On behalf of the Swedish Environmental Protection Agency (Naturvårdsverket), SMED has updated and further developed an Excel tool that can be used by government agencies to calculate emission of carbon dioxide and climate gasesfrom business travel and other use of vehicles and machinery.

The purpose of this task was to meet the requirements of the Regulation (2009:907) on Environmental Management in Government Agencies, as well as to be used as a general tool to calculate the climate emissions of agency transportation/vehicle and machinery use. The tool in its early form included only carbon dioxide emissions, not other greenhouse gases, during transportation/use and did not account for related indirect emissions, such as emissions during the production of fuel or the vehicle. In 2018, the tool was further developed, including changes in categories and interfaces to better describe the current transportation situation. More climate gases were included (methane and nitrous oxide), and emission values to produce fuels were added. Additionally, the high-altitude effect for aviation was included in the climate impact value. From 2019 to 2022, a simpler update and minor development of the tool took place.

In the 2023 version of the tool, some new categories for emission calculations have been added, such as diesel trains in Europe, as well as the flight routes Stockholm-Visby and Stockholm-Umeå.

In the 2023 version of the tool, most emission values for road traffic are significantly lower (about 2-20 %) than in the 2022 version, reflecting continued electrification and an increased share of biofuels in diesel and gasoline. For taxis, values have decreased by 10-20% per kilometer, explained by an increased number of electric cars in the fleet. At the same time, emissions per unit of currency increase in Stockholm, attributed to lower prices in the city center.

For buses in public transportation, emissions per passenger-kilometer decreased significantly, on average for the whole of Sweden by 38 %. This reflects a rapid change towards increasing fossil-free fuels and electrification. In a few counties, emissions increase, especially in Gothenburg (+42 %), due to a return to fossil diesel and decreased use of HVO. For public transportation buses, emission values can change significantly between two years, reflecting rapid changes in bus fleets, fuel combinations, and occupancy rates in each region.

For rail traffic, emissions are generally lower than the previous year (about 20 %), but the differences vary by about +/- 60 %. Changes can be attributed to changes in energy use and occupancy, especially for trams, after the pandemic has subsided. For aviation, emissions decrease relatively significantly for most flight routes (0-40 %) based on data from ICAO's tool and previous calculations. This can be explained by higher occupancy after the pandemic. The exception is Malmö – Östersund, which increases significantly for unknown reasons.

SMED refrains from providing specific recommendations for the tool's future improvement and development as the Regulation (2009:907) on Environmental Management in Government Agencies is under review. SMED intends to return with suggestions on how the tool can be developed once any new regulation is decided. 

The updates made in this work are considered sufficient for the tool to maintain a generally good quality for emission calculations in the year 2023.

This manual describes the QA/QC procedures applied in the work with the preparation of the Swedish air pollutant and greenhouse gas emission inventories (hereinafter referred to as emission inventories) annually submitted to the UNFCCC, to the Paris agreement, to the EU Monitoring Mechanism Regulation, to the EU NEC directive and to the CLRTAP. The emission inventories cover historical data from 1990 as well as emission projections to 2050.

The manual is part of the Swedish National System, and is in accordance with the quality manual for the Swedish national inventory arrangements for inventory and reporting according to the Paris Agreement and decisions within the EU, and with SMED´s overall quality system. This manual is updated regularly to reflect changes in international reporting requirements and reporting guidelines. 

This document covers the activities performed by SMED (Swedish Environmental Emissions Data) and the collaboration between SMED and the Swedish Environmental Protection Agency (EPA). It briefly touches the QA performed within the Swedish national system for greenhouse gas inventories.

Sweden is applying a Tier III method for estimating changes in soil organic carbon (SOC) stocks for cropland remaining cropland on mineral soils within the Land Use, Land-Use Change and Forestry (LULUCF) sector. For that purpose, SLU is using the Introductory Carbon Balance Model (ICBM) for calculating annual SOC stock change rates for eight Swedish agricultural production (PO8) regions and aggregating results to the national level. The initial stocks of SOC are derived from SOC concentrations data in a national soil-monitoring program and pedo-transfer functions for computing dry soil bulk density. The model accounts for the effect of climate on SOC decomposition by using gridded daily weather records from the Swedish Meteorological and Hydrological Institute (SMHI). The estimates for annual carbon inputs to soil from crop residues and manures are based on yearly agricultural census data concerning crop types, yields, aboveground biomass and manure management. In the LULUCF sector, carbon input from manures are based on activity data from a series of reports published by Statistics Sweden (“Use of fertilizers and animal manure in agriculture”), Gödselmedelsundersökningen (GMU). The GMU is reporting the total amount of manure (wet weight) spread on arable land for different livestock categories and type of manure storage forms. The agricultural sector is using an animal number based (ANB) approach, which is calculating the amount of manure (excreted volatile solids) produced per animal type per year.

The objectives of this report were to compare the time series for manures with the two approaches, with the goal to explore the possibility of using the ANB approach also for the LULUCF sector. This was done by adapting the ANB approach and we examined possibilities for improving the time series with the GMU approach, for which we have observed some inconsistencies. We were conducting these calculations for each of the Swedish PO8s, results were also scaled up to the country level. Using a hypothesis that the dry matter content in manures were lower from mid 2000s because of more water from milk- and washing robots (i.e., which is entering the manures) provides an improved time series with the GMU approach, in the sense that there is at least an overall decreasing trend over time. This is in agreement with the trend using the ANB approach.

The variability in estimates of carbon from manures between years remain high for some of the data from the latest GMU reports, while the estimates using the ANB approach provides a smoother and more consistent time series. Both approaches are presenting various elements of uncertainty relating to the availability of precise activity data. It is feasible and would be advantageous adapting the ANB approach for use within LULUCF with the ICBM model, in order to harmonize the activity data used so that it can be the same as that used in the agricultural sector. However, this is still requiring some minor investigations. Especially regarding the amount of peat and sawdust used as bedding material, which needs to be added to the estimated amount of carbon from manures with the ANB approach. 

This report presents results from a methodological study on how to calculate and interpolate the time series of carbon stock changes in the fine litter and soil organic carbon pools on forest land remaining forest land and grassland remaining grassland for reporting of the LULUCF sector in the annual greenhouse gas inventory submitted to the UNFCCC and the EU.

Since rather small variations along the time series estimated by the present calculation and interpolation method results in huge national variations in sinks or sources of CO₂ at the national level, there has been a wish to find an alternative way of calculating the time series that reduces the effect of recalculations between submission years. The aim of the project was to evaluate alternative calculation methods for the time series for changes in the litter and soil carbon pools. The ambition is that the new method should result in time series that are as close as possible to the long-term trends in litter and soil carbon pool change by reducing the effect of random variation and interpolation methods within and between the submissions. The changes should be calculated so that the reporting is still representing carbon pool changes based on repeated measurements on the permanent plots of the Swedish Forest Soil Inventory.

Two different calculation methods were tested and compared with the method used up to submission 2022. The “average change method” estimates the average annual change for each plot over the whole time series instead of between each measured occasion. This results in a time series at plot level with no variations. However, when changes from all plots are averaged for each year in the time series some variation is introduced due to that plots may enter or leave the time series due to land-use change. For forest land remaining forest land, the method worked best and it significantly reduced variation along the time series and inter annual variation was also reduced compared to the previous method. The method was less effective on grassland remaining grassland where the large proportion of plots involved in land-use change contributed significantly to time series variation. The total aggregated carbon pool changes over the whole time series for this method is identical with the method that has been used up to submission 2022. 

The second method tested is based on regression of the carbon pool over the entire time series instead of estimating change on each inventory plot by repeated measurement. This method essentially leads to a total elimination of the variation over the reported time series but the difference between different submissions were more substantial. The aggregated amount of reported changes differ a lot from the other methods.

The conclusion from this project is to recommend the average change method to be used for calculating and interpolating the time series for fine litter and soil organic carbon for forest remaining forest and grassland remaining grassland.

This report summarizes the development work carried out in five different studies between 2017 and 2021 aiming to improve emission calculations for small-scale biomass combustion in the Other sector (CRF/NFR 1A4) for reporting to the UNFCCC and the CLRTAP. The most important measures include a more detailed and transparent structure of activity data as well as updated emission factors for different biofuels and heating technologies. Reports and documentation from earlier studies are attached.

Sweden’s emissions of greenhouse gases and air pollutants from stationary combustion, reported to among others UNFCC and CLRTAP, are in most cases calculated from fuel amounts and calorific values from Swedish energy statistics, and emission factors developed by SMED.

During autumn 2021 and spring 2022, several projects with the purpose to revise and update certain emission factors for stationary combustion have been conducted. The results on these projects (UF-27 Updates of certain emissions factors for stationary combustion, UF-31 Revision of certain emission factors for stationary combustion – stage 2 and parts of UF-07 Updates of emission factors for pulp and paper industry NFR 2H1) have been summarized in the current report.

The most significant effect of the revision of emission factors is seen for emissions of dioxin in NFR 1A1 and 1A2, where the annual average decrease would account to 65 % and 80%, respectively. Large effects are also seen for emissions of N₂O in CRF 1A4, where the average decrease was 35 %, and for NH₃ emissions in NFR 1A1 with an average increase of 10 %. Certain effects on the biogenic CO₂ emissions are also seen, with an average increase of 6 % to 9 % in NFR 1A1, 1A2 and 1A4. Emissions of SO₂ decrease on average with 3 % to 8 % for NFR 1A1, 1A2 and 1A4, while corresponding decrease for CO was between 4 % and 6% for the same NFR codes. Zink emissions in NFR 1A1 had an average decrease of 5 %. The effect on other emissions is relatively small, under ±5 % in average change of emissions. 

As a result of the projects UF-27, UF-31 and parts of UF-07 the following revisions are proposed:

• Emissions factor for biogenic CO₂ from combustion of black liquor from 96 kg/GJ is suggested to be 105 kg/GJ,
• Revisions of emission factors for CO₂ and wood fuels, pine and pitch oil and other biomass within stationary combustion (CRF/NFR 1A1, 1A3, and 1A4),
• Emission factor for dioxin from combustion of wood fuels, peat and other biomass is suggested to be revised,
• Emission factors for other fuel categories are suggested to be revised,
• All recommendations of revised emission factors with the reference of “SEPA 1995”,
• Revision of emission factors updated in the EMEP/EEA Guidebook 2019.