Data for International (Fair) Trade in Air-Quality-Related Mortality

Note: The GTAP dataset includes a total of 140 regions, some of which are aggregated regions. For all map-related supplementary files (S11, S12, S13), we assign values to each individual country to enhance visualization. Countries within the same aggregated region are assigned the same regional value to maintain consistency across the map.

Data S1 (separate file): S1.csv

File Format: CSV file detailing production-related deaths for the GTAP dataset.

Rows: Each row represents a country where deaths occur as a result of production activities.

Columns: Each column represents a country-sector pair on the production side.

Values: The values indicate the number of deaths caused by production activities in the country-sector listed in each column and occurring in the country listed in each row.

Data S2 (separate file): S2.csv

File Format: CSV file detailing production-related deaths for the EORA dataset.

Structure: The file has the same structure as S1.csv.

Data S3 (separate file): S3.csv

File Format: CSV file detailing consumption-related deaths for the GTAP dataset.

Rows: Each row represents a country where deaths occur as a result of consumption activities.

Columns: Each column represents a consumption country.

Values: The values indicate the number of deaths caused by consumption activities in the country listed in the column and occurring in the country listed in the row.

Data S4 (separate file): S4.csv

File Format: CSV file detailing consumption-related deaths for the EORA dataset.

Structure: The file has the same structure as S3.csv.

Data S5 (folder of files): S5/

File Format: A folder containing CSV files, each named after a country's 3-digit code (e.g., USA.csv, CHN.csv), representing production-related spatial PM₂.₅ concentration patterns for all GTAP countries.

Rows: Each row corresponds to a grid cell.

Columns: Each column represents an industrial sector. The final column, "geometry," contains the spatial coordinates (latitude and longitude) for each grid cell.

Values: Each value indicates the PM₂.₅ concentration level (in µg/m³) attributable to emissions from the specified sector in the given country, as they occur in each grid cell.

Data S6 (folder of files): S6/

File Format: A folder containing CSV files, each named after a country's 3-digit code, representing production-related spatial PM₂.₅ concentration patterns for all EORA countries.

Structure: Each file follows the same format as those in S5/, with rows representing grid cells and columns representing industrial sectors, plus a "geometry" column containing spatial coordinates.

Data S7 (separate file): S7.csv

File Format: CSV file containing consumption-related spatial PM₂.₅ concentration patterns for all GTAP countries.

Rows: Each row represents a grid cell.

Columns: Apart from the last column ("geometry"), which contains spatial information for each grid cell in latitude-longitude coordinates, each column represents a consumption country.

Values: Each value indicates the PM₂.₅ concentration level caused by each country’s consumption process and occurring in each grid cell, measured in µg/m³.

Data S8 (separate file): S8.csv

File Format: CSV file containing consumption-related spatial PM₂.₅ concentration patterns for all EORA countries.

Structure: The file has the same structure as S7.csv.

Data S9 (separate file): S9.csv

File Format: CSV file listing the total net bidirectional export of deaths for all countries in GTAP, displaying only positive values.

Columns:

"from": The country that exports more consumption-related deaths.

"to": The country that imports more consumption-related deaths.

"values": The net export of deaths between these two countries, calculated as the difference between the deaths flowing from "from" to "to" and those from "to" to "from."

Data S10 (separate file): S10.csv

File Format: CSV file listing the total net bidirectional export of deaths for all countries in EORA, displaying only positive values.

Structure: The file has the same structure as S9.csv.

Data S11 (separate file): S11.csv

File Format: CSV file listing the Value of Statistical Lives (VSLs), and consumption-related externalities under three scenarios—Business as Usual (BAU), Global Community (GC), and Fair Trade in Deaths (FTD)—along with externalities per GDP and their differences for GTAP countries.

Columns:

VSL, BAU_Externality, GC_Externality, FTD_Externality

BAU_Ext_perGDP, GC_Ext_perGDP, FTD_Ext_perGDP

Diff_GC_BAU, Diff_FTD_BAU, Diff_FTD_GC

Data S12 (separate file): S12.csv

File Format: Same as S11.csv, but for EORA countries.

Structure: Identical to S11.csv.

Data S13 (separate file): S13.csv

Purpose: Includes data used to generate Figures 1, 2, 3, and 5 in the main text.

Columns:

country_code: 3-letter country code

GTAP_region, continent, population, GDP, GDP_capita, VSL

export_of_death, import_of_death, net_export, net_export_capita

allforeign_world, G50foreign_world, G100foreign_world

cause_allforeign_world, cause_L30foreign_world, cause_L50foreign_world

BAU_Externality, GC_Externality, FTD_Externality

BAU_Ext_perGDP, GC_Ext_perGDP, FTD_Ext_perGDP

Diff_GC_BAU, Diff_FTD_BAU, Diff_FTD_GC

geometry (used for visualization)

Data S14 (separate file): S14.xlsx
This Excel file contains six sheets summarizing cross-model Pearson correlation coefficients between sectoral economic activity fractions and transboundary mortality impact metrics, based on both GTAP and EORA datasets.

Sheets:

Output_fraction_GTAP

Direct_demand_fraction_GTAP

Final_demand_fraction_GTAP

Output_fraction_EORA

Direct_demand_fraction_EORA

Final_demand_fraction_EORA

Rows: Each row represents an economic sector.

Columns:

G50foreign_world: Fraction of deaths attributable to final demand from regions where demand per capita is more than 50% higher than in the current country.

cause_L50foreign_world: Fraction of deaths caused by consumption within the current country but occurring in countries with more than 50% lower demand per capita.

Values: Each value represents the Pearson correlation between the sectoral fraction and the corresponding transboundary mortality metric.

Data S15 (separate file): S15.csv

File Format: CSV file derived from the GTAP dataset, containing Monte Carlo simulation results
(500 draws) for the uncertainty analysis of production-based premature deaths.

Column Producer: The producing country–sector pair responsible for the emissions leading to health impacts.

Column Affected Country: The country where the resulting premature deaths occur.

Column Deaths: The estimated number of deaths corresponding to the one used in the main analysis.

Columns Deaths_median, Deaths_low95, Deaths_high95: The median, 2.5th percentile, and 97.5th percentile values across 500 Monte Carlo draws of the GEMM θ parameter, representing the 95% confidence interval for each producer–affected country pair.

Data S16 (separate file): S16.csv

File Format: CSV file derived from the GTAP dataset, containing Monte Carlo simulation results (500 draws) for the uncertainty analysis of consumption-based premature deaths.

Column Consumer: The consuming country whose final demand drives the global production and associated health impacts.

Column Affected Country: The country where the resulting premature deaths occur.

Column Deaths: The estimated number of deaths corresponding to the one used in the main analysis.

Columns Deaths_median, Deaths_low95, Deaths_high95: The median, 2.5th percentile, and 97.5th percentile values across 500 Monte Carlo draws of the GEMM θ parameter, representing the 95% confidence interval for each consumer–affected country combination.