We provide here a description and outline of the methodology and key data sources used to compile our process life-cycle inventory (PLCI) database. Three life-cycle impact categories are included for all the materials, products and processes in the database: greenhouse gas emissions (embodied carbon), primary energy use (embodied energy) and water use (embodied/virtual water). The methodology used to build the database is consistent with international standards (the ISO 14040, PAS 2050, GHG Protocol).
Data obtained from external data sources include both activity data and emission factors. Activity data characterize the life cycle of a product/process by accounting for all material and energy inputs consumed, material and energy outputs generated, transport, storage, and waste outputs generated throughout the life cycle. The system boundary for the process modeling is "cradle to factory gate", "cradle to farm gate", "well to wheel", or "cradle to grid" as appropriate. Emission factors are then used to translate this process activity data into GHG emissions, covering both energy-related and non-energy-related GHG emissions (CO2, as well as other significant non-CO2 greenhouse gas emissions). Variations due to different production methods and geographical regions are preserved as much as possible in the database.
Modeling Carbon Storage
Carbon storage in products/processes lasting more than a year -- including a variety of wood-based manufactured products, textiles incorporating natural fibers, concrete structures, planted trees, etc. -- are modeled based on methodologies and parameters adapted from IPCC tier 1/2 and the Danish Technological Institute.
Modeling Waste Processing
Both solid waste and waste water streams are modeled in detail based on methodologies and parameters adapted from IPCC tier 1/2 for a broad range of industries. Solid waste modeling includes aerobic/anaerobic landfilling, incineration, composting, and recycling/reuse. Waste water modeling includes aerobic and anaerobic treatments. Methane and energy recovery options are included with waste processing steps. Recycling is modeled using the "recycled content" method which allocates the costs and benefits of recycling to the input side of product systems; the system boundaries are drawn such that the system that produces the recyclable waste is responsible up to the point of delivering the waste to a recycling facility, and then any subsequent transport, processing and use of that material is included within other systems that use the material in some form. Other types of waste material that may be useful elsewhere, such as manure from animal systems, are handled in a similar manner: The product systems that use the material, such as organic crop systems that use manure as a substitute for fertilizers, get credit for avoiding the resource use and emissions associated with fertilizer manufacture; and these systems also bear the burden of actually applying the manure and the subsequent nitrous oxide emissions from the soil.
Modeling Land Use
Land use modeling is based on methodologies and parameters adapted from IPCC tier 1/2, and includes two specific scenarios: unchanged land use and changes in land use. All common land use categories are included. Factors considered include changes in above-ground and below-ground biomass and changes in soil carbon -- during land use change as well as over time.
Modeling Dynamic Emissions/Sequestration
Dynamic, or time-dependent, GHG emissions and carbon sequestration are modeled by explicitly considering the time dimension over a specific assessment period (such as the standard 100 years). Emission and sequestration events are weighted according to the timing of the events within the assessment period, as part of our Deep Carbon Footprinting™ methodology.
Allocation Methods
Allocation of resource use and emissions between co-products is performed by dividing a process into distinct sub-processes, or by using mass-weighted economic value or a biophysical measure (such as mass, energy or nutrition content) as appropriate. We have generally avoided system expansion because of the inherent difficulties and uncertainties involved in identifying and characterizing appropriate marginal product systems. Mass-weighted economic value has proven to be the most reliable method of allocation in many real-world scenarios, particularly for product systems that produce highly dissimilar co-products. Low-value (and often high-volume) waste outputs that may be useful elsewhere, such as recyclable material waste or manure from animal systems, are handled as part of the waste processing algorithm described above.
Modeling Agricultural Processes
Agricultural processes such as the production of food commodities are modeled uniformly based on a detailed inventory of inputs and outputs as indicated below:
Emission Factors
Emission factors for extraction and combustion of primary fuels -- as well as non-energy-related emission factors for GHG emissions inherent in industrial, agricultural, transport, and other processes -- are consistently derived and calculated using the following two sources:
Global Warming Potentials
Global warming potentials for greenhouse gases are based on the IPCC Fifth Assessment Report (AR5):
Electricity
Primary energy use and GHG emissions per unit of electricity supplied through the grid are calculated using activity data -- consisting of fuel and power plant mixes for various grid regions (both US and international), as well as transmission losses and other details -- from these sources:
Transport
Primary energy use and GHG emissions per tonne-km of freight transport for all transport modes (road, rail, ocean, and air) are calculated using activity data from these sources:
Refrigeration
Primary energy use and GHG emissions for refrigerated storage in warehouses and transportation are calculated using activity data from EPA Energy Star.
Materials
GHG emissions, primary energy use, and water use for many basic manufacturing processes and materials (including metals, plastics, cement, timber, fibers, fabrics, etc.) used in construction, manufacturing and packaging are calculated through an analysis of the US Life-Cycle Inventory Database. Additional data sources for materials include the Inventory of Carbon and Energy, Eco-Profiles of the European Plastics Industry, Copper Development Association LCA, NIST BEES Software, peer-reviewed research publications, LCA/LCI studies available in the public domain, and industry sources. Data for fertilizer production are based on IFA publications, pesticide data are derived from the Encyclopedia of Pest Management, and water/wastewater treatment data are from ACEEE and IPCC.
Food and Agriculture
GHG emissions, primary energy use, and water use for food and agricultural products and processes are calculated using activity and other data from numerous credible sources, including university agricultural extensions, agro-economics departments, government agencies and peer-reviewed research publications. Representative data sources include:
This section outlines the methodology and key data sources used to compile our environmentally extended input-output life-cycle inventory (EEIOLCI) database. Three life-cycle impact categories are included for all the goods and services in the database: greenhouse gas emissions (embodied carbon), primary energy use (embodied energy) and water use (embodied/virtual water). At the core of the database is the USEEIO dataset which provides input-output life cycle model for 385 goods and services in the US economy based on one 2013 US dollar of producer price as reference. Our implementation enhances this model to cover purchaser price, inflation, and other necessary extensions in order to make it usable for product life cycle assessments and corporate GHG inventory analyses. Our EEIOLCI database uses one 2019 US dollar of purchaser price as reference (the reference year will be updated annually). It also uses the structure of the distribution channel used to deliver the good or service to determine the profit margins at retail and distribution in order to calculate the price actually paid to the producer. The distribution channel can be one of four possibilities: Direct from producer (also applies to most services), online, wholesale and retail. It is the producer price that is finally mapped to the environmental flows and life-cycle impact categories associated with the production of a good or service. The system boundary is cradle-to-gate.
The EEIOLCI database provides data on every good or service in the economy, so the LCA or GHG inventory models that use this data are complete by definition and do not suffer from cutoffs or other data gaps. In particular, the EEIOLCI database can be used to efficiently model scope 3 emissions (examples: purchased materials and services, waste disposal, third-party freight transport, warehousing and retail, etc.) that may be difficult and time-consuming to model using PLCI. A hybrid methodology combining PLCI (for accurate scope 1/2 emissions and some scope 3 emissions) and EEIOLCI (for 100% coverage of most scope 3 emissions) can be very effective in quickly compiling corporate GHG inventories and for offsetting corporate or personal emissions.
The EEIOLCI database also supports point-of-sale carbon offsetting by mapping either the North American Industry Classification System (NAICS) code of specific industry sectors/subsectors OR the Visa Merchant Category Code (MCC) to the greenhouse gas emissions per dollar of purchaser price.
A limitation of EEIOLCI is that it cannot be used for evaluating the relative impacts of process improvements or material substitutions within an industry sector, or for comparing the environmental impacts of two similar products, due to the data aggregation by industry sector. Imported commodities are assumed to have the same input structure and the same production characteristics as comparable products of equal value produced domestically (after including any tariffs in the price of imports), and are modeled using domestic production as proxy.
US Environmentally-Extended Input-Output Dataset
The USEEIO dataset from the Environmental Protection Agency combines data on economic transactions between 389 industry sectors with environmental data for these sectors covering various resource uses and emissions, to build a life cycle model of 385 US goods and services. The dataset represents the 50 US states. The original model was intended to represent conditions for the year 2017 as closely as possible. Resource uses and emissions reflect the environmental burdens of producing one 2013 US dollar worth of a good or service.
Gross Domestic Purchases Price Index
The gross domestic purchases price index is the featured measure of inflation in the US economy and is produced by the Bureau of Economic Analysis (US Department of Commerce). The index measures the prices of goods and services purchased by US residents, regardless of where the goods and services were produced.
Operating and Net Margins by Sector
This dataset provides the gross, net and operating margins for 94 industry sectors in the US consisting of a total of over 7000 firms. This includes manufacturers, distributors, retailers and service providers.