Appendix F  
Development of Default Emissions Factors and Sensitivity Analyses

 

This appendix is divided into two primary sections. The first section outlines the Commission’s approach to generating default emissions factors for use in the Commission’s calculation of the emissions intensity estimates of covered steel products. The emissions factors calculated using this approach were used as default global and country-specific emissions factors and were multiplied by activity data (i.e., quantities of inputs into the production of steel) to generate upstream scope 3 emissions for reporting facilities.[575] The steps of the approach to develop these factors and the data sources used to do so are described.

The second section of this appendix presents the results of sensitivity analyses that explore the impact that modifications to the Commission’s methods, respondent population, and input parameters have on the overall emissions intensity estimates.

Development of Default Emissions Factors for Materials Used by Steel Facilities

As described in chapter 3, this investigation collected default emissions factors for use in calculating consuming facilities’ scope 3 emissions. This appendix details how default emissions factors were developed that cover materials within the steel system boundary using a methodology referred to in this report as the partial life cycle inventory (LCI) approach (“partial LCI approach”). The partial LCI approach used public source information and LCI analysis principles to create a database of (1) global emissions factors for upstream products used as material inputs by steel-producing facilities and (2) country- and production pathway-specific default emissions factors that capture distinctions in how international industries produce pig iron and steel. In particular, this approach sought to account for the following factors that drive differences between countries’ steel product emissions intensities:

·         Mixes of electricity generation sources: many processes in the steel production life cycle are electricity intensive, particularly steelmaking in electric arc furnaces.

·         Apparent efficiency of steel production processes.

·         Types of fuel used predominantly in each country’s steel industries.

·         Use rates of alloying materials in stainless steel production (specifically nickel and chromium).

In addition to providing a mechanism for capturing greenhouse gas (GHG) emissions intensity differences between countries and production pathways, the partial LCI approach also allowed for the development of default emissions factors that adhered as closely as possible to the methodologies and system boundary that governed the broader approach used to calculate each facility’s product-level emissions intensity values. In addition, the Commission’s calculation of its own default emissions factors for use in this investigation allowed for the publication of these emissions factors. These emissions factors are presented in appendix G.

Overview of Partial LCI Approach

For each product used as a material input ( $material$ ) in the steel system boundary, the goal of the partial LCI approach was to collect or calculate one or more default emissions factors ( $DefaultE{F}_{material}$ ) covering the amount of GHG emissions that occur in the production of one unit of that product from cradle (i.e., far-upstream production practices) to product gate (i.e., the end point of the production process for the product).579F[576] This term is designed to capture the inventory of direct emissions that occur in the manufacturing of that product as well as indirect emissions that occur in the generation of energy and the production of upstream material inputs (referred to throughout this section as “inputs”) used to produce that product.

For many of the nonsteel products included in the steel system boundary, the Commission calculated global emissions factors ( $DefaultE{F}_{material}$ ) using publicly available emissions factors (see step 1 below, which also contains of a list of these products). No country-specific emissions factors were calculated for these products.

The Commission also calculated country-specific emissions factors ( $DefaultE{F}_{material,country}$ ) for iron sinter, pig iron, and all steel products.580F[577] The methods for calculating country-specific emissions factors are described in greater detail in steps 2 and 3 and are summarized in equation F.1 below. Equation F.1 sums the emissions intensity of specific production processes used to directly produce $material$ as well as the emissions associated with the use of inputs (including both emissions embedded in upstream inputs and direct emissions that occur when using those inputs).

$$\begin{array}{c}DefaultE{F}_{material,country}=\\ UE{F}_{material,country}+{{\displaystyle \sum }}_{input}\left(Intensit{y}_{input,material}*\left(DefaultE{F}_{input,country}+DirectE{F}_{input}\right)\right)\left(F.1\right)\end{array}$$

For each product ( $material$ ):

·         $UE{F}_{material,country}$ refers to the unit process emissions factor for $material$, or the amount of emissions from fuel and energy consumption that is attributed to the discrete production process used to make that product in a particular country (the “unit process”). This term covers direct emissions from consumption of fuels during the unit process, as well as indirect emissions from the generation of energy used in the unit process.581F[578] $UE{F}_{material,country}$ is expressed as metric tons of carbon dioxide-equivalent (mt CO₂e) per metric ton of production of $material$.

·         $Intensit{y}_{input,material}$ refers to the rate at which a specific input ( $input$ ) is used in the unit process that produces $material$ (i.e., the “intensity” of the input in the product’s unit process). $Intensit{y}_{input,material}$ is expressed as the quantity of $input$ used (generally metric tons, except for gases, which are standard cubic feet), per metric ton of $material$ produced. $Intensit{y}_{input,material}$ is the same for all countries under this approach.

·         $DefaultE{F}_{input,country }$ refers to the default emissions factor of $input$. $DefaultE{F}_{input,country}$ is equivalent to the value of $DefaultE{F}_{material}$ for the input product itself, and is expressed as mt CO₂e per unit of input used. This term differs by country if the upstream input’s own default emissions factors was calculated at the country level; otherwise, that emissions factor was global.

·         $DirectE{F}_{input}$ refers to the amount of direct emissions that occurs from the use of $input$ in production of other products. For example, $DirectE{F}_{lstone}$ refers to the amount of emissions that occurs when a quantity of limestone is consumed in the production of another product.582F[579] $DirectE{F}_{input}$ is expressed as mt CO₂e per unit of input used. $DirectE{F}_{input}$ is the same for all countries under this approach.

The methods and sources for calculating $UE{F}_{material}$ and $Intensit{y}_{input,material}$ were adapted from parts of the methodology used in a 2023 study by the European Commission’s Joint Research Centre, Greenhouse Gas Emission Intensities of the Steel, Fertilisers, Aluminium and Cement Industries in the EU and its Main Trading Partners (JRC 2023). JRC 2023 estimated the emissions intensities of steel and other products made by the EU’s largest trading partners for these products.583F[580] In particular, the partial LCI approach uses JRC’s technique of dividing GHG emissions from fuel and energy consumption in the iron and steel sector among specific unit processes that correspond with product categories (see the description of step 2 below).584F[581] In addition, the partial LCI approach uses multiple data sources. These include:

·         Default and direct emissions factors for upstream inputs from the World Steel Association’s CO₂ Data Collection User Guide and Annex 5 of the ResponsibleSteel International Production Standard Version 2.1 (RS Standard 2.1).585F[582]

·         The International Energy Agency’s (IEA’s) Extended Energy Balances data, which include quantities of fuel and energy inputs and outputs within different industrial sectors including blast furnaces and the broader iron and steel industry.586F[583]

·         The IEA’s Emissions Factors 2023 data, which include emissions factors for electricity generation by country.587F[584]

·         Direct emissions factors for fuel combustion produced by the 2006 Intergovernmental Panel on Climate Change (IPCC) Guidelines for National Greenhouse Gas Inventories, Volume 2 (2006 IPCC Guidelines).588F[585]

·         Fuel, energy, and input intensity data produced by JRC, both in its 2023 report (JRC 2023) and in a 2013 JRC study that included surveys of the European steel industries (JRC 2013).589F[586]

·         Production data from the World Steel Association’s (worldsteel) Steel Statistical Yearbook (worldsteel Statistical Yearbook), the European Foundry Association, and EUROFORGE.590F[587]

The following sections, organized into steps, describe in greater detail how these data were used in equation F.1 to calculate default emissions factors. Step 1 describes how default emissions factors and direct emissions factors for certain materials were collected from public sources. Step 2 describes how $UE{F}_{material,country}$ was calculated for iron sinter, pig iron, and all steel products. Step 3 describes in practical terms how equation F.1 was used to calculate $DefaultE{F}_{material,country}$ for those products. The emissions factors themselves are available in appendix G.

Step 1 Emissions Factors Collected from Public Sources

The Commission collected two types of emissions factors directly from public sources for use in scope 3 analysis and in the partial LCI approach. These were:

·         $DirectE{F}_{input}$ collected from the World Steel Association’s CO₂ Data Collection User Guide for all inputs for which emissions occur from the use of those inputs.591F[588] These inputs include non-calcined limestone and dolomite, pig iron, direct reduced iron, ferroalloys, and carbon electrodes.592F[589] The Commission used all values for $DirectE{F}_{input}$ without additional modifications.

·         $DefaultE{F}_{material}$ (or $DefaultE{F}_{input}$ where such products are used as inputs in the calculation of a downstream product’s value for $DefaultE{F}_{material}$ ) collected for all products other than iron sinter, pig iron, and steel products.593F[590] The specific sources for each of these emissions factors are described below.

The Commission assigned $DefaultE{F}_{material}$ for most materials based on emissions factors from RS Standard 2.1. These include emissions factors assigned to oxygen, argon, nitrogen, hydrogen, iron ore, iron pellets, metallurgical coke, non-calcined limestone and dolomite, calcined lime, calcined dolime, aluminum metal, copper metal, nickel pig iron, ferromanganese, manganese, ferromolybdenum, molybdenum metal, ferrosilicon, silicon, ferrovanadium, silicomanganese, tin metal, carbon electrodes, and direct reduced iron.594F[591] The Commission selected RS Standard 2.1 as the source of most emissions factors for upstream materials for several reasons. First, this standard is one of only a small number of resources that publicly reports a comprehensive set of default emissions factors for most materials used by the steel sector. Second, RS Standard 2.1 reports emissions factors derived from data provided by CRU and Sphera Solutions, Inc. (Sphera), private organizations that are commonly used by U.S. companies to provide reputable sources of emissions factor data.595F[592] Third, reported emissions factors from RS Standard 2.1 are generally consistent with the broader methodology used by the Commission in this investigation. The emissions factors are based on cradle-to-gate processes within a system boundary that extends far upstream to processes such as mining.596F[593] The emissions factors are also expressed as mt CO₂e/unit of material used in steel production. Therefore, the Commission’s selection of emissions factors for upstream materials from RS Standard 2.1 satisfied many of the criteria for selecting default emissions factors described in appendix E (“II.D.2. Selection of Default Emissions Factors”).

The Commission made several modifications to the emissions factors from RS Standard 2.1 for use in its own investigation. RS Standard 2.1 increases certain default emissions factors by 20$–$60 percent over source data in order to encourage reporters under the standard to use primary data in lieu of default emissions factors.[594] Because that incentive does not apply to the Commission’s methodology, the Commission divided the RS Standard 2.1 default emissions factors with these add-ons applied by 1.2 or 1.6, depending on the material.[595]

The Commission calculated the default emissions factors for direct reduced iron ( $DefaultE{F}_{dri}$ ) using the natural gas- and coal-based emissions factors from RS Standard 2.1. Although direct reduced iron can be made using both fuel types, $DefaultE{F}_{dri,global}$ was derived directly from the natural gas-based emissions factors after the reduction of the add-on described above.599F[596] Although U.S. facilities producing covered steel products did not report external receipts of direct reduced iron from India, a separate term $DefaultE{F}_{dri,India}$ was calculated for use in calculating downstream values of $DefaultE{F}_{material,India}$ in step 3 using equation F.1. India is one of the world’s largest producers of direct reduced iron and is also unique in its widespread use of the coal-based method for producing this material.600F[597] The direct reduced iron emissions factors for India is the weighted average of coal- and natural gas-based direct reduced iron emissions factors from RS Standard 2.1 (with the add-on removed), weighted on the basis of 77.8 percent of India’s direct reduced iron production being coal based and 22.2 percent being natural gas based.601F[598]

Emissions factors for zinc, chromium, ferrochromium, nickel, and ferronickel were derived from other sources. The Commission selected the emissions factor for zinc from the World Steel Association’s 2020 Life Cycle Inventory (LCI) Study.602F[599] The Commission also sought emissions factors for ferronickel and ferrochromium that took into account assumed grades of these ferroalloys. For ferrochromium, the Commission used an estimate from a report published by the International Chromium Development Association (ICDA) that found that each kilogram of chromium metal in ferrochromium has embedded emissions of 10 kilograms of carbon dioxide equivalent (CO₂e).603F[600] For ferronickel, the Commission used an estimate from a 2023 report published by the Nickel Institute that found that each kilogram of nickel metal within ferronickel has embedded emissions of 45 kilograms of CO₂e.604F[601] Both of these reports were based on analysis provided to those institutions by Sphera.605F[602] The Commission calculated emissions factors for ferrochromium and ferronickel using these estimates and assumed chromium and nickel content for each of these ferroalloys of 53 percent and 30 percent, respectively.606F[603] The Commission also used the reported emissions factor for nickel metal from the 2023 Nickel Institute report.607F[604] The Commission used the calculated default emissions factor for ferrochromium for chromium metal as well, as no emissions factor for chromium metal was identified.

Step 2 Unit Process Emissions Calculations for Iron and Steel Products

The partial LCI approach uses an approach adapted from JRC 2023 to calculate unit process emissions factors ( $UE{F}_{material,country}$ ) for each iron sinter, pig iron, and steel production process in each country.608F[605] Unless otherwise stated, all equations and variables described in the discussion of step 2 include both country-specific and global derivatives without those being explicitly noted. $UE{F}_{material}$ (in mt CO₂e/mt) is calculated in equation F.2 using the total GHG emissions (in mt CO₂e) from fuel and energy use associated with the unit process that produces the material ( $UGH{G}_{material}$ ) and the total output of the material $\left(Outpu{t}_{material}\right)$.609F[606]

$$\begin{array}{c}UE{F}_{material}=\frac{UGH{G}_{material}}{Outpu{t}_{material}}\left(F.2\right)\end{array}$$

$UGH{G}_{material}$ is calculated using equation F.3, which divides total emissions from fuel and energy use within a country’s iron and steel sector proportionally across the sector.

$$UGH{G}_{material}={{\displaystyle \sum }}_{fuel}(GH{G}_{fuel}* Shar{e}_{fuel,material})+{{\displaystyle \sum }}_{energy}(GH{G}_{energy}* Shar{e}_{energy,material}) \left(F.3\right)$$

Total GHG emissions (in mt CO₂e) from the iron and steel sector’s consumption of a specific fuel type (i.e., the iron and steel sector’s direct emissions from use of that $fuel$ type, or $GH{G}_{fuel}$ ) and total GHG emissions from energy (electricity or purchased heat) consumed by the iron and steel sector (i.e., the iron and steel sector’s indirect emissions from use of that $energy$ type, or $GH{G}_{energy}$ ) are calculated using the methods described in step 2.1. The consumption of a fuel or energy type in the material’s unit process as a share of total consumption of that fuel or energy type ( $Shar{e}_{fuel,material}$ or $Shar{e}_{energy,material}$, respectively $)$ is calculated using the methods described in step 2.2.

Step 2.1 Calculating GHG Emissions for Each Fuel and Energy Type in the Iron and Steel Sector

In the partial LCI approach, equations F.4 and F.5 calculate $G{G}_{fuel}$ and $GH{G}_{energy}$ (in mt CO₂e/terajoules [TJ]) by multiplying the total quantity of fuel and energy use in blast furnaces and the iron and steel sector more broadly ( $IndustryUs{e}_{fuel}$, $IndustryUs{e}_{energy}$ ) by emissions factors ( $DirectE{F}_{fuel}, IndirectE{F}_{energy}$ ).

$$\begin{array}{c}GH{G}_{fuel}=IndustryUs{e}_{fuel}*DirectE{F}_{fuel}\left(F.4\right)\end{array}$$

$$\begin{array}{c}GH{G}_{energy}=IndustryUs{e}_{energy}*IndirectE{F}_{energy}\left(F.5\right)\end{array}$$

Step 2.1.1 Calculating Fuel and Energy Use in the Iron and Steel Sector

$IndustryUs{e}_{fuel}$ and $IndustryUs{e}_{energy}$ are based on data from the IEA Extended Energy Balances database, which quantifies (in TJ) fuel and energy use and generation by year, country, type of fuel or energy, and “flow.” Dozens of fuel types and two energy types$—$electricity and heat$—$are covered in this database.610F[607] A flow refers to a sector or activity that uses or generates fuel or energy. Certain flows are unique to the iron and steel sector and are used to calculate $IndustryUs{e}_{fuel}$ and $IndustryUs{e}_{energy}$ for the sector. These include:

·         Transformation Processes in Blast Furnaces: These data include fuel inputs used directly in blast furnaces as well as blast furnace gas (BFG) and basic oxygen furnace gas (BOFG) outputs. The fuel inputs reported within this flow include the portion of the feedstock coke and other fuels (e.g., coal, natural gas, and oil) used in blast furnaces that contribute calorific energy to BFG and BOFG.611F[608]

·         Energy Sector Own Use in Blast Furnaces: These data include fuel and energy inputs used primarily for auxiliary purposes to support blast furnaces.612F[609] Country reporting within this flow is uneven, and some of this fuel and energy use may be reflected in the other two flows listed here (Final Consumption in the Iron and Steel Sector or Transformation Processes in Blast Furnaces).613F[610]

·         Final Consumption in the Iron and Steel Sector: These data include fuel and energy inputs used in the iron and steel sector.614F[611] As noted above, data under this flow may include fuel and energy use in blast furnaces that is not covered under either of the two flows described above.615F[612]

To avoid double counting emissions, no fuel use or output data are used for the BFG or BOFG fuel types. This is the first of several steps taken to avoid double counting associated with the use of fuels to generate BFG and BOFG, which are themselves fuels used in the iron and steel sector (see box F.1).

 

 

Step 2.1.2 Selection of Direct and Indirect Emissions Factors for Fuel and Energy Types

Emissions factors for fuel ( $DirectE{F}_{fuel}$ ) and energy $\left(IndirectE{F}_{energy}\right)$ are from the following sources:

Direct emissions factors for fuel types: All direct emissions factors for fuels (not electricity or heat) are from the 2006 IPCC Guidelines.616F[613] The Commission used the same default emissions factors from this source for all countries.617F[614]

Indirect emissions factors for energy types: Indirect emissions factors for electricity are from the IEA Emissions Factors 2023 database.618F[615] These indirect emissions factors are country-specific measures of the amount of CO₂e (in grams) for each kilowatt-hour of electricity produced across each country’s mix of generation sources.619F[616] The Commission averaged indirect emissions factors from this database over the 2017$–$21 period.620F[617] The IEA calculated the indirect emissions factors for electricity using the same data sources used in the partial LCI approach to calculate sector-level emissions: the IEA World Energy Balances database (which provides information on fuel use and energy generation by country and flow) and the 2006 IPCC Guidelines.621F[618]

Indirect emissions factors for heat are derived using an adaptation of the equation that IEA uses to calculate emissions factors for electricity within the Emissions Factors 2023 database.622F[619] For each country and year, equation F.6 calculates the emissions factors for heat (in mt CO₂e/TJ) by summing across fuel types the product of total fuel use to generate heat $\left(Heatplant Us{e}_{fuel,heat}+CHP Us{e}_{fuel,heat}+Plant OwnUs{e}_{fuel,heat}\right)$ and a direct emissions factor for each fuel type $\left(DirectE{F}_{fuel}\right)$, and dividing by total heat output from all sources ( $Outpu{t}_{heat})$.623F[620] For more information on how fuel inputs are allocated in facilities with combined heat and power, see box F.2.

$$\begin{array}{c}IndirectE{F}_{heat}=\\ \frac{{{\displaystyle \sum }}_{fuel}\left((Heatplant Us{e}_{ful,heat}+CHP Us{e}_{fuel,heat}+Plant OwnUs{e}_{fuel,heat}\right)*DirectE{F}_{fuel})}{Outpu{t}_{heat}}\left(F.6\right)\end{array}$$

 

 

Step 2.1.3 Calculation and Aggregation of Energy- and Fuel-Specific Emissions from the Iron and Steel Sector

The Commission calculated $GH{G}_{energy}$ for both heat ( $GH{G}_{heat}$ ) and electricity ( $GH{G}_{electricity}$ ) using equation F.5. In addition, the Commission calculated $GH{G}_{fuel}$ for narrow categories of fuels that adhere to the specificity of IPCC emissions factors and data from the IEA Extended Energy Balances database. $GH{G}_{fuel}$ is then aggregated into broader fuel types corresponding with fuel intensity values discussed in step 2.2. These broader fuel types, and the specific fuels as classified under IPCC included within each, are based on a similar mapping provided by Joint Research Centre (JRC) 2023.624F[621] Specifically:

·         $GH{G}_{COG}$ solely includes the industry’s emissions from use of coke oven gas.

·         $GH{G}_{mcoke}$ covers the industry’s emissions from use of metallurgical coke, including lignite coke and coke oven coke (i.e., coke produced in coke ovens).

·         $GH{G}_{natgas}$ solely includes the industry’s emissions from use of natural gas.

·         $GH{G}_{coal}$ covers the industry’s emissions from use of coal, including anthracite, charcoal, coking coal, lignite, other bituminous coal, patent fuel, and sub-bituminous coal.625F[622]

·         $GH{G}_{oil}$ covers the industry’s emissions from use of oil products, including gas or diesel oil, naphtha, and residual fuel oil.

·         $GH{G}_{otherfuel}$ covers the industry’s emissions from use of all other fuels, including biodiesels, biogasoline, brown coal briquettes, crude oil, ethane, gas biomass, gas coke, gas works gas, industrial wastes, jet kerosene, liquified petroleum gases, lubricants, motor gasoline, municipal wastes (biomass fraction), municipal wastes (non-biomass fraction), natural gas liquids, other kerosene, other liquid biofuels, other petroleum products, other primary solid biomass, peat, petroleum coke, refinery gas, shale oil, and white spirit and special boiling point industrial spirits.

As described above in box F.1, the partial LCI approach does not calculate emissions for use of BFG or BOFG in the iron and steel sector.

Step 2.2: Calculating the Share of Fuel and Energy Used in Specific Unit Processes

The aggregated $GH{G}_{fuel}$ and $GH{G}_{energy}$ terms from step 2.1.3 are then allocated between unit processes associated with the production of individual materials using $Shar{e}_{fuel,material}$ and $Shar{e}_{energy,material}$ (equation F.7).626F[623] These terms cover use of a fuel or energy type in the material’s unit process as a share of total use of that fuel or energy type, by country.627F[624]

$$Shar{e}_{fuel,material}=\frac{Outpu{t}_{material}*Intensit{y}_{fuel, material}}{{{\displaystyle \sum }}_{material}\left(Outpu{t}_{material}*Intensit{y}_{fuel,material}\right)} \left(F.7\right)$$

$Outpu{t}_{material}$ refers to the sector’s total quantity of production for that material in metric tons.628F[625]

$Intensit{y}_{fuel,material}$ refers to the typical intensity (use rate) of that fuel in the unit process for the material, measured as gigajoules (GJ) of fuel used per metric ton of material output.629F[626] Data for each fuel type and material combination are presented in tables F.1 and F.2.630F[627]

In effect, equation F.7 determines the proportional fuel use for unit processes corresponding with materials, using two scaling factors: (1) the relative output of each material and (2) measures of the typical use of that fuel type in the production of each material. If a country produces a large quantity of a material or the material typically uses a lot of a specific fuel type in its associated unit process, then $Shar{e}_{fuel,material}$ will be higher.631F[628] Nonetheless, $Shar{e}_{fuel,material}$ may be overstated or understated for countries that use disproportionately high or low quantities of specific fuel types relative to use in other countries. For these countries, allocation of emissions associated with those specific fuel types to unit processes using the fuel intensity data from JRC 2023 (which is based primarily on production practices in Europe) is subject to higher levels of uncertainty.632F[629]

Table F.1 Fuel and energy intensities for unit processes that produce iron and semifinished steel products, by fuel and energy type and by product

In gigajoules per metric ton (gj/mt). BOF = basic oxygen furnace; DRI = direct reduced iron; EAF = electric arc furnace.

Product	Mapping with Joint Research Centre 2023 process	Metallurgical coke intensity	Coal intensity	Oil intensity	Electricity intensity	Heat intensity	Natural gas intensity	Coke oven gas intensity	Total fuel and energy intensity
Iron sinter	Sinter plant	1.28	0.00	0.00	0.16	0.00	0.02	0.02	1.51
Pig iron	Blast furnace	10.81	1.00	1.27	0.27	0.16	0.17	0.28	11.18
Carbon and alloy semifinished steel (BOF method)	BOF, continuous casting (carbon steel)	0.01	0.00	0.00	0.16	0.00	0.27	0.00	0.44
Carbon and alloy semifinished steel (EAF method)	EAF, continuous casting (carbon steel)	0.00	0.52	0.00	2.22	0.71	0.86	0.00	4.31
Stainless semifinished steel	EAF, continuous casting (high alloy steel)	0.00	0.52	0.00	2.22	0.71	0.86	0.00	4.31
DRI (natural gas method)	DRI-EAF (natural gas-based)	0.00	0.00	0.00	0.35	0.00	9.40	0.00	9.74
DRI (coal method)	DRI-EAF (coal-based)	0.00	23.93	0.00	0.26	0.00	0.00	0.00	24.17
Iron castings	Cast iron melting, foundry casting (iron)	3.42	0.00	0.00	0.36	5.84	0.00	0.00	9.62
Steel castings	Steel melting, foundry casting (steel)	0.00	0.00	0.00	2.34	0.00	10.49	0.00	12.83
Iron and steel forgings	Forging (carbon steel, iron)	0.00	0.00	0.00	2.06	0.00	27.03	0.00	29.09

Sources: USITC compiled from Vidovic et al., Greenhouse Gas Emission Intensities of the Steel, Fertilisers, Aluminium and Cement Industries in the EU and Its Main Trading Partners, September 18, 2023, 46$–$47.

Note: A fuel intensity value of 0 indicates that the Commission assumed$—$for purposes of calculating scope 3 emissions factors$—$that the unit process corresponding with the product specified in the row did not use the fuel specified in the column. “Total fuel and energy intensity” is measured as the total use of energy and fuel by each product’s unit process. Total use of fuel and energy includes use of all of the specified fuel and energy types in this table as well as blast furnace gas and basic oxygen furnace gas. The sum of all underlying intensity values may differ slightly from total fuel and energy intensity due to rounding. Total fuel and energy intensity is used as a proxy value for “other fuel intensity.” The partial life cycle inventory (LCI) approach used fuel and energy intensity values for DRI derived from Joint Research Centre (JRC) 2023 intensities that include both EAF and DRI processes combined, with EAF intensity data subtracted from the JRC data. The partial LCI approach used fuel and energy intensity values for semifinished steel unit processes that produce solid semifinished steel products. JRC 2023 separated the processes used to produce semifinished steel products into EAF and BOF operations that produced liquid steel in addition to a continuous casting process that produced solid semifinished steel products. JRC 2023 also assumes that 1.06 mt of liquid steel are used to produce solid semifinished steel products. JRC 2023 converted JRC 2023’s fuel and energy intensity data for these processes into consolidated unit processes for EAF and BOF production. In the partial LCI approach, JRC 2023 fuel and energy intensity data for EAFs and BOFs were multiplied by a factor of 1.06 and added to the intensity data for continuous casting.

Table F.2 Fuel and energy intensities for unit processes that produce finished steel mill products, by fuel and energy type and by product

In gigajoules per metric ton (gj/mt).

Product	Mapping with Joint Research Centre 2023 process	Electricity intensity	Natural gas intensity	Total fuel and energy intensity
Stainless hot-rolled flat steel	Hot rolled mill (high alloy steel)	0.35	2.33	2.68
Carbon and alloy hot-rolled flat steel	Hot rolled mill (carbon steel)	0.18	1.40	1.59
Cold-rolled flat steel (carbon and alloy, stainless)	Cold rolled mill, annealing (carbon and high alloy steel)	0.30	1.06	1.36
Carbon and alloy coated flat steel	Finishing flat products	0.10	1.07	1.17
Stainless hot-worked long steel	Bars and rods mills (high alloy steel)	0.86	3.49	4.35
Carbon and alloy hot-worked long steel	Bars and rods mills (carbon steel)	0.31	1.21	1.52
Cold-formed long steel (carbon and alloy, stainless)	Wire mill (carbon and high alloy steel)	0.53	0.00	0.53
Seamless tubular steel products (carbon and alloy, stainless)	Beams, billets, rails and tubes mills (carbon and high alloy steel), annealing	0.18	2.77	2.96
Non-seamless tubular steel products (carbon and alloy, stainless)	Cold rolled mill, annealing (carbon and high alloy steel)	0.30	1.06	1.36

Source: USITC compiled from Vidovic et al., Greenhouse Gas Emission Intensities of the Steel, Fertilisers, Aluminium and Cement Industries in the EU and Its Main Trading Partners, September 18, 2023, 47.

Note: A fuel intensity value of 0 indicates that the Commission assumed$—$for purposes of calculating scope 3 emissions factors$—$that the unit process corresponding with the product specified in the row did not use the fuel specified in the column. “Total fuel and energy intensity” is measured as the total use of energy and fuel by each product’s unit process. Total use of fuel and energy includes use of all of the specified fuel and energy types in this table as well as blast furnace gas and blast oxygen furnace gas. The sum of all underlying intensity values may differ slightly from total fuel and energy intensity due to rounding. Total fuel and energy intensity is used as a proxy for “other fuel intensity.” Where a product category is modified with “carbon and alloy, stainless,” the fuel and energy intensities apply to both the carbon and alloy steel and stainless steel types of that product.

Step 2.3 Final Calculation of Unit Process Emissions Factors for Iron Sinter Plants, Blast Furnaces, and Steel Production Processes

Step 2.1 generated the terms $GH{G}_{cog}$, $GH{G}_{mcoke}$, $GH{G}_{coal}$, $GH{G}_{oil}$, $GH{G}_{natgas}$, $GH{G}_{otherfuel}$, $GH{G}_{electricity}$, and $GH{G}_{heat}$ covering total emissions from use of each type of fuel by the iron and steel sector. Under equation F.3, these terms are multiplied by corresponding terms for $Shar{e}_{fuel,material}$ and $Shar{e}_{energy,material}$ and subsequently summed to calculate unit process emissions associated with fuel and energy use in the production of specific materials ( $UGH{G}_{material}$ ).

An additional step adjusts emissions associated with electricity use in $UGH{G}_{pig}$ (unit process emissions for pig iron production in blast furnaces). As described in box F.1, all emissions associated with BFG and BOFG generation and combustion are implicitly allocated to blast furnaces and BOFs. If a country’s integrated facilities self-generate electricity using the waste gases generated from blast furnaces and BOFs (BFG and BOFG), those facilities should not also have positive contributions to $UGH{G}_{pig}$ from indirect emissions from the national grid for that electricity. To correct for this, after $GH{G}_{electricity}$ is allocated to the pig iron unit process using $Shar{e}_{electricity,pig}$, $UGH{G}_{pig}$ is reduced by the amount of electricity that the iron and steel industry self-generated from BFG and BOFG multiplied by the indirect emissions factors for electricity.[630]

Using equation F.2, the partial LCI approach calculates unit process emissions factors ( $UE{F}_{material}$ ) for production of iron sinter, pig iron, and every steel product listed in tables F.1 and F.2. These unit process emissions factors are then incorporated into equation F.1, which generates country-specific emissions factors for these materials, as described in step 3.

Step 3 Material Flow Analysis and Calculation of Default Scope 3 Emissions Factors for Iron Sinter, Pig Iron, and Steel Products

As described in step 1, many of the furthest upstream materials included in the steel system boundary have assigned default emissions factors using publicly available data. However, for iron sinter, pig iron, and each steel product, equation F.1 is used to calculate country-specific emissions factors for that material.634F[631] For each material, $UE{F}_{material}$ is added to a sum of emissions associated with upstream inputs that are used to produce the material to calculate $DefaultE{F}_{material}$. The amount of emissions associated with upstream inputs is based on a combination of upstream emissions factors associated with those inputs ( $DirectE{F}_{input}$ and $DefaultE{F}_{input}$ ) and input intensity data ( $Intensit{y}_{input,material}$ ), which captures the rate at which inputs are used to make downstream materials.635F[632]

Step 3 describes how $DefaultE{F}_{material}$ is calculated for each material type and provides values for $Intensit{y}_{input,material}$ using material-specific derivatives of equation F.1. These calculations occur sequentially: for example, the default emissions factor for iron sinter ( $DefaultE{F}_{sinter}$ ) is used in the calculation of the default emissions factor for pig iron ( $DeaultE{F}_{pig}$ ), which in turn is used in the calculation of the default emissions factor for carbon and alloy semifinished steel. All values of $DirectE{F}_{input}$ and $DefaultE{F}_{input}$ other than those values of $DefaultE{F}_{input}$ generated in step 3 itself are from tables G.3 and G.4 in appendix G.636F[633]

Step 3.1 Default Emissions Factors for Iron Sinter

U.S. steel-producing facilities do not receive iron sinter from other sources. Therefore, use of iron sinter is not a source of scope 3 emissions when calculating the emissions intensity of U.S. steel products. However, iron sinter is an important input in pig iron production. Therefore, country-specific measures of $DefaultE{F}_{sinter}$ (equation F.8) are calculated that are subsequently used in the calculation of country-specific measures of $DefaultE{F}_{pig.}$

$$\begin{array}{c}DefaultE{F}_{sinter}=UE{F}_{sinter}+\left(Intensit{y}_{ore,sinter}*DefaultE{F}_{ore}\right)\\ +\left(Intensit{y}_{lstone,sinter}*\left(DefaultE{F}_{lstone}+DirectE{F}_{lstone}\right)\right)\\ + \left(Intensit{y}_{lime,sinter}*DefaultE{F}_{lime}\right)\left(F.8\right)\end{array}$$

$Intensit{y}_{ore,sinter}$ is set at 0.813 metric tons of iron ore; $Intensit{y}_{lstone,sinter}$ is set at 0.131 metric tons of non-calcined limestone and dolomite; and $Intensit{y}_{lime,sinter}$ is set at 0.010 metric tons of calcined lime.637F[634]

Step 3.2 Default Emissions Factors for Pig Iron

$DefaultE{F}_{pig}$ is calculated using equation F.9.

$$\begin{array}{c}DefaultE{F}_{pig}=UE{F}_{pig}+\left(Intensit{y}_{oxygen,pig}*DefaultE{F}_{oxygen}\right)\\ +\left(Intensit{y}_{ore,pig}*DefaultE{F}_{ore}\right)+\left(Intensit{y}_{sinter,pig}*DefaultE{F}_{sinter}\right)\\ +\left(Intensit{y}_{pellets,pig}*DefaultE{F}_{pellets}\right)+\left(Intensit{y}_{lime,pig}*DefaultE{F}_{lime}\right)\left(F.9\right)\end{array}$$

Values for $Intensit{y}_{input,pig}$ are shown for each input in table F.3.

 

Table F.3 Input intensity of inputs in the production of pig iron, by input category

In metric tons of inputs used per metric ton of material produced (mt input/mt material) for solid materials and thousand cubic feet per metric ton of material produced for gases (Mcf/mt material).

Input category	Variable name	Unit of measure	Input intensity value
Oxygen	$Intensit{y}_{oxygen,pig}$	Mcf/mt material	1.598
Iron ore	$Intensit{y}_{ore,pig}$	mt input/mt material	0.180
Iron sinter	$Intensit{y}_{sinter,pig}$	mt input/mt material	1.088
Iron pellets	$Intensit{y}_{pellets,pig}$	mt input/mt material	0.358
Calcined lime	$Intensit{y}_{lime,pig}$	mt input/mt material	0.026

Source: USITC compiled from Drnevich, Messina, and Selines, “Production and Use of Industrial Gases for Iron and Steelmaking,” 1998, 292; Remus et al., Best Available Techniques (BAT) Reference Document, January 24, 2013, 304.

Note: “Input intensity value” refers to the quantity of inputs used to produce one metric ton of pig iron.

Step 3.3 Default Emissions Factors for Carbon and Alloy Semifinished Steel

Carbon and alloy semifinished steel is the foundation of all other carbon and alloy steel products within the steel system boundary of this investigation. To capture differences in production methods that have substantial impacts on the emissions intensities of steel products, the partial LCI approach calculates three separate types of default emissions factors covering carbon and alloy semifinished steel:

·         $DefaultE{F}_{car semi, bof}$ refers to default emissions factors for carbon and alloy steel produced using BOFs.

·         $DefaultE{F}_{car semi, eaf,long}$ refers to default emissions factors for carbon and alloy steel produced using EAFs that is then used to make long steel products and seamless tubular steel products. These values are also used for the default emissions factors for $DefaultE{F}_{carsemi, eaf}$ generally, assuming that most receipts of carbon and alloy semifinished steel from EAF facilities are billets and ingots rather than slabs.

·         $DefaultE{F}_{car semi, eaf,flat}$ refers to default emissions factors for carbon and alloy steel produced using EAFs that is subsequently used to make flat steel products. The partial LCI approach did not use these emissions factors directly to calculate scope 3 emissions for U.S. steel facilities. These emissions factors were used only to determine the upstream emissions contribution for carbon and alloy semifinished steel in the calculation of default emissions factors for flat steel mill products (e.g., hot-rolled flat steel) produced using an EAF production pathway (these are calculated in step 3.5).

Each type of carbon and alloy steel described above is calculated using equation F.10 below.

$\begin{array}{c}DefaultE{F}_{car semi}=UE{F}_{car semi}+\left(Intensit{y}_{oxygen,car semi}*DefaultE{F}_{oxygen}\right)\\ +\left(Intensit{y}_{nitrogen,car semi}*DefaultE{F}_{nitrogen}\right)\\ +\left(Intensit{y}_{argon,car semi}*DefaultE{F}_{argon}\right) \\ +\left(Intensit{y}_{hydrogen,car semi}*DefaultE{F}_{hydrogen}\right)\\ +\left(Intensit{y}_{lstone,car semi}*\left(DefaultE{F}_{lstone}+DirectE{F}_{lstone}\right)\right)\\ +\left(Intensit{y}_{lime,car semi}*DefafultE{F}_{lime}\right)\\ +\left(Intensit{y}_{alloys,car semi}*\left(DefaultE{F}_{alloys}+DirectE{F}_{alloys}\right)\right)\\ +\left(Intensit{y}_{electrode,car semi}*\left(DefaultE{F}_{electrode}+DirectE{F}_{electrode}\right)\right)\\ +\left(Intensit{y}_{pig,car semi}*\left(DefaultE{F}_{pig}+DirectE{F}_{pig}\right)\right)\\ +\left(Intensit{y}_{dri,car semi}*\left(DefaultE{F}_{dri}+DirectE{F}_{dri}\right)\right)(F.10)\end{array}$

Values for $Intensit{y}_{input,carsemi}$ are shown for each input and for each production pathway in table F.4. The main differences between each production pathway involve the use of pig iron and direct reduced iron. BOF steelmaking uses considerably more pig iron than EAF steelmaking.638F[635] Although they do not use as much pig iron as BOFs, EAF facilities that produce flat steel products use considerably more pig iron and direct reduced iron than those that produce long steel products.639F[636] When production pathway cannot be determined for a steel product input, the Commission use the steps explained in box F.3 to generate an emissions factor.

 

Table F.4 Input intensity of upstream inputs in the production of carbon and alloy semifinished steel, by input category

In metric tons of inputs used per metric ton of material produced (mt input/mt material) for solid materials and thousand cubic feet per metric ton of material produced for gases (Mcf/mt material). BF-BOF = blast furnace and basic oxygen furnace; EAF = electric arc furnace.

Input category	Variable name	Unit of measure	Input intensity value (BF-BOF pathway)	Input intensity value (EAF pathway)
Oxygen	$Intensit{y}_{oxygen,car semi}$	Mcf/mt material	2.375	1.620
Nitrogen	$Intensit{y}_{nitrogen,car semi}$	Mcf/mt material	2.298	0.134
Argon	$Intensit{y}_{argon,car semi}$	Mcf/mt material	0.027	0.024
Hydrogen	$Intensit{y}_{hydrogen,car semi}$	Mcf/mt material	0.010	0.010
Calcined lime	$Intensit{y}_{lime,car semi}$	mt input/mt material	0.051	0.087
Non-calcined limestone and dolomite	$Intensit{y}_{lstone,car semi}$	mt input/mt material	0.015	0.000
Ferroalloys and alloying metals	$Intensit{y}_{alloys,car semi}$	mt input/mt material	0.018	0.027
Carbon electrodes	$Intensit{y}_{electrodes,car semi}$	mt input/mt material	0.000	0.004
Pig iron	$Intensit{y}_{pig,car semi}$	mt input/mt material	0.943	0.000$–$0.162
Direct reduced iron and hot briquetted iron	$Intensit{y}_{dri,car semi}$	mt input/mt material	0.000	0.000$–$0.228

Source: USITC compiled from Remus et al., Best Available Techniques (BAT) Reference Document, January 24, 2013, 369, 429.

Note: “Input intensity value” refers to the quantity of inputs used to produce one metric ton of carbon and alloy semifinished steel. An input intensity value of 0 indicates that the Commission assumed$—$for purposes of calculating scope 3 emissions factors$—$that steel producers using the production pathway specified in the column did not use the material specified in the row.

 

Step 3.4 Default Emissions Factors for Stainless Semifinished Steel

Stainless steel encompasses a highly diverse family of steels with a wide range of alloy mixtures, sources of alloying metal, production processes, scrap intensities, and corresponding grades of stainless steel produced. In seeking to develop standard default emissions factors for stainless semifinished steel, the partial LCI approach made the following simplifying assumptions:

1.       The default emissions factors for all stainless semifinished steel would be based on an EAF production pathway based on the prevailing method for global production of stainless steel.640F[637]

2.       The default emissions factors would be based on a grade that captures the essential characteristics of stainless steel$—$at least 10.5 percent chromium content and a generally high percent of nickel content. ASTM Grade 304 was selected because it is the most common grade of stainless steel produced globally at the time of this investigation. Grade 304 is defined in this investigation as having 19 percent chromium and 9.3 percent nickel.641F[638]

3.       The input intensities of nickel and chromium were dictated by the amount of stainless steel scrap used.642F[639] The amount of scrap used for the production of stainless steel was assumed to be 58 percent of total metallic inputs. This percentage represents a mid-point between relatively high scrap use in the United States and Europe$—$the main sources of most of the U.S. industry’s external receipts of stainless steel products$—$and lower scrap use globally.643F[640] This use of scrap led the Commission to assume that the input intensities of nickel and chromium provided by ferroalloys and other alloying metals (i.e., not from scrap) were 0.048 mt nickel and 0.095 mt chromium per metric ton of stainless semifinished steel produced.644F[641]

4.       Ferrochromium was the source of all chromium needed to produce stainless semifinished steel not supplied by scrap. Ferronickel and nickel metal were the sources of all nickel not supplied by scrap, accounting for 66.9 percent and 33.1 percent of the remaining nickel needed, respectively.645F[642] The metallic content of these materials was 53 percent chromium content for ferrochromium and 30 percent nickel content for ferronickel, consistent with the assumptions used to calculate the default emissions factors for these materials described in step 1.646F[643]

The assumptions described above allowed the Commission to calculate broad approximations of the input intensities of ferronickel, nickel metal, and ferrochromium in the production of stainless semifinished steel (see table F.5). For China and Indonesia, however, the assumptions described above do not adequately capture a significant difference in production practices in these two countries that results in far higher emissions intensities for stainless steel produced in these countries than in other countries. Both countries rely to a lesser extent on scrap and to a greater extent on ferroalloys and other alloying metals as a source of alloying metals. In particular, these two countries produce stainless steel using nickel pig iron, a highly emissions-intensive form of nickel used in stainless steel production.647F[644]

Table F.5 Input intensity of ferroalloy and other alloying metal inputs in the production of stainless semifinished steel for China, Indonesia, and the rest of the world, by input category

In metric tons of inputs used per metric ton of material (mt input/mt material).

Input category	Variable name	Input intensity value (China)	Input intensity value (Indonesia)	Input intensity value (rest of world)
Ferrochromium	$Intensit{y}_{fecr,ss semi}$	0.268	0.312	0.180
Ferronickel	$Intensit{y}_{feni, ss semi}$	0.000	0.000	0.107
Nickel	$Intensit{y}_{nimetal,ss semi}$	0.013	0.000	0.016
Nickel pig iron	$Intensit{y}_{nipig,ss semi}$	0.505	0.717	0.000

Sources: USITC estimates based on partial life cycle inventory (LCI) approach using data compiled from Gyllenram and Wei, 304 Stainless Steel Carbon Footprint Comparison: EU, Indonesia and China, October 2022; Vidovic et al., GHG Emission Intensities of the Steel, Fertilisers, Aluminium and Cement Industries, 2023; worldstainless, Global Life Cycle of Stainless Steel, June 26, 2023.

Note: “Input intensity value” refers to the quantity of inputs used to produce one metric ton of stainless semifinished steel. An input intensity value of 0 indicates that the Commission assumed$—$for purposes of calculating scope 3 emissions factors$—$that steel producers in the country specified in the column did not use the material specified in the row.

To capture the clear distinction in China and Indonesia’s sourcing of the nickel and chromium content in stainless semifinished steel production, assumptions 3 and 4 described above were modified in order to estimate different ferroalloy input intensities for these countries. Specifically, the Commission used a study by Gyllenram and Wei for its estimated contributions of nickel and chromium from scrap and ferroalloys from these countries. This study assumed the Chinese and Indonesian steel industries use lower quantities of stainless scrap, ferronickel, and nickel metal but higher quantities of ferrochromium and high quantities of nickel pig iron in the production of stainless semifinished steel (see table F.5).648F[645]

The Commission used equation F.11 to calculate the default emissions factor for stainless semifinished steel ( $DefaultE{F}_{ss semi}$ ) for each country. Different $Intensit{y}_{input, ss semi}$ values for China, Indonesia, and the rest of the world are shown for each alloying metal input in table F.5. For non-alloying metal inputs (industrial gases, calcined lime, and carbon electrodes), values for $Intensit{y}_{input,ss semi}$ are the same as the EAF-specific values for $Intensit{y}_{input,car semi}$ shown in table F.4 above.

$\begin{array}{c}DefaultE{F}_{ss semi}=UE{F}_{ss semi}+\left(Intensit{y}_{oxygen, ss semi}*DefaultE{F}_{oxygen}\right)\\ +\left(Intensit{y}_{nitrogen,ss semi}*DefaultE{F}_{nitrogen}\right)\\ +\left(Intensit{y}_{argon,ss semi}*DefaultE{F}_{argon}\right)\\ +\left(Intensit{y}_{hydrogen,ss semi}*DefaultE{F}_{hydrogen}\right)\\ +\left(Intensit{y}_{ss semi}*DefaultE{F}_{lime}\right)\\ +\left(Intensit{y}_{electrode,ss semi}*\left(DefaultE{F}_{electrode}+DirectE{F}_{electrodes}\right)\right)\\ +\left(Intensit{y}_{feni, ss semi}*\left(DefaultE{F}_{feni}+DirectE{F}_{feni}\right)\right)\\ +\left(Intensit{y}_{nimetal,ss semi}*\left(DefaultE{F}_{nimetal}+DirectE{F}_{nimetal}\right)\right)\\ +\left(Intensit{y}_{nipig,ss semi}*\left(DefaultE{F}_{nipig}+DirectE{F}_{nipig}\right)\right)\\ +\left(Intensit{y}_{fecr,ss semi}*\left(DefaultE{F}_{fecr}+DirectE{F}_{fecr}\right)\right)(F.11)\end{array}$

Step 3.5 Default Emissions Factors for Steel Mill Products

For both stainless steel and carbon and alloy steel mill products, default emissions factors generally are calculated as the sum of the unit process emissions for that product and the upstream emissions associated with the use of a corresponding steel substrate that would typically be used to produce the steel mill product. For example, in equation F.12 below, the default emissions factors for carbon and alloy hot-rolled flat steel are calculated using the emissions that occur during hot-rolling and upstream emissions associated with carbon and alloy semifinished steel.649F[646]

$$\begin{array}{c}DefaultE{F}_{car hr}=UE{F}_{car hr}+\left(Intensit{y}_{car semi,car hr}*DefaultE{F}_{car semi}\right)\left(F.12\right)\end{array}$$

The partial LCI approach uses the following input intensity measures derived from JRC 2023 to determine the quantities of steel inputs used to make other forms of steel:

·         Production of 1 mt of hot-rolled flat steel, hot-worked long steel, or seamless tubular steel requires 1.03 mt of semifinished steel.650F[647]

·         Production of 1 mt of cold-rolled flat steel requires 1.04 mt of hot-rolled flat steel.

·         Production of 1 mt of cold-formed long steel requires 1.04 mt of hot-worked long steel.651F[648]

·         Production of 1 mt of non-seamless tubular steel products requires 1 mt of hot-rolled flat steel.

Default emissions factors for carbon and alloy coated flat steel includes upstream emissions from cold-rolled flat steel ( $car cr$ ) as well as from three commonly used metallic coating materials: zinc, tin, and aluminum (equation F.13).

$$\begin{array}{c}DefaultE{F}_{car coat}=UE{F}_{car coat}+\left(Intensit{y}_{car cr,car coat}*DefaultE{F}_{car cr}\right)\\ +\left(Intensit{y}_{zinc,car coat}*DefaultE{F}_{zinc}\right) \\ +\left(Intensit{y}_{tin,car coat}*DefaultE{F}_{tin}\right)\\ +\left(Intensit{y}_{aluminumcar coat}*DefaultE{F}_{aluminum}\right) \left(F.13\right)\end{array}$$

Production of 1 mt of carbon and alloy coated flat steel requires 1 mt of cold-rolled flat steel.652F[649] The input intensity of coating metals used to produce coated flat steel products is assumed to be 0.069 mt of coating metal used to produce 1 mt of coated flat steel.653F[650] The input intensity of coating metals is divided into three coating metal types, with 0.047 mt of zinc, 0.016 mt of tin, and 0.005 mt of aluminum assumed to be used in the production of 1 mt of coated flat steel.654F[651]

Sensitivity Analyses

To explore the effects that certain parameters, methods, and respondents had on the overall emissions intensity estimates, the Commission ran sensitivity analyses of its emissions intensity calculations. The results of these analyses are presented here to provide insight into the aspects of the calculations that are driving the emissions intensity estimates, which may help inform future efforts to produce similar calculations. In its sensitivity analyses, the Commission recalculated its emissions intensity results using four alternative methods: (1) using the market-based method for generating scope 2 emissions, (2) incorporating fugitive methane emissions associated with coal and natural gas used in the production of covered steel and aluminum products, (3) restricting the population of responses only to those firms reporting to the Greenhouse Gas Reporting Program (GHGRP), and (4) using only default emissions factors to generate emissions intensities. Compared to the presentation of results in the chapters and in appendix I, results under method 1 represent a change to the methodology and input data, results under method 2 represent a change to the methodology and input data via an expansion of the Commission’s system boundaries, results under method 3 represent a change in the population, and results under method 4 represent a change to the scope 3 emissions factors that the Commission uses to calculate emissions intensities.

Market-Based Method

As summarized in box 1.2 in chapter 1, the GHG Protocol has two recommended methods for estimating emissions from purchased electricity: the location-based method and the market-based method. These methods are also each covered in the U.S. Environmental Protection Agency (EPA) guidance on calculating scope 2 emissions.655F[652] The market-based method is often used for company-level emissions reporting as it allows for instances where companies take actions, such as purchasing renewable energy certificates, to reduce their scope 2 emissions. Because the Commission’s calculations are U.S.-wide averages rather than company-specific calculations and because of challenges with occasional double-counting of emissions attributes under the market-based method, the main calculations used to generate the results in chapters 4 and 5 use the location-based method. However, the questionnaire collected the information needed to apply the market-based method. This section presents the approach for calculating these market-based method emissions and the results of that analysis.

The EPA guidance identifies six different electricity emissions factors in order of preference that may be used for the market-based method. The first three are for emissions factors that are not always relevant to the facility, but should be used when relevant and available. The Commission’s questionnaire collected data for each of these factors: energy attribute certificates, contracts to purchase electricity from specific generating facilities, and emissions factors specific to the utility or retail energy supplier from which the facility sourced their electricity. After these three emissions factors are applied, any remaining purchased electricity receives a default emissions factor. In order of preference, these default factors should use a residual mix, regional, or national factor. The residual mix factor represents the average emissions from electricity generation for a geographic area after excluding the attributes of electric generation in that area that are already counted in the certificates, contracts, and utility or retail supplier-specific factors. Residual mix emissions factors are not widely available. As such, the calculations use the same subregional emissions factors that were used in the location-based method.

The Green-e certification program publishes U.S. residual mix emissions factors for subregions in the EPA’s Emissions and Generation Resource Integrated Database (eGRID) that exclude the attributes of all Green-e certified renewable energy generation. However, these data were not available for the year 2022 in time for inclusion in this report. According to the 2021 residual mix data, the differences compared to the eGRID subregional emissions factors were small$—$less than 1 percent and sometimes less than 0.1 percent$—$in many of the subregions where electricity use was most concentrated. One exception is the eGRID SRMW subregion (which spans parts of Missouri, Illinois, and Iowa), where Green-e’s residual mix emissions factor was 1.66 percent higher in 2021 than the eGRID emissions factor.656F[653]

Steel

For steel produced in the United States, using the market-based method results in slightly different average emissions intensities than those generated by using the location-based method. The differences between the intensities calculated under either method were not statistically significant for any product category. On a facility-by-facility basis, emissions intensities calculated using the market-based method could be higher than those under the location-based method if facilities reported an emissions factor for their purchased electricity from their utility or retail energy supplier that was higher than the eGRID subregional emissions factor. By contrast, zero-emission energy attribute certificates, as were reported by multiple facilities with EAF steelmaking, served to decrease a facility’s emissions intensity under the market-based method relative to the location-based method. These certificates reduced the scope 2 emissions for those facilities and reduced the scope 3 emissions for U.S. facilities that sourced steel inputs from them.

Aluminum

As with the steel product categories, the average emissions intensities for aluminum product categories did not significantly differ between those generated by the market-based method and those generated by the location-based method. Purchases of zero-emission energy attribute certificates were much more common among facilities producing covered steel products, particularly electric arc furnaces, than among facilities producing covered unwrought aluminum products.657F[654] Unlike semifinished steel, many facilities producing unwrought aluminum can manufacture a relatively low emissions intensity product (secondary aluminum) without using large quantities of electricity, which may explain why the use of certificates was not as common.

Fugitive Emissions Associated with Coal and Natural Gas Used in Steel and Aluminum Production

In the main calculation methodology for product-level emissions intensities, the Commission assigns direct emissions factors that capture combustion emissions related to scope 1 and 2 coal and natural gas use. Coal and natural gas production generates additional emissions upstream of the point of combustion, which include fugitive methane emissions.658F[655] This section presents alternative results of average emission intensities for covered steel and aluminum product categories that incorporate estimates of fugitive methane emissions from cradle-to-gate coal and natural gas production activities.

The Intergovernmental Panel on Climate Change (IPCC) defines fugitive emissions as an intentional or unintentional release of gases from anthropogenic sources, excluding fuel combustion.659F[656] These emissions are mostly methane, and in fossil fuel production they are often released from pipeline leaks, venting and flaring in mining and drilling activities, and storage.660F[657] The EPA estimates that in 2022 fugitive methane emissions from U.S. natural gas production and processing were 104.1 million metric tons of CO₂e, while fugitive methane emissions from coal mining were 43.6 million metric tons of CO₂e.661F[658]

Fugitive emissions, often disperse or resulting from accidental leaks, are inherently difficult to measure, leading to higher degrees of uncertainty in emissions reporting.662F[659] Research has shown that national emissions inventories likely vastly undercount fugitive methane emissions associated with fossil fuel production, particularly oil and gas systems.663F[660] Advancements in emissions measurement$—$such as methane detection technologies using satellites, airplanes, and vehicles$—$have led to estimates of fugitive methane emissions from natural gas production and processing that are up to eight times higher than the amount reported in the EPA Inventory of U.S. Greenhouse Gas Emissions and Sinks (GHGI).664F[661] Increased ability to measure these data has encouraged federal government efforts toward updating methane emissions monitoring and reporting.665F[662]

The Commission conducted desk research and interviews with subject matter experts to inform its development of U.S.-specific fugitive emissions factors for coal and natural gas.666F[663] The Commission applied four sets of natural gas and coal emissions factors to capture uncertainty in fugitive methane emissions accounting, using factors with both a 20-year and 100-year global warming potential (GWP).667F[664] The factors use natural gas and coal production data from U.S. Department of Energy (USDOE) and EPA, respectively.668F[665] The main sets of factors use EPA GHGI data on methane emissions for coal and natural gas, while the upper-bound sets incorporate the USDOE Greenhouse Gases, Regulated Emissions, and Energy Use in Technologies (GREET) model’s estimates of fugitive emissions in natural gas production and processing along with the EPA’s high estimate for fugitive emissions from coal mining. The GREET model estimates fugitive methane emissions from natural gas production and natural gas processing to be 44$–$47 percent and 60 percent higher than what EPA reports, respectively.669F[666] Fugitive emissions factors used in the analysis presented below are reported in table G.21 of appendix G.

The calculations that incorporate fugitive emissions in the average product-level emissions intensity estimates largely follow the same allocation steps as scope 1 and scope 2 emissions associated with coal and natural gas combustion presented in appendix E. In this analysis, a separate subprocess-specific total of scope 3 fugitive emissions approximated from activity data on scope 1 and scope 2 coal and natural gas use ( $S3FGH{G}_{subproc}$ ) is added to the unit process emissions totals calculated in the main approach (equations F.14 and F.15).

$$S3FUGH{G}_{product}=S3FGH{G}_{subproc}* \frac{Outpu{t}_{product}}{Outpu{t}_{subproc}} \left(F.14\right)$$

$UGH{G}_{product}= S1P UGH{G}_{product}+S1FC UGH{G}_{\begin{array}{c}product\\ \end{array}} +S2 UGH{G}_{product}+S3 UGH{G}_{product}+ S3FUGH{G}_{product} (F.15)$

For fugitive emissions associated with scope 1 coal and natural gas use, the approach applies a fugitive emissions factor to quantities of natural gas and bituminous coal reported in section 3 and section 5 of the Commission questionnaire.670F[667]

Estimation of the coal and natural gas use embedded in a facility’s purchases required several additional calculation steps to the scope 2 emissions calculations presented in appendix E.671F[668] Natural gas and coal use associated with a facility’s plant-specific electricity purchase are calculated using plant-specific data on total plant generation and fuel mix from EPA’s eGRID.672F[669] The Commission used a similar approach to obtain the coal and natural gas use associated with plant-specific purchases of useful thermal output (UTO). The main calculation approach yielded total annual heat input from combustion and associated emissions from UTO purchases, which are further split in this analysis according to the coal and natural gas generation mix at the cogeneration plant.

The approach uses plant-level and subregional data in the eGRID database to assign coal and natural gas use to a respondent’s facility-level electricity purchases from the grid.673F[670] The sum of fuel-specific purchases from the grid and via direct-line connection yield total coal and natural gas quantity activity data. The Commission then multiplied these fuel quantities by their respective fugitive emissions factors to generate facility-level estimates of scope 3 coal and natural gas fugitive emissions from scope 2 purchases.

Marginal Aluminum and Steel Emissions Intensities from Fugitive Methane Emissions Associated with Coal and Natural Gas Use

Table F.6 presents the additional marginal emissions intensities from fugitive methane associated with coal and natural gas use for steel and aluminum aggregate product categories. The Commission calculated additional emissions intensity estimates from fugitive methane using a base and high-end estimate of total fugitive methane emissions associated with coal and natural gas production in the United States in 2022, applying both a 100-year GWP and 20-year GWP to methane.674F[671]

Higher relative coal use for certain product categories, an increased share of emissions intensity attributed to on-site fuel combustion and process emissions, and a lower baseline emissions intensity all contribute to a higher relative impact of fugitive methane emissions on the total emissions intensity.675F[672] Inclusion of Commission estimates of fugitive methane emissions associated with coal and natural gas use have the largest relative impact on the average emissions intensity of carbon semifinished steel among all product categories (increasing average emissions intensity by 6.1$–$23.9 percent). Primary unwrought aluminum has the highest absolute increase in its average emissions intensity when accounting for fugitive methane. A higher share of product-level emissions intensity attributed to use of coal and natural gas (through on-site activities and purchased energy) increases the marginal emissions intensities contribution of fugitive methane. On-site coal use was most prevalent in primary unwrought aluminum production and stainless steel and carbon semifinished production.

Table F.6 Steel and aluminum production: marginal product-level emissions intensities due to fugitive emissions from scope 1 and scope 2 coal and natural gas use

In metric tons of CO₂ equivalent per metric ton (mt CO₂e/mt) of steel. GWP = Global warming potential.

Product category	Average emissions intensity, main method	Marginal emissions intensities, fugitive 100-year GWP	Marginal emissions intensities, fugitive 20-year GWP	Percent increase to the average emissions intensity due to inclusion of fugitive emissions, 100-year GWP to high-end 20-year GWP
Primary unwrought aluminum	14.52	0.23	0.87	1.6$–$6.0
Secondary unwrought aluminum	2.46	0.02	0.09	0.7$–$3.5
Wrought aluminum	6.23	0.04	0.16	0.6$–$2.6
Carbon and alloy semifinished steel	1.02	0.06	0.24	6.1$–$23.9
Carbon and alloy flat steel	1.83	0.06	0.25	3.4$–$13.5
Carbon and alloy long steel	0.75	0.04	0.14	4.8$–$19.1
Carbon and alloy tube	1.50	0.01	0.04	0.6$–$2.8
Stainless steel	2.78	0.01	0.06	0.5$–$2.0

Source: USITC estimates based on its calculation methodology, see appendix E.

Note: The rightmost column provides the range in percent increase of product categories’ average emissions intensities when including marginal fugitive emissions from scope 1 and 2 natural gas and coal use. For example, the addition of marginal fugitive emissions from coal and natural gas increases the U.S. primary unwrought aluminum average emissions intensity of 14.52 mt CO₂e/mt by 1.6 to 6.0 percent, depending on the GWP and source estimates for total fugitive emissions from coal and natural gas production in the U.S. in 2022.

 

Greenhouse Gas Reporting Program Reporters Only

During the development of the data collection and analysis plan for this investigation, several industry stakeholders echoed the Trade Representative’s request for the Commission to avail themselves of public emissions data sources. Some stakeholders even urged the Commission to limit its analysis to only established, publicly available data, with the idea that these data would be more reliable and lead to the production of a more replicable and transparent methodology.676F[673]

Using this perspective, this sensitivity analysis compares production totals and average emissions intensities of covered products made by a well-identified group of facilities$—$those reporting to the EPA’s Greenhouse Gas Reporting Program (GHGRP)$—$to production totals and emissions intensities of covered products made by all facilities in the Commission’s survey population.677F[674] Both sets of emissions intensities were calculated using the same methods described throughout this report, including a combination of data from the GHGRP, the Commission’s questionnaire responses, and other data sources to cover direct and indirect emissions.678F[675] The only difference between the approaches is the population of facilities in each set. The results of this analysis show the impact on average emissions intensity of restricting the sample of facilities to only those reporting to the GHGRP. The results also provide a sense of how much production in a given product category is produced by GHGRP reporters, and where additional responses from the Commission’s survey population frame (that were incorporated as a result of the Commission’s additional research to develop a population list) contributed most to filling out the population coverage. There are no restrictions on what products may be produced by GHGRP reporters, however, the program’s annual reporting requirement thresholds means that GHGRP reporters are well-represented among the surveyed facilities that produce of emission-intensive intermediate products (i.e., those from steel mills and aluminum smelters) as shown in table F.7 and F.8.

Steel

The average emissions intensities of GHGRP reporters only were not significantly different than those from all facilities for any product category shown in table F.7.679F[676] As described in chapter 2, steelmaking and upstream on-site production processes are significant sources of direct emissions; as a result, most facilities with EAFs and all facilities with BF-BOFs are GHGRP reporters. Table F.7 shows that most production of semifinished and flat steel categories, as well as carbon and alloy hot-worked long steel products and carbon and alloy seamless tubular products, occurred at facilities that were GHGRP reporters. Most production of further downstream product categories was performed by non-GHGRP reporters.

Table F.7 Number and share of surveyed facilities reporting to the GHGRP that produced covered steel products and their share of overall production, by product category

In number and percentages (%).

Product category	Surveyed facilities reporting to the GHGRP (number)	Share of surveyed facilities reporting to the GHGRP (%)	Share of total production of surveyed facilities comprised by facilities reporting to the GHGRP (%)
Carbon and alloy semifinished	84	96.6	100.0
Carbon and alloy hot-rolled flat	41	87.2	99.4
Carbon and alloy cold-rolled flat	29	70.7	97.9
Carbon and alloy coated flat	33	73.3	86.5
Carbon and alloy hot-worked long	50	71.4	97.0
Carbon and alloy cold-formed long	8	8.1	30.6
Carbon and alloy seamless tubular	10	47.6	85.4
Carbon and alloy non-seamless tubular	3	3.1	5.8
Stainless semifinished	10	58.8	96.4
Stainless hot-rolled flat	10	71.4	98.9
Stainless cold-rolled flat	8	53.3	98.2

Sources: EPA, OAP, “FLIGHT Database, 2022 Greenhouse Gas Emissions from Large Facilities,” accessed various dates. USITC, Greenhouse Gas (GHG) Emissions Intensities Questionnaire: Facility-Level, 2024, responses to question 2.1.1.

Note: The number of GHGRP reporters in the stainless hot-worked long, cold-formed long, seamless tubular, and non-seamless tubular product categories was small enough to require data suppression, so these rows are not included above.

 

Aluminum

The average emissions intensities of only GHGRP reporters producing covered aluminum products were not significantly different than the averages from all facilities for any product category shown in table 5.8. Only 56 facilities producing covered aluminum products reported to the GHGRP, and some of the GHGRP reporters made both secondary and wrought products so they are included in both totals in table F.8. 680F[677] All (100 percent) facilities producing primary unwrought aluminum reported to GHGRP, while only a third of facilities producing secondary unwrought aluminum, and less than 10 percent of wrought aluminum-producing facilities were GHGRP reporters.681F[678] Although few facilities producing secondary unwrought and wrought aluminum reported to GHGRP, these reporters’ production accounted for a large share of total domestic aluminum production for each of these product categories (table F.8).

Table F.8 Number and share of surveyed facilities reporting to the GHGRP that produced covered aluminum products and their share of overall production, by product category

In number and percentages (%).

Product category	Surveyed facilities reporting to the GHGRP (number)	Share of surveyed facilities reporting to the GHGRP (%)	Share of total production of surveyed facilities comprised by facilities reporting to the GHGRP (%)
Unwrought	41	37.3	72.1
Primary unwrought	6	100.0	100.0
Secondary unwrought	35	33.7	69.5
Wrought	39	9.4	64.1

Sources: EPA, OAP, “FLIGHT Database, 2022 Greenhouse Gas Emissions from Large Facilities,” accessed various dates. USITC, Greenhouse Gas (GHG) Emissions Intensities Questionnaire: Facility-Level, 2024, responses to questions 2.2.1, 2.2.2, and 2.2.3.

Note: Facility counts may not sum to total because facilities may produce more than one product category.

Default Emissions Factors Only

Default emissions factors used in scope 3 analysis are inherently subject to uncertainty. For example, as described in this report, U.S. facilities producing covered steel and aluminum products operate under a wide range of energy efficiencies, purchase electricity from providers with different emissions profiles, have multiple production pathways, and use varying quantities of material inputs from a variety of sources. As a result, the emissions intensity of each U.S. supplier facility’s production of a given material can differ significantly from U.S.-specific or global default emissions factors for that material.

As described in chapter 3 and appendix E, the Commission calculated scope 3 emissions estimates using U.S. supplier-specific activity data (external receipts) and emissions factors for consuming facilities’ receipts of pig iron, steel products, and primary unwrought aluminum. For consuming facilities’ other receipts of these materials, the Commission used default emissions factors that were country specific or in some cases (for steel materials) product pathway specific. The Commission also used a country-specific emissions factor for alumina sourced from the United States. For all other materials in the system boundaries of this investigation, the Commission used global emissions factors. This approach is referred to here as the “main method” for calculating scope 3 emissions. In this section, a sensitivity analysis is conducted to examine the effects of data specificity and the choice of default emissions factors in calculating industry-wide emissions intensity estimates.

Under this analysis, referred to as the “default factors only method,” the Commission used country-specific and global default emissions factors in lieu of supplier-specific emissions factors to calculate scope 3 emissions for receipts from specific suppliers.682F[679] Specifically, equations E.45 (for pig iron), E.48 (for steel products), and E.53 (for primary unwrought aluminum) are replaced with equation F.15 below for each material ( $material$ ) received from a specific U.S. supplier facility ( $supplier$ ).683F[680]

$$\begin{array}{c}S{3}_{material,supplier}= Receipt{s}_{material,supplier}*DefaultE{F}_{material,us}\left(F.15\right)\end{array}$$

Steel

Except for a few product categories, emissions intensity estimates of steel product categories are largely unaffected by the approach used to calculate scope 3 emissions when comparing the results of the main method to the default factors method (see tables F.9 and F.10). Under the main method, the total scope 3 emissions associated with receipts from specific U.S. suppliers accounts for over half of all facility-level scope 3 emissions for the steel sector.684F[681] Because most scope 3 emissions were calculated using supplier-specific emissions factors under the main method, use of a default factors-only method (i.e., not using supplier-specific emissions factors) will increase or decrease the emissions intensities of products depending on whether suppliers are more or less emissions intensive than default emissions factors. The use of these default emissions factors did not have a significant impact on the resulting emissions intensity estimates for carbon and alloy semifinished, flat, or long product categories (table F.9), or for 6 out of the 7 stainless steel product subcategories (table F.10). By contrast, the average emissions intensity estimates of carbon and alloy non-seamless steel tubular products and stainless hot-rolled steel products are significantly less using the main method than using the default factors only method. This means that facilities making non-seamless steel tubular products receive flat steel products from U.S. suppliers that are less emissions intensive than default emissions factors for those substrate products.

Table F.9 Carbon and alloy: average product-level emissions intensity under the main method and the default-factors-only-method

In metric tons of carbon dioxide-equivalent per metric ton (mt CO₂e/mt) of steel. ** = rounds to zero (less than 0.005); * indicates the averages for each method are statistically significantly different at the 0.05 significance level.

Product category	Average emissions intensity, main method	Average emissions intensity, default factors only	Difference, main-default
Semifinished	1.02	1.03	**
Flat	1.83	1.75	0.08
Hot-rolled flat	1.59	1.54	0.05
Cold-rolled flat	1.91	1.79	0.12
Coated flat	2.17	2.00	0.18
Long	0.75	0.76	**
Hot-worked long	0.67	0.70	−0.02
Cold-formed long	1.25	1.16	0.09
Tubular	1.50	1.67	−0.17*
Seamless tubular	1.09	1.12	−0.03
Non-seamless tubular	1.71	1.96	−0.24*
Non-seamless OCTG	1.52	2.01	−0.49*
All other non-seamless	1.74	1.96	−0.21*

Source: USITC estimates based on its calculation methodology, see appendix E.

Table F.10 Stainless steel: average product-level emissions intensity under the main method and the default-factors-only-method

In metric tons of carbon dioxide equivalent per metric ton (mt CO₂e/mt) of steel. * indicates the averages for each method are statistically significantly different at the 0.05 significance level.

Product category	Average emissions intensity, main method	Average emissions intensity, default factors only	Difference, main-default
Stainless steel	2.78	3.02	−0.24*
Semifinished	2.23	2.22	0.01
Hot-rolled flat	2.31	2.75	−0.44*
Cold-rolled flat	3.08	3.26	−0.18
Hot-worked long	2.93	2.77	0.16
Cold-formed long	3.55	3.61	−0.06
Seamless tubular	4.07	4.04	0.04
Non-seamless tubular	3.16	3.42	−0.26

Source: USITC estimates based on its calculation methodology, see appendix E.

Aluminum

The emissions intensity for secondary unwrought and wrought aluminum product categories calculated using a global default primary aluminum factor were not found to be significantly different from those calculated using the U.S. supplying-smelter-specific emissions factors under the main method. This similarity of estimates under either method is in part explained by the input sourcing of surveyed facilities. Only about a quarter of input primary aluminum was reported as being sourced domestically by these facilities, and less than that was identified as being produced at a specific U.S. smelter.685F[682] Given the high share of non-U.S. primary aluminum as inputs into covered secondary unwrought and wrought aluminum production, the additional data granularity of U.S. smelter-specific emissions factors had a relatively small impact on the final emissions intensity.

Bibliography

Air Products. “Hydrogen Weight and Volume Equivalents.” Accessed November 12, 2024. https://microsites.airproducts.com/gasfacts/hydrogen.html?_ga=2.155415035.177972454.1662132901-582580380.1649040173.

Alvarez, Ramón A., Daniel Zavala-Araiza, David R. Lyon, David T. Allen, Zachary R. Barkley, Adam R. Brandt, Kenneth J. Davis, et al. “Assessment of Methane Emissions from the U.S. Oil and Gas Supply Chain.” Science 361, no. 6398 (July 13, 2018): 186$–$88. https://doi.org/10.1126/science.aar7204.

American Iron and Steel Institute (AISI). “AIS-10 Net Shipments of Steel Mill Products, All Grades Including Carbon, Alloy, and Stainless, December 2022.” January 31, 2023. https://www.steel.org/.

Arguillarena, Andrea, María Margallo, Ane Urtiaga, and Angel Irabien. “Life-Cycle Assessment as a Tool to Evaluate the Environmental Impact of Hot-Dip Galvanisation.” Journal of Cleaner Production 290 (March 25, 2021): 125676. https://doi.org/10.1016/j.jclepro.2020.125676.

Association of Iron & Steel Technology (AIST). 2022 Directory of Iron and Steel Plants. Warrendale, PA, 2022. https://www.aist.org/publications-advertising/directory%E2%80%94iron-and-steel-plants.

Burnham, Andrew. “Updated Natural Gas Pathways in GREET 2023.” Argonne National Laboratory (ANL), October 1, 2023. https://greet.anl.gov/publication-update_ng_2023.

Bussewitz, Cathy. “Difficulty Measuring Methane Slows Plan to Slash Emissions.” AP News, January 31, 2023. https://apnews.com/article/drones-business-climate-and-environment-0a1ab9be3427818cfe5f33521b9b05c2.

Denis-Ryan, Amandine. Gross Under-Reporting of Fugitive Methane Emissions Has Big Implications for Industry. Institute for Energy Economics and Financial Analysis, July 2023. https://ieefa.org/sites/default/files/2023-07/Gross%20under-reporting%20of%20fugitive%20methane%20emissions%20has%20big%20implications%20for%20industry_2.pdf.

Drnevich, R. F., C. J. Messina, and R. J. Selines. “Production and Use of Industrial Gases for Iron and Steelmaking.” In The Making, Shaping and Treating of Steel, 11th Edition: Steelmaking and Refining Volume. The AISE Steel Foundation, 1998. https://www.academia.edu/37434608/Steelmaking_and_Refining_Volume_AISE_1998.

EUROFORGE. International Statistics 2022. Accessed April 18, 2024. https://www.euroforge.org/wp-content/uploads/2022/11/International-Statistics-2022.pdf.

European Commission (EC). Directorate-General Taxation and Customs Union (DG-TAXUD). “Default Values for the Transitional Period of the CBAM Between 1 October 2023 and 31 December 2025.” December 22, 2023. https://taxation-customs.ec.europa.eu/system/files/2023-12/Default%20values%20transitional%20period.pdf.

European Foundry Association (CAEF). “The European Foundry Industry 2022.” November 2023. https://www.caef.eu/downloads-links/.

Green-e. “2023 Residual Mix Emissions Rates (2021 Data).” December 12, 2023. https://www.green-e.org/2023-residual-mix.

Gyllenram, Rutger, and Wenjing Wei. 304 Stainless Steel Carbon Footprint Comparison: EU, Indonesia and China. October 2022. https://www.researchgate.net/publication/370214813_304_Stainless_steel_Carbon_Footprint_Comparision-EU_Indonesia_and_China?enrichId=rgreq-4eff9998d6374d7572d847c96ac31524-XXX&enrichSource=Y292ZXJQYWdlOzM3MDIxNDgxMztBUzoxMTQzMTI4MTE1MzA4NTk0NEAxNjgyMzI3MDY2NjEz&el=1_x_2&_esc=publicationCoverPdf.

Intergovernmental Panel on Climate Change (IPCC). 2006 IPCC Guidelines for National Greenhouse Gas Inventories, Volume 2. 2006. https://www.ipcc-nggip.iges.or.jp/public/2006gl/vol2.html.

Intergovernmental Panel on Climate Change (IPCC). “Glossary- 2019 Refinement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories.” June 19, 2019. https://www.ipcc.ch/site/assets/uploads/2019/06/19R_V0_02_Glossary_advance.pdf.

International Chromium Development Association (ICDA). Quantifying the Improvement in the Environmental Impact of the Production of High Carbon Ferrochromium (HC FeCr). April 2022. https://www.icdacr.com/wp-content/uploads/2022/10/ICDA-HC-FeCr-LCA-summary-2022.pdf.

International Energy Agency (IEA). “Methane and Climate Change.” Methane Tracker, 2021. https://www.iea.org/reports/methane-tracker-2021/methane-and-climate-change.

International Energy Agency (IEA). Emission Factors 2023: Database Documentation. September 2023. https://iea.blob.core.windows.net/assets/bf862218-7fd8-4637-aca6-5a347b6ca4f1/IEA_Methodology_Emission_Factors_2023.pdf.

International Energy Agency (IEA). “Emissions Factors 2023.” September 2023. https://www.iea.org/data-and-statistics/data-product/emissions-factors-2023.

International Energy Agency (IEA). “World Energy Balances.” July 2024. https://www.iea.org/data-and-statistics/data-product/world-energy-balances.

International Energy Agency (IEA). World Energy Balances, 2024 Edition: Database Documentation. July 2024. https://iea.blob.core.windows.net/assets/feea6879-1796-41c1-b8a1-2da8c181cba0/WORLDBAL_Documentation.pdf.

Janjua, Rizwan, and Felipe Maciel. CO₂ Data Collection User Guide, Version 11. World Steel Association, May 30, 2024. https://worldsteel.org/wp-content/uploads/CO2_User_Guide_V11.pdf.

Koolen, Derck, and Danko Vidovic. Greenhouse Gas Intensities of the EU Steel Industry and Its Trading Partners. June 22, 2022. https://doi.org/10.2760/170198.

Laconde, Thibault. Fugitive Emissions: A Blind Spot in the Fight Against Climate Change. Global Observatory on Non-State Climate Action, 2018. https://www.climate-chance.org/wp-content/uploads/2019/03/new-fugitive-emissions-a-blind-spot-in-the-fight-against-climate-change.pdf.

McVay, Renee. Methane Emissions from U.S. Gas Pipeline Leaks. Environmental Defense Fund, August 2023. https://www.edf.org/sites/default/files/documents/Pipeline%20Methane%20Leaks%20Report.pdf.

Myers, Andrew. “Methane Emissions from Major U.S. Oil and Gas Operations Higher than Government Predictions.” Stanford Report, March 13, 2024. https://news.stanford.edu/stories/2024/03/methane-emissions-major-u-s-oil-gas-operations-higher-government-predictions.

Nduagu, Experience I., Deepak Yadav, Nishant Bhardwaj, Sabarish Elango, Tirtha Biswas, Rangan Banerjee, and Srinivasan Rajagopalan. “Comparative Life Cycle Assessment of Natural Gas and Coal-Based Directly Reduced Iron (DRI) Production: A Case Study for India.” Journal of Cleaner Production 347 (May 1, 2022): 131196. https://doi.org/10.1016/j.jclepro.2022.131196.

Nickel Institute. Life Cycle Data: Executive Summary. January 2023. https://nickelinstitute.org/media/fbmdel4y/lifecycledata-summary-updatejan2023.pdf.

Norgate, T. E., S. Jahanshahi, and W. J. Rankin. “Alternative Routes to Stainless Steel - A Life Cycle Approach.” Tenth International Ferroalloys Congress 7, no. 3 (January 2004): R126. https://www.researchgate.net/publication/242689888_Alternative_routes_to_stainless_steel-a_life_cycle_approach.

Outokumpu Stainless USA, LLC. Written submission to the U.S. International Trade Commission in connection with Inv. No. 332-598, Greenhouse Gas Emissions Intensities of the U.S. Steel and Aluminum Industries at the Product Level. December 21, 2023. https://edis.usitc.gov/external/search/document/810868.

Remus, Rainer, Miguel A. Aguado-Monsonet, Serge Roudier, and Luis Delgado Sancho. Best Available Techniques (BAT) Reference Document for Iron and Steel Production. January 24, 2013. https://doi.org/10.2791/98516.

ResponsibleSteel. ResponsibleSteel International Production Standard: Version 2.1. May 21, 2024. https://www.responsiblesteel.org/standards.

Riddick, Stuart, and Denise Mauzerall. “Likely Substantial Underestimation of Reported Methane Emissions from United Kingdom Upstream Oil and Gas Activities.” Energy & Environmental Science 16, no. 1 (2023): 295$–$304. https://doi.org/10.1039/D2EE03072A.

Rutherford, Jeffrey S., Evan D. Sherwin, Arvind P. Ravikumar, Garvin A. Heath, Jacob Englander, Daniel Cooley, David Lyon, Mark Omara, Quinn Langfitt, and Adam R. Brandt. “Closing the Methane Gap in US Oil and Natural Gas Production Emissions Inventories.” Nature Communications 12, no. 1 (August 5, 2021): 4715. https://doi.org/10.1038/s41467-021-25017-4.

Silverado Policy Accelerator. Written submission to the U.S. International Trade Commission in connection with Inv. No. 332-598, Greenhouse Gas Emissions Intensities of the U.S. Steel and Aluminum Industries at the Product Level. November 17, 2023. https://edis.usitc.gov/external/search/document/808917.

Sphera Solutions, Inc. “LCA Database.” Life Cycle Assessment Datasets, 2024. https://lcadatabase.sphera.com/.

Steel Manufacturers Association (SMA). Steelmaking Emissions Report 2022. June 14, 2022. https://steelnet.org/steelmaking-emissions-report-2022/.

Total Materia. “Production of Stainless Steel: Part Two.” March 2008. https://www.totalmateria.com/page.aspx?ID=CheckArticle&site=kts&NM=220.

United Nations, ed. International Standard Industrial Classification of All Economic Activities (ISIC), Rev. 4. Statistical papers. Series M. New York: United Nations, 2008. https://unstats.un.org/unsd/publication/seriesm/seriesm_4rev4e.pdf.

Universal Industrial Gases, LLC (UIG). “Oxygen Quantity Conversions Calculator.” Universal Industrial Gases. Accessed November 12, 2024. https://www.uigi.com/calculators/oxygen-quantity-conversions-calculator/.

U.S. Environmental Protection Agency (EPA). Greenhouse Gas Inventory Guidance: Indirect Emissions from Purchased Electricity. December 2023. https://www.epa.gov/sites/default/files/2020-12/documents/electricityemissions.pdf.

U.S. Environmental Protection Agency (EPA). Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990$–$2022. EPA 430-R-24-004. 2024. https://www.epa.gov/system/files/documents/2024-04/us-ghg-inventory-2024-main-text_04-18-2024.pdf.

U.S. Environmental Protection Agency (EPA). “Greenhouse Gas Reporting Program (GHGRP). Envirofacts GHG Query Builder.” Accessed September 18, 2024. https://enviro.epa.gov/query-builder/ghg.

U.S. Environmental Protection Agency (EPA). “Methane Emissions Reduction Program and GHGRP Subpart W (Petroleum and Natural Gas Systems).” Accessed October 21, 2024. https://www.epa.gov/inflation-reduction-act/methane-emissions-reduction-program-and-ghgrp-subpart-w-petroleum-and.

U.S. Environmental Protection Agency (EPA). Office of Atmospheric Protection (OAP). “Greenhouse Gas Reporting Program (GHGRP). Facility-Level Information on Greenhouse Gases Tool (FLIGHT) Database.” Accessed various dates. https://ghgdata.epa.gov/ghgp/main.do?site_preference=normal.

U.S. International Trade Commission (USITC). Greenhouse Gas (GHG) Emissions Intensities Questionnaire: Facility-Level. 2024. https://www.usitc.gov/research_and_analysis/documents/sa_emissions/emissions_facility-level_questionnaire_aluminum_and_steel_combined.pdf.

Vidovic, Danko, Alain Marmier, Lovro Zore, and Jose A. Moya. Greenhouse Gas Emission Intensities of the Steel, Fertilisers, Aluminium and Cement Industries in the EU and Its Main Trading Partners. European Commission. Joint Research Centre (JRC), 2023. https://data.europa.eu/doi/10.2760/359533.

World Resources Institute (WRI). GHG Protocol Scope 2 Guidance. 2015. https://ghgprotocol.org/sites/default/files/2023-03/Scope%202%20Guidance.pdf.

World Steel Association (worldsteel). Life Cycle Inventory (LCI) Study: 2020 Data Release. May 2021. https://worldsteel.org/wp-content/uploads/Life-cycle-inventory-LCI-study-2020-data-release.pdf.

World Steel Association (worldsteel). Steel Statistical Yearbook 2023. Accessed September 21, 2024. https://worldsteel.org/publications/bookshop/ssy_subscription-2023/.

worldstainless. “Stainless Steel in Figures 2020.” (2020). https://www.worldstainless.org/Files/issf/non-image-files/PDF/ISSF_Stainless_Steel_in_Figures_2020_English_public_version.pdf.

worldstainless. Global Life Cycle of Stainless Steel. June 26, 2023. https://www.worldstainless.org/news/global-life-cycle-of-stainless-steel/.

worldstainless. “Stainless Steel CO₂ Emissions Report.” (August 2023). https://www.worldstainless.org/files/issf/non-image-files/PDF/worldstainless_CO2_Emissions_Report.pdf.