Accounting carbon emissions from electricity generation: a review and comparison of emission factor-based methods

Overview

Accurate estimation of greenhouse gas emissions is essential to meet carbon-neutrality targets, particularly through calculating direct CO2 emissions from electricity generation. The basic emission formula is E = activity data × emission factor; Tier 1 uses IPCC default factors, Tier 2 country-specific factors and Tier 3 plant/technology-specific data. Fuel emission factors used (IPCC default): fossil coal 94.60 tCO2/TJ, natural gas 56.10, fossil oil 77.7, derived gas 107.07; the Italian (ISPRA) derived-gas factor is 163.36 tCO2/TJ.

The methods produce very different totals: three simple formulations underestimate emissions by about 73% against the best estimates. North zone 2022: Method 5 (Tier 2, ISPRA factors, with CO2/C molecular-weight ratio) gave the highest estimate, about 57.5 Mt CO2; Method 4 was about 2% lower, Method 6 (Tier 3-type Carbon Monitor adjustment) almost 11% lower, and Methods 1-3 about 73% lower. Average yearly emission factor (AEF) for the North zone by Method 5 rose from 0.1627 tCO2/MWh in 2016 to 0.4709 in 2022, then fell to 0.3936 in 2023.

For 2018 in Northern Italy, the plant-level Tier 3 approach of Beltrami et al. gave 36,625,956 tCO2 and an AEF of 0.2840 tCO2/MWh, differing from Method 5 by 8% in emissions and 9% in AEF. 2023 AEF for the island zones depends strongly on the factor source: Sicily 0.5939 (Method 4) vs 0.4937 tCO2/MWh (Method 5), and Sardinia 0.9984 vs 0.7265, because of the derived-gas factor (IPCC 0.39 vs ISPRA 0.59 tCO2/MWh). Adding the carbon oxidation fraction (Method 3) made little difference, and a Tier 3 label did not guarantee higher accuracy: the Tier 3-type Method 6 underestimated relative to Tier 1/2 Methods 4-5.

The finding is that accuracy does not rank cleanly by tier, and country- or zone-specific factors matter most where fuels such as derived gases are significant; applied in the correct context, Tier 1 methods can also perform well. The paper focuses on generation-side (production-based) average emission factors and does not address marginal factors.

Key points

1. The basic emission formula is E = activity data × emission factor; Tier 1 uses IPCC defaults, Tier 2 country-specific factors, Tier 3 plant/technology-specific data.
2. Fuel emission factors (IPCC default): fossil coal 94.60 tCO2/TJ, natural gas 56.10, fossil oil 77.7, derived gas 107.07; the Italian (ISPRA) derived-gas factor is 163.36 tCO2/TJ.
3. North zone 2022: Method 5 gave the highest estimate (about 57.5 Mt CO2); Method 4 about 2% lower, Method 6 almost 11% lower, Methods 1-3 about 73% lower.
4. AEF for the North zone by Method 5 rose from 0.1627 tCO2/MWh in 2016 to 0.4709 in 2022, then fell to 0.3936 in 2023.
5. For 2018 in Northern Italy, the plant-level Tier 3 approach gave 36,625,956 tCO2 and an AEF of 0.2840 tCO2/MWh, differing from Method 5 by 8% in emissions and 9% in AEF.
6. 2023 AEF for the island zones depends strongly on the factor source: Sicily 0.5939 (Method 4) vs 0.4937 (Method 5), Sardinia 0.9984 vs 0.7265 tCO2/MWh; the cause is the derived-gas factor (IPCC 0.39 vs ISPRA 0.59 tCO2/MWh).
7. Adding the carbon oxidation fraction (Method 3) made little difference; a Tier 3 label did not guarantee higher accuracy.

Using it in design

• In data-centre carbon accounting, state the method (Tier 1, 2, 3) and factor source (IPCC or national) when choosing a grid emission factor; simple methods can underestimate generation emissions by about 73%.
• Use country- or zone-specific factors where derived gas and similar fuels have a significant share; in the island zones factor choice changed AEF by 20-37%.
• Note in the report that a production-based average factor does not represent marginal impact and that marginal approaches are needed for load-shifting or siting decisions.

Who should read it

Sustainability analysts choosing electricity emission-factor methods and doing data-centre Scope 2 accounting.

Document contents

1. 1 Introduction
2. 2 Review of emission-factor-based methods under the IPCC Tier 1/2/3 framework
3. 3 Comparison of methods: six methods, Italian market zones, hourly ENTSO-E data 2016-2023
4. 4 Results
5. 5 Discussion
6. 6 Conclusions

Related standards

• GHG Protocol Corporate Standard
• IPCC 2006 Guidelines for National Greenhouse Gas Inventories

Read the document

Publication details

Authors

Marina Bertolini; Pierdomenico Duttilo; Francesco Lisi

Publisher

Applied Energy (Elsevier BV)

Year

2025

Document number

arXiv:2411.13663

Volume

392

Pages

125992

Document language

English

Source type

Scientific

Access

Open licence — downloadable

DOI

10.1016/j.apenergy.2025.125992

Licence

CC BY 4.0 — share with attribution

Licence evidence

arxiv.org

Citation