Development of toxicity footprints at national level

Alexandra Segolene Corinne Leclerc

Research output: Book/ReportPh.D. thesisResearch

37 Downloads (Pure)


Sustainable development has become a priority for our society, as illustrated by the adoption of the Sustainable Development Goals by the United Nations and the increasing environmental awareness of citizens. Both international organisations and scientific research have identified chemical pollution as a barrier to sustainable development, degrading both human health and the environment. Yet, they lack comprehensive data and indicators to gauge the extent of this problem. Which chemicals do we release into the environment? In what quantities? Where? From which activities? To satisfy which demand? What are their impacts? These are some of the questions that this PhD thesis tries to answer. Such a project requires (i) the development of national inventories of chemical emissions with a comprehensive coverage of substances, emission sources, environmental compartments and countries, (ii) their integration into a modelling structure enabling consumption-based accounting, and (iii) the assessment of their impacts on human health and ecosystems. With more than 100,000 chemicals on the market and being used in a wide variety of applications, the quantification of their releases into the environment is however a challenge. At product level, process Life Cycle Inventory (LCI) databases are used to quantify all the chemical emissions occurring in the life cycle of a product, from the extraction of the raw materials, through its manufacturing and use, until its endof-life. Despite their richness in terms of substance coverage, process LCI databases have not been used to develop national inventories of chemical emissions. To investigate if process LCI databases could effectively advance the development of national emission inventories, this approach was tested with the case of global heat and electricity generation in 1995-2014. This case study led to the general observations
that producing national and yearly inventories from process LCI databases was a resource-consuming procedure hampered by the need for physical data to upscale the processes, and by the lack of time differentiation in the process emission data. In addition, it was identified that process LCI databases may not be adapted to evaluate chemical emissions from some sectors such as chemical manufacturing or the health care sector, due to the low availability of representative

To overcome this barrier, publicly available databases were thus analysed and used to build national inventories of chemical emissions to air, water and soil, with a global coverage wherever possible. Industries, households, and agriculture were identified to be among the main sources of chemical releases in the environment, with a potential for developing comprehensive inventories of chemical emissions. With differences in the availability of public data, tailored methods were
formalized and applied for each of these emission sources. Industrial emissions were extrapolated at sectoral level from Pollutant Release and Transfer Registers using proxies such as the economic output, carbon dioxide emissions or particulate matter emissions. Direct releases to water from households were extrapolated from reports of diffuse emissions using demographic statistics. Emissions of pharmaceutical substances were derived from consumption data, knowing their removal in the human body and in wastewater treatment plants. The combination of quantities applied and pollutant concentrations allowed estimating the emissions to soil resulting from the agricultural use of manure and sewage sludge. Finally, agricultural emissions of pesticides were derived from application data and harvested area. At European level, thanks to the large availability of public data, national inventories were produced for each European Union Member States from 2000 to 2014, covering altogether the releases of 805, 572 and 468 chemicals to air, water and soil, respectively. At global scale, the complementary use of publicly vailable databases resulted in globally differentiated inventories for the year 2011, with a coverage of chemical emissions never achieved before. Yet, the development of national inventories of agricultural pesticide emissions with a global coverage is prevented by the paucity of data on the application of pesticide active ingredients in non-European countries. These inventories of territorial emissions help evaluate the sustainability of our production systems. Yet, with international trade, supply chains are stretching around the world to satisfy one’s consumption, and with them, chemicals emissions. A territorial approach is therefore not sufficient, and must be complemented with a more global perspective, where we account for the environmental pressures and impacts driven by our consumption. Such an analysis is possible with the use of environmentally extended multi-regional inputoutput (EE-MRIO) models. The consumption-based accounting of chemical emissions was thus operationalized by integrating the newly built global inventories of chemical emissions in the state-of-the-art EEMRIO model EXIOBASE 3.7 for the year 2011. This procedure is constrained by challenges such as the accounting rules in input-output analysis, the allocation to economic sectors or the global coverage. Altogether, the releases of 693 compounds to air, water and soil, were integrated in EXIOBASE 3.7, covering emissions from transportation, industries, households and agriculture (yet excluding pesticides). In a third step, these inventories were translated into toxic using the consensus life cycle impact assessment (LCIA) model USEtox v2.11. The PhD work resulted in the production of national toxicity footprint indicators evaluating the toxic impacts on human health and freshwater ecosystems resulting from total consumption in 44 countries. With their broader coverage of industrial, residential and transport-related emissions, the global inventories developed in this thesis are associated with larger toxicity impacts than in previous studies. They show however a high sensitivity to the extrapolation of industrial emissions, and need to expanded with global inventories of agricultural pesticide emissions.
While the territorial approach allows identifying which chemicals emissions should be curbed in priority to reduce territorial impacts, the toxicity footprints can provide a global perspective to the chemical pollution problem and give a fair account of one’s true impacts by accounting for international trade. By identifying where demand-driven impacts occur, one can evaluate the potential influence of local regulations and/or international chemical management to efficiently
reduce impacts on human health and freshwater ecosystems. Beyond the obtained toxicity footprint results, which carry a number of uncertainties and limitations, the PhD work provides a number of advances towards consistent toxicity footprints by formalizing inventorying methods to build national inventories of chemical emissions, systematically evaluating their potential for improvements, and operationalizing their integration with EE-MRIO models and LCIA methodologies.
Original languageEnglish
Number of pages304
Publication statusPublished - 2020


Dive into the research topics of 'Development of toxicity footprints at national level'. Together they form a unique fingerprint.

Cite this