Canada’s Black Carbon Inventory Report 2021

3 Black carbon inventory development

As mentioned in the introduction, the black carbon (BC) inventory is based on the Air Pollutant Emissions Inventory (APEI) (Environment and Climate Change Canada [ECCC], 2021). This chapter gives an overview of the development of the black carbon inventory. For more details on the air pollutant emissions inventory development, refer to Chapter 3 [PDF]of the APEI.

3.1 Black carbon as a fraction of particulate matter less than or equal to 2.5 microns in diameter

Two important assumptions underlie the present inventory: black carbon is predominantly emitted in particulate matter less than or equal to 2.5 microns in diameter (PM2.5), and only PM2.5 emissions resulting from combustion contain significant amounts of black carbon. Therefore, the basis for the black carbon inventory is the PM2.5 emitted from combustion processes, multiplied by the BC/PM2.5 fractions specific to each type of source. Although important in some cases, PM2.5 emissions from non-combustion sources, such as dust raised by traffic on paved and unpaved roads or by wind and machinery on open fields or mine sites, are not considered sources of black carbon.

For example, diesel engines have relatively high emission rates of PM2.5 per unit energy, and the fraction of black carbon in these PM2.5 emissions is also relatively high. The majority of diesel fuel in Canada is used for mobile sources, particularly in off-road applications. Other combustion sources with high PM2.5 emissions include solid fuel combustion units, such as coal- and wood-fired boilers and wood fireplaces. Industrial sources are generally equipped with highly effective PM2.5 controls on boiler emissions, with PM-control efficiencies often in the 90% range. This is reflected in the lower PM2.5 emissions compared with other sources. In contrast, the smaller and markedly different equipment used for residential wood combustion (fireplaces, wood stoves or furnaces) have poorer PM2.5 control efficiencies than larger units, notwithstanding the different types of fuel and firing practices used for burning firewood. Given their lower efficiency, combined with the lack of treatment of stack gases for many existing residential wood-burning devices, such devices are by far the largest source of combustion-related PM2.5 emissions in Canada. Nonetheless, black carbon emissions from residential wood burning are only slightly more than one third that of mobile sources due to a lower BC/PM2.5 fraction for wood devices than for diesel engines.

The dataset that breaks down the PM2.5 emitted from a particular source (e.g. diesel engine emissions) into its different components, including black carbon and organic carbon, is known as a speciation profile. Most speciation profiles contain a fraction for elemental carbon; these fractions are commonly used as a surrogate to quantify black carbon emissions. The current inventory relies primarily on the United States Environmental Protection Agency’s (U.S. EPA) SPECIATE database (U.S. EPA, 2014a) to calculate black carbon emissions from compiled combustion PM2.5 emissions. Several PM2.5 speciation profiles are specific to the combustion processes or technologies (e.g. appliance types for residential wood combustion), to the fuel type (e.g. diesel, gasoline, natural gas) or to the application (e.g. natural gas use for electrical power generation).

Where readily available, the PM2.5 emissions data from combustion were used directly with BC/PM2.5 fractions to estimate black carbon emissions. Annex 2 lists all BC/PM2.5 fractions used in this inventory. Separating combustion from non-combustion sources of PM2.5 remains a challenge in some cases because of a lack of data on activities (i.e. quantity of fuel burned) and on non-combustion sources (e.g. rock dust at a mine). In those cases, separating combustion PM2.5 from non-combustion PM2.5 is done on the basis of expert knowledge of the relevant activities prior to applying BC/PM2.5 fractions.

To estimate emissions from mobile sources, bottom-up approaches were adopted, i.e. applying fuel-specific emission factors to disaggregated activity data, such as vehicle or equipment data sorted by class, age or model year. In all cases, PM2.5 was estimated first, and BC/PM2.5 fractions were subsequently applied. The methods for estimating PM2.5 emissions from mobile sources are described in the APEI Report [PDF] (ECCC, 2021).

3.2 Use of facility reported emissions

Only PM2.5 emissions resulting from combustion contain significant amounts of black carbon. In the APEI, PM2.5 emission estimates are calculated using a variety of data sources, notably emission estimates reported by Canadian facilities to the National Pollutant Release Inventory (NPRI). For sources that are incompletely covered by PM2.5 estimates reported to the NPRI, PM2.5 emissions are calculated in-house using activity data, statistics and emission factors. For this inventory, emissions from Manufacturing, Electric Power Generation as well as Ore and Mineral Industries are estimated using facility data. Oil and Gas Industry estimates are based on facility-reported data used in combination with the results of independent studies (EC, 2014; ECCC, 2017; Quadram, 2019). Emissions due to agricultural, construction and residential (wood and other) fuel combustion are estimated from data on fuel consumption and combustion technologies. Commercial Fuel Combustion is estimated using a combination of facility-reported and other data sources.

Stack emissions of PM2.5 reported by facilities form the basis of black carbon estimates from facility-reported data. For each individual stack, the appropriate black carbon speciation factor (or factors) was applied to the combustion-related PM2.5 (Annex 2). The emissions are then summed at the facility level and aggregated to form the sectoral emission estimate.

3.3 Recalculations

As new data and methodologies become available, emission estimates from previous inventory editions are recalculated. Table 3–1 presents the main improvements to the estimation methodologies for this year’s inventory.

Table 3–1: Summary of methodological changes, refinement or improvements
Description Impact on emissions
Ore and Mineral Industries

Recalculations occurred in the Aluminium Industry and the Iron and Steel Industry sectors for years 2013 to 2018 as a result of better understanding of processes in these sectors, allowing for more accurate assignment of speciation factors.

The recalculations to the Aluminium Industry sector resulted in increases to the sector-specific emission totals for all years in the time series, ranging from 0.020 tonnes (0.06%) in 2016 to 2.4 tonnes (9%) in 2018.

The recalculations in the Iron and Steel Industry sector occured for all years of the time series, ranging from 27 tonnes (18%) in 2016 to 61 tonnes (30%) in 2018.

Oil and Gas Industry

In order to reflect the regional variability in gas composition, black carbon emissions from flaring in Alberta are estimated using recently developed natural gas composition data for the upstream oil and gas industry in Alberta by the Energy and Emissions Research Laboratory (EERL) of Carleton University (Tyner and Johnson, 2020). The EERL study uses measured gas composition data from approximately 400 000 wells in Alberta taken over a span of several decades across the province’s many oil and gas producing regions to generate gas compositions and higher heating values (HHV) by Alberta township. The township-level HHV data from the EERL study is used in conjunction with flared volumes extracted from the Petrinex (2020) reporting system and the empirical relationship between black carbon and HHV, derived in the Quadram (2019) study, to estimate black carbon emissions for the following upstream oil and gas sectors in Alberta: Natural Gas Production and Processing, Light/Medium Crude Oil Production, Heavy Crude Oil Production and In-situ Oil Sands Production.

These recalculations resulted in minor changes to emissions estimates for the oil and gas sectors, with increases in 2013 and 2014 and decreases from 2015 to 2018. A maximum increase of 1.8 tonnes (0.1%) occurred in 2014, and a maximum decrease of 10.5 tonnes (0.5%) occurred in 2015.
Manufacturing

Recalculations occurred in the Pulp and Paper Industry sector and Wood Products sector due to the inclusion of missing data from the previous submission.

Changes to Manufacturing are an increase of 36 tonnes (15%) in 2018.
Transportation and Mobile Equipment – Aviation

Recalculations occured in the aviation section due to updates to the aviation model. Data sources were updated to include new/current information. Also, aerodromes and aircrafts were further defined to include additional information. Finally, the emissions are now calculated by flight mode (taxi in/out, takeoff, climb-out, climb, cruise, descent and landing). In order to calculate emissions at this level of detail, some emission factors were adjusted to account for each mode.

The recalculations resulted in significant changes for the whole time series. The change will results in an apparent increase of 8.6 tonnes (4%) for 2013, and an apparent increase of 12 tonnes (5%) for 2018.

Transportation and Mobile Equipment – Marine

Recalculations occured because updated vessel activity data was incorporated into the marine model. The Marine Emissions Inventory Tool (MEIT) updated their 2015 model and produced data for the 2016, 2017, 2018 calendar years. Provincial estimates were redeveloped based on 2015, 2016, 2017 and 2018 port origin/destination pairs. Emissions associated with international navigation were removed from the report total in order to conform to the national total reported in the NFR tables.

The updated MEIT models resulted in significant changes from 2013 to 2018. The redevelopment port origin/destination pairs had a significant impact on provincial estimates for the whole time series. The change resulted in an apparent decrease of 3373 tonnes (68%) for 2013, and an apparent decrease of 1233 tonnes (44%) for 2018.

Commercial/Residential/Institutional – Home Firewood Burning

Recalculations occurred in the residential sector from home firewood burning. New firewood consumption data was developed based on data collected from the Statistics Canada Household and Environment Survey (Statistics Canada, 2017). This survey runs every other year, which allows for data coverage throughout the time series.

The recalculations resulted in a decrease of 4 kt from home firewood burning for each year of the time series.

3.4 Sources of uncertainty

A key source of uncertainty associated with black carbon inventories is the inconsistencies between definitions and measurements of black carbon (Bond et al., 2013). Scientists use different methods to measure black carbon particle emissions at the source and in the atmosphere, and therefore measured quantities are not strictly comparable.

Although not quantified, uncertainty in the black carbon estimates in this inventory stems partly from the uncertainty around the BC/PM2.5 fractions. There is large variability in the size of measurement samples used to derive these fractions; the same fractions can by default be applied to several different technologies. An example of the limitation of available BC/PM2.5 fractions can be seen with the application of the diesel BC/PM2.5 fraction for aviation turbo fuel in jet aircraft, as there is no available fraction specific to aviation turbo fuel. Similarly, a single BC/PM2.5 fraction is applied to all residential wood combustion appliances except wood furnaces (Annex 3, Table A3–1). The refinement of BC/PM2.5 fractions is dependent on new measurements. Assignment of fraction to sector or equipment type is made using engineering knowledge and judgment based on limited available information (such as stack names), with varying degrees of accuracy.

There is considerable uncertainty in determining the proportion of combustion PM2.5 emissions from industrial sources. The primary data source for estimating PM2.5 emissions from many industrial sources is the NPRI, in which emissions are reported by facilities by stack or as one aggregate value for the facility as a whole and are not broken down between combustion and non-combustion emissions. For some sectors (such as Aluminium, Pulp and Paper, and Cement and Concrete industries), it is assumed that the PM2.5 emissions are combustion-related when emissions of both CO and NOx are reported from the same stack; this assumption contributes to the overall uncertainty.

3.5 Considerations for future editions of this inventory

Future improvements will focus on expanding current coverage, as well as improving the accuracy of emission estimates, including the following:

Report a problem or mistake on this page
Please select all that apply:

Thank you for your help!

You will not receive a reply. For enquiries, contact us.

Date modified: