Insights into the Spatial Distribution of Global, National, and Regional Nitrogen Oxide Emissions
The Importance of Understanding Spatially Resolved NOx Emissions
To mitigate the impact of greenhouse gas (GHG) and air pollutant emissions, it is of utmost importance to understand where emissions occur. In the real world, atmospheric pollutants are produced by various human activities from point sources (e.g., power plants and industrial facilities) but also from diffuse sources (e.g., residential activities and agriculture). However, as tracking all these single sources of emissions is practically impossible, emission inventories are typically compiled using national-level statistics by sector, which are then downscaled at the grid-cell level using spatial information.
In this work, we develop high-spatial-resolution proxies for use in downscaling the national emission totals for all world countries provided by the Emissions Database for Global Atmospheric Research (EDGAR). In particular, in this paper, we present the latest EDGAR v8.0 GHG, which provides readily available emission data at different levels of spatial granularity, obtained from a consistently developed GHG emission database. This has been achieved through the improvement and development of high-resolution spatial proxies that allow for a more precise allocation of emissions over the globe.
A key novelty of this work is the potential to analyze subnational GHG emissions over the European territory and also over the United States, China, India, and other high-emitting countries. These data not only meet the needs of atmospheric modelers but can also inform policymakers working in the field of climate change mitigation. For example, the EDGAR GHG emissions at the NUTS 2 level (Nomenclature of Territorial Units for Statistics level 2) over Europe contribute to the development of EU cohesion policies, identifying the progress of each region towards achieving the carbon neutrality target and providing insights into the highest-emitting sectors.
Developing High-Resolution Spatial Proxies for Emissions Downscaling
Knowing where emissions are released is essential to supporting the design of effective mitigation actions and for atmospheric modeling purposes. Emission inventories are typically developed at the national level and provide sector-specific emission estimates. In order to disaggregate national emissions over high-resolution grids, information on the location of the different emission sources (e.g., point, linear, and area sources) must be collected, and “spatial proxies” should be developed and applied to national sector-specific emission totals to downscale them over grid maps.
The correct allocation of point source emissions is essential to avoid misplacing high emission levels. However, gathering information on point sources covering the entire globe and a wide temporal domain (1970 to present) is challenging because of limitations in data availability, in the accuracy of the reporting (real location vs. legal address, etc.), and in the completeness of data.
The Emissions Database for Global Atmospheric Research (EDGAR) provides global greenhouse gas (GHG) and air pollutant emissions over the global grid map at a 0.1°×0.1° resolution, obtained through a downscaling process of national emissions using high-resolution spatial data. The development and maintenance of the EDGAR grid maps is essential since several regional and global databases rely on the EDGAR emission grid maps to disaggregate national emissions to the grid.
This work is an update of previous EDGAR publications dealing with spatial data and describes all the new developments in the spatialization of the emissions from EDGAR v8.0 onwards, focusing on not only high-emitting sectoral point sources, such as power plants and industrial activities, but also more diffuse sources, such as residential activities.
Key Improvements in EDGAR v8.0 Spatial Proxies
The main novelties of this work are:
-
Update on Emission Point Sources: An update on emission point sources using global datasets (e.g., Global Energy Monitor) to improve the representation of power plants, industrial facilities, and other high-emitting point sources.
-
Gap-Filling Method for Non-Population-Based Sources: The development of a gap-filling method for non-population-based sources using built-up surface information for non-residential areas from the Global Human Settlement Layer (GHSL). This approach helps to better allocate emissions from industrial and other non-residential activities.
-
Update of Population-Based Proxies: An update of population-based proxies using the latest GHSL data, including a weighting for the temperature-dependent need for heating. This improves the allocation of emissions from residential and agricultural activities.
-
Update on International Ship Tracks: An update on international ship tracks and weights by vessel type, which is crucial for accurately representing emissions from the shipping sector.
-
Inclusion of Subnational Information: The incorporation of subnational information (e.g., for Europe at the NUTS 2 level) when developing the new spatial proxies for EDGAR, allowing for a more accurate allocation and analysis of subnational emissions.
These updates, applied from EDGAR v8.0 onwards, aim to provide a more precise spatial representation of emissions across different sectors and regions, enabling a better understanding of emission patterns and trends at both the national and subnational levels.
Updating Spatial Proxies for Key Emitting Sectors
Power Plants
Power plants represent a major source of fossil-fuel-derived CO2 and other GHG emissions globally, nowadays contributing around 38% and 18%, respectively, of the corresponding global totals. It is therefore of utmost importance to spatially allocate these emissions correctly at the global level and understand their trends over time in order to design and implement adequate emission mitigation measures.
In EDGAR v8.0, fuel-specific spatial proxies have been developed using data from the Global Coal Plant Tracker and Global Oil and Gas Plant Tracker of the Global Energy Monitor (for coal and gas), the Global Power Plant Database v1.3.0 (for oil and biofuels), and the Carbon Monitoring for Action database (CARMA v3.0) for autoproducers (i.e., plants and industries producing power for their own use). The time frame covered by the new power plant spatial proxy datasets developed in EDGAR v8.0 is 1970-2022, which includes, for each plant, information on opening and closing years, capacity, and main fuel type.
Industrial Activities
Industrial activities cover a wide range of sectors, encompassing not only the production of iron and steel, cement, glass, metals, chemicals, and fertilizers but also the use of solvents and intensive animal farming. Gathering information on industrial activities at the global level is challenging, in part because of confidentiality and data protection issues.
In EDGAR v8.0, the latest E-PRTR (E-PRTR v18) locations for all industrial facilities (with the exception of power plants, iron and steel facilities, and coal mines) have been included. Several manual adjustments were made to overcome data quality issues related to missing spatial information and inconsistencies. Additionally, the global locations of iron and steel plants have been updated using the Global Steel Plant Tracker of the Global Energy Monitor.
Coal Mines
Coal mines are also a relevant source of fugitive emissions of GHGs and air pollutants (e.g., volatile organic compounds). In EDGAR v8.0, the information on coal mines at the global level has been updated using the Global Coal Mine Tracker of the Global Energy Monitor and the EIA data for the United States.
Gas Flaring
Gas flaring is the burning of the natural gas that results from oil extraction. It is a highly polluting practice and a source of GHG and air pollutant emissions. In EDGAR v8.0, data from the World Bank Global Gas Flaring Tracker Report have been used for estimating both the emissions and the location of global flaring activities from 2012 to 2022.
Agriculture
Agriculture includes a variety of activities that are typically distributed over large areas (e.g., crop areas and animal pastures). However, several agricultural activities can be defined as hotspots or point sources and include intensive animal farming and manure management practices.
In EDGAR v8.0, the infrared atmospheric sounding interferometer (IASI) satellite-derived NH3 point source database has been included to map emissions from animal farming and fertilizer production with yearly information for the period 2008-2022.
Gap-Filling for Industrial Emissions
A significant improvement in EDGAR v8.0 is the development and use of a new spatial proxy to gap-fill missing information for all industry-related emissions. Until EDGAR v7.0, population-related proxies were used as backup information when no spatial data were available to represent the emissions for a sector within a country.
In EDGAR v8.0, the non-residential built-up surface information developed by the GHSL has been used as a backup proxy to distribute the emissions of all the activities not related to small-scale combustion, for which no point source information was available. This methodological assumption is a key novelty of this work and supports the validity of this approach by aligning with methodologies already applied in regional inventories, such as in Europe.
Residential and Agricultural Emissions
Small-scale combustion emissions are mostly related to non-industrial activities, such as those from the residential, commercial, agricultural, and fishing sectors. EDGAR v8.0 aims to couple population distribution with heating degree-days (HDDs) to better represent the spatial distribution of these emissions, as the amount of emissions is dependent on not only the number of people living in a certain area but also the meteorological conditions and the need for heating indoor spaces.
Subnational Emissions Analysis
A key strength and novelty of EDGAR v8.0 is the inclusion of subnational information, representing emissions at the regional level for various countries. This allows for a more accurate allocation and analysis of emissions at the subnational scale, which is essential for the development of effective climate mitigation policies.
Case Study: European Union
The EDGAR v8.0 subnational data are regularly used as input to the European semesters and contribute to climate action territorial and cohesion policies through the EU cohesion reports. For example, the analysis of GHG emissions at the NUTS 2 level over Europe shows that out of 242 EU regions, 155 regions have shown a downwards trend in emissions since 1990. However, in 2021, only 34 regions achieved GHG emissions of less than 5 t CO2 equivalent per person, which is the average value needed to achieve the 2030 EU climate targets.
Case Study: United States
EDGAR v8.0 also includes GHG emission estimates at the subnational level for the United States, with emissions estimates for each US state. Based on the analysis, Texas emitted 11.5% of the total US GHG emissions in 2022, followed by California with a contribution of 7.7%, and Florida, with a share of 4.6%.
Case Study: China and India
For China, the five most-emitting provinces contributed around 40% of China’s total GHG emissions in 2022, with Shandong, Guangdong, Jiangsu, Hebei, and Nei Mongol being the top contributors. In India, five states contributed around 50% of the country’s total GHG emissions, with Maharashtra, Tamil Nadu, Uttar Pradesh, Gujarat, and Chhattisgarh being the top emitters.
These subnational analyses highlight the importance of understanding emission patterns at finer spatial scales to develop targeted and effective climate mitigation strategies.
Conclusion
The EDGAR v8.0 GHG global emission maps and subnational emission data can be accessed at https://edgar.jrc.ec.europa.eu/dataset_ghg80. These data not only meet the needs of atmospheric modelers but can also inform policymakers working in the field of climate change mitigation by providing a detailed understanding of emission patterns and trends at both the national and subnational levels.
The key strengths of the EDGAR v8.0 spatial proxies include the improved representation of point sources, the development of a gap-filling method for non-population-based sources, the update of population-based proxies, the inclusion of subnational information, and the update on international ship tracks. These improvements aim to provide a more precise spatial representation of emissions across different sectors and regions, enabling a better understanding of emission patterns and trends.
The availability of high-resolution, quality-controlled global and subnational emission data is essential for supporting the design of effective climate change mitigation actions and for atmospheric modeling purposes. The EDGAR v8.0 dataset represents a significant step forward in providing such data to the scientific community and policymakers.
