compared by the source of the data that are used in the calculation model (

Table 7. Comparison of two methods used in Port of Tallinn.

Sources for Scopes/Method	PoT Method	EVISA Tool Method
General approach	Bottom-up	Combination of bottom-up and top-down approach
Use	Only for PoT	General, open for everyone
Emissions factors considered in calculations	Yes	Yes
Emissions factors are updated automatically in calculations	Yes	No
On-road heavy-duty vehicles and trucks/vehicles and mobile devices emissions	Based on the Smart Port automatic number plate recognition of vehicles	Approach based on fuel consumption OR based on activity per mode (idling and running)
Passenger vehicles and commuting	Based on the Smart Port automatic number plate recognition of vehicles	Combination of distance-based method and average-data method
Electricity consumption emissions	Yes	Yes
(District) heating consumption emissions	Yes	Yes
(District) heating production emissions	Yes	No
Water consumption emissions	Yes	Yes
Wastewater treatment emissions	No	Yes
Ocean-going vessels data	Automatically (AIS and EMSA)	Manual
Ocean-going vessels emissions	Yes	Yes
Cargo handling equipment/stationary devices emissions (cranes, forklifts, etc.)	Yes	Yes
Harbour crafts (icebreakers, tugboats and similar)	Automatically (AIS and EMSA)	Yes
Train emissions included	Yes	Yes
Stationary energy sources (power plant, boiler, generator)	Yes	Yes
Construction equipment (dredges, paving equipment and similar)	No	Yes

). For ocean-going vessel emissions, both tools calculate emissions as the sum of manoeuvring and hotelling emissions. The EVISA tool incorporates emission factors based on the vessel’s fuel type and engine type to determine these emissions. These emission factors are sourced from the Fourth IMO GHG Study for CH₄ and the Third IMO GHG Study for N₂O and CO₂, ensuring accurate and consistent calculations [3,38,41]. The PoT method uses data from the Automatic Identification System (AIS) and the European Maritime Safety Agency (EMSA) database for reporting vessel GHG emissions. The AIS system provides information on a ship’s location and status (manoeuvring or hotelling), while the EMSA supplies emission factors specific to each state, ensuring accurate emission calculations [39]. Similar to the EVISA tool, the PoT also calculated harbour crafts separately from other visiting (ocean-going) vessels [38,39].

In the EVISA tool method, emissions calculations for on-road heavy-duty vehicles and trucks follow two approaches: one based on fuel consumption and the other on activity by mode, with the two modes being idling and running [38]. In contrast, the PoT method gathers data on individual vehicle movements from the Smart Port vehicle identification system and assesses emissions by retrieving vehicle-specific emission factors from existing databases using license plate information. The method then calculates emissions based on the known travel time and distance [39].

When comparing the emissions results between the EVISA and PoT tools across each scope, there are notable differences. In Scope 1, the PoT tool shows emissions that are roughly twice as high as those calculated by the EVISA tool (

Table 8. Comparison of emissions results of Port of Tallinn’s two methods.

Tool/Scope	Scope 1, t CO2e	Scope 2, t CO2e	Scope 3, t CO2e	Total, t CO2e	Year
EVISA	562	8651	100,832	110,044	2019
PoT tool	1199	1208	126,507	128,914	2021

). In the EVISA model, among all the criteria listed in Table 6, only data on harbour craft were used for emissions calculations. In contrast, the PoT model also includes heat and machinery in its calculations. For Scope 2, although the EVISA tool focuses solely on energy, its emissions values are more than seven times higher than those of the PoT tool. This discrepancy is likely because emissions from different energy sources, including electricity, heating, cooling, and water consumption, are accounted for in this step. Additionally, water consumption emissions were not accounted for in the EVISA tool due to missing data, and similarly, the PoT tool does not calculate emissions from water consumption. In Scope 3, the PoT tool reports slightly higher emissions than the EVISA tool, though the difference is less significant. Differences might occur due to the PoT tool using an activity-based approach, whereas EVISA relies on ship call-based calculations.

5. Conclusions

In this paper, we have studied greenhouse gas emissions calculation in seven Baltic Sea ports, and we found that the results of these assessments are incomparable, all the ports used the open-source tool differently, and report different numbers of emissions.

A wide variety of methodologies exist for carbon footprint calculation, with different ports relying on different approaches. While many of the existing methods share fundamental similarities—drawing from standards and guidelines such as the GHG Protocol, IPCC guidelines, or ISO 14064—port operations vary widely depending on activities and geographical location, leading to different countries and ports adopting distinct methods for estimating carbon emissions [16].

Harmonising GHG emission mapping methodologies is essential for enabling ports to effectively monitor their environmental impact and implement targeted reduction strategies. The research demonstrates that existing tools like EVISA and PoT offer valuable insights, but inconsistencies in data availability, emission categories, and calculation methods hinder meaningful comparisons. A standardised approach, incorporating clear guidelines on emission categories, minimum data detail requirements, and preferred methodologies, would promote greater alignment across ports. This would not only facilitate benchmarking and policy alignment but also support global efforts to reduce emissions from maritime transport. Comprehensive guidelines should not only specify the emission categories to be included but also detail how the mapping should be conducted, covering aspects such as the minimum required level of detail, preferred methodologies, and transitioning between different methods. Such standardisation is crucial for enabling meaningful comparisons of carbon emissions across ports.

This study highlights the pressing need for standardised greenhouse gas (GHG) emission mapping methodologies in ports, particularly within the Baltic Sea region. The comparative analysis of the EVISA emissions assessment tool and the Port of Tallinn (PoT) tool, along with the evaluation of EVISA’s application across various ports, highlights significant variations in data collection methods, calculation principles, and the scope of emissions considered across different ports. These differences, while reflective of port-specific operational and geographical factors, pose significant challenges to achieving consistency and comparability in GHG assessments. The findings emphasise the importance of developing a unified, user-friendly tool that adheres to established standards such as the GHG Protocol, IPCC guidelines, and ISO 14064. Baltic Sea area ports must collaborate to develop and implement standardised methods for measuring GHG emissions within port areas, leveraging international projects and private funding as potential resources.

By establishing a standardised approach, ports can enhance their accountability and transparency in GHG reporting, which in turn builds trust among stakeholders and aligns with broader sustainability goals. Such advancements would not only benefit the maritime sector but also serve as a model for other industries seeking to address climate change impacts through harmonised methodologies. Additionally, implementing these standardised methods could empower ports to better track their progress towards emission reduction targets and prioritise initiatives that yield the greatest environmental and economic benefits.

Finally, by identifying the strengths and limitations of current methodologies, this research provides a foundation for the development of a robust, practical framework for port emissions assessment. Such a framework is critical for advancing sustainable port management, guiding policy decisions, and contributing to the broader goals of climate change mitigation within the maritime sector.

This study has limitations regarding access to the methods and data utilised in the ports. It was not possible to verify all data sources and therefore we relied on the information provided by the port representatives.

At the time of writing this article, the PoT method utilised the previously established specific emission factors. Since ports typically update their emissions calculations in line with the renewal of these factors, this does not present any significant issues.

For further research, we recommend investigating the possibilities for developing unified solutions to calculate CO₂ emissions for all ports. The first step would be to map the needs of ports and find out common interests and possibilities regarding the methodology. Future studies can also expand the unified methodology to account for the unique characteristics of GHG emissions in dry port–seaport systems [42], including additional transportation emissions and the transfer of emissions from loading and unloading mechanisms outside urban boundaries.