HAL-INERIS
Not a member yet
8807 research outputs found
Sort by
Source apportionment of PM10 particles in the urban atmosphere using PMF and LPO-XGBoost
No full textInternational audienceAtmospheric particulate matter (PM), as a leading part of air pollution, affects health in many ways. Thus, identifying and quantifying the contribution of atmospheric particulate matter sources of PM is vital for developing effective air quality management strategies. Positive Matrix Factorization (PMF) is one of the most common methods for source apportionment. However, PMF has some limitations, particularly its assumption that each source contributes linearly. In reality, some sources may exhibit nonlinear behaviors, which can compromise the accuracy of source apportionment. This study introduces a Lung Performance Optimization-based XGBoost (LPO-XGBoost) model, which leverages adaptive optimization principles inspired by lung function to enhance classic PM source apportionment. We demonstrate the potential for efficient, real-time application of the LPO-XGBoost model across 21 monitoring sites in 6 European countries. Trained and validated on extensive environmental datasets, the model is capable of predicting major pollution sources, including road traffic, biomass burning, crustal, industrial, nitrate-rich particles, sulfate-rich particles, heavy fuel oil, and sea salt. It outperforms other machine learning models with an overall predictive coefficient of determination (r2 = 0.88). Notably, the model performs exceptionally well in predicting sources such as sea salt (r2 = 0.97) and biomass burning (r2 = 0.89), but shows lower accuracy for the sulfate-rich particles source (r2 = 0.75). Comparative analyses with models including Random Forest (RF), Support Vector Machine (SVM), and their LPO-enhanced variants confirm that LPO-XGBoost provides the most reliable performance in estimating pollution source contributions, offering scalability and robustness ideal for high-time-resolution observational data. This model has significant potential to support targeted air quality management strategies. Future research should focus on expanding key species measurements at monitoring sites, ensuring consistent temporal coverage, and optimizing the model for improved mixed-source predictions to strengthen its applicability in comprehensive urban air quality assessments
Source apportionment of PM<sub>10</sub> based on offline chemical speciation data at 24 European sites
No full textInternational audienceThis study applied Positive Matrix Factorization (PMF) to PM10 speciation datasets from 24 urban sites across six European countries (France, Greece, Italy, Portugal, Spain, and Switzerland) to perform a detailed source apportionment (SA) analysis. By using a consistent source apportionment tool for all datasets, the study enhances the comparability of PM10 SA results across urban Europe. The results identified seven major PM10 sources including road traffic, biomass burning, crustal/mineral sources, secondary aerosols, industrial emissions, sea salt, and heavy oil combustion (HOC). Road traffic emerged as the predominant source of PM10 in urban areas, with contributions varying by location, but representing as much as 41% in high-traffic zones. Biomass burning was detected at 23 sites, contributing 8% to 41% on yearly averages, with substantial increase in winter. Crustal sources were present at all sites (3–33%). Industrial sources contributed relatively less PM10 mass, which was identified at 10 sites with contributions ranging from 2% to 14%. Secondary inorganic and organic aerosol, consisting primarily of ammonium nitrates and sulfates, and organic matter, formed a portion of the PM10 mass (5–41%). These secondary factors are primarily influenced by anthropogenic emissions, including the various combustion processes. Sea salt, predominantly found in coastal areas, contributed between 4% and 21%, reflecting the impact of the marine environments on air quality. This source was very often ‘aged’ (mixed with anthropogenic pollutants from different origins). Additionally, HOC, especially emits from shipping activities, and traced by V and Ni, was also a frequent contributing source (2–15% for 9 sites), indicating a need for more stringent emission controls. The chemical comparison is performed which indicates road traffic and secondary aerosols, showed consistent chemical profiles across sites, while industrial, HOC, and crustal sources displayed significant site-specific variability. These findings underscore the need for tailored air quality strategies according to local sources of emissions and the importance of long-term PM speciation monitoring for effective pollution control
A decision-support tool for repurposing hydrocarbon wells for geothermal application in France
No full textInternational audienceThe opportunities and barriers relating to the conversion of hydrocarbon wells in France for the development of geothermal energy have been studied by Ineris and BRGM since several years, particularly following the promulgation of Law No. 2017-1839 of 30 December 2017 putting an end to the exploration and the exploitation of hydrocarbons in France. The law also provides for the study of the potential for repurposing hydrocarbon facilities for other uses of the subsoil and, in particular, the opportunities for developing geothermal energy projects.In this context, BRGM and Ineris have developed, with the support of the General Direction of Risk Prevention (DGPR), a decision-support tool for public authorities based on flowcharts to identify the eligibility of hydrocarbon wells for conversion to geothermal uses. The repurposing requires to verify several criteria as the well accessibility, the energy needs identified near the facility, the presence of geothermal resources, the technologies and energy exchange systems for exploiting these resources that can be developed, the well integrity (cementing and casings) and potential work to apply before repurposing, the environmental and operational risks associated with the conversion and the regulatory framework in France
evaltools
No full textThe Python package `evaltools` is designed to assess surface atmosphere composition prediction models against in-situ observations. This package provides several tools to compute model scores and plot them. It is used for evaluation of air quality models from the Copernicus Atmosphere Monitoring Service (CAMS).The concept of `evaltools` is to compare observations (measured over time at fixed lat/lon locations) with simulations (wich may have a forecast horizon of several days) computed over a period of several days. Therefore, it can be suitable for other data types such as AERONET data, but will not handle data with a vertical component.Up-to-date source code and documentation are available at https://github.com/UMR-CNRM/EvalTool
Experimental characterisation of cellular materials and sandwich panels filled with cellular core materials for structural protection against blast loadings
No full textInternational audienceExperimental characterisation of cellular materials and sandwich panels filled with cellular core materials for structural protection against blast loading
Using Machine Learning for Predicting Collapse extending in Abandoned Underground Mines
No full textInternational audienceThe main objective of this work is to explore the advantages of applying machine learning tools to the analysis of collapse related to abandoned underground mines. To this end, the present work focuses on the analysis of data recorded by Ineris for 143 collapses cases, with particular attention to the extension of the collapse to the surface. This work presents the procedure followed from the creation of a clean database, through the selection of features, to the proposal of a tool capable of estimating the extension of the collapse to the surface. This exploratory work confirms the significant potential of using machine learning on geotechnical data
Évaluation de la neurotoxicité des contaminants des sites et sols pollués: Apport de la modélisation de données à l’évaluation du risque pour le cerveau
No full textInternational audienceAlors que de nombreux travaux sont menés sur les outils et méthodes de caractérisation de la pollution des sols (diagnostic de sites) et sur les modalités de gestion des sites et sols pollués (méthodes de réhabilitation), l’étude des dangers et des effets toxicologiques des substances retrouvées au niveau de ces sites reste limitée. Le but du projet BrainSol a été d’évaluer les effets neurotoxiques de l’exposition à des substances présentes au niveau de sites et sols pollués (hydrocarbures aromatiques polycycliques (HAP), métaux, composés organochlorés) en population générale. Le projet a permis le développement d’une méthodologie innovante reposant sur la modélisation des données, d’une part pour caractériser les expositions aux sites et sols pollués, les scenarii d’exposition et les niveaux d’exposition, et d’autre part les effets neurotoxiques des substances seules ou en mélange. Les résultats relatifs à l’exposition ont permis d’identifier les HAP, le plomb et le cadmium comme étant les composés les plus préoccupants en matière de risque au niveau des sites et sols pollués. Des formalismes de type Adverse Outcome Pathway de la neurotoxicité de ces substances ont été construits et ont permis d’identifier une effet neurotoxique commun qui sont des troubles cognitifs liés à la mémorisation et aux émotions. Ces démarches de modélisation constituent des approches prometteuses en matière d’évaluation du risque sanitaire au niveau des sites et sols pollués
Impact of Radiofrequency Electromagnetic Fields on Cardiac Activity at Rest: A Systematic Review of Healthy Human Studies
No full textInternational audienceRadiofrequency electromagnetic fields (RF-EMF) exposure is increasingly prevalent, raising concerns about potential non-thermal health effects. This systematic review synthesizes current evidence regarding RF exposure effects on cardiac activity, focusing on heart rate (HR) and heart rate variability (HRV). Studies on healthy individuals were selected based on strict methodological criteria, including experimental design, control for confounding variables, and sufficient details on exposure parameters. Articles were included if they compared healthy subjects with and without exposure and provided cardiac measurements, specific absorption rate, or exposure measurement. A total of 28 articles were analyzed. This review included studies with RF exposure ranging from 100 to 110,000 MHz and exposure durations varying from short periods to 7 nights, with most studies lasting between 5 and 50 min. Most studies demonstrated no significant effects of RF exposure on HR, regardless of the exposure system, frequency, duration, age, sex, distance, or subject position. Findings for HRV were more nuanced, with most studies indicating no significant impact on key HRV parameters. However, some position-dependent variations emerged, particularly in antenna-based studies. Additionally, our analysis suggests that RF exposure may particularly interfere with cardiac regulatory mechanisms when the cardiovascular system is challenged and required to adapt, such as during postural changes or physiological maneuvers, although there are insufficient comparable studies to validate this hypothesis. Importantly, all included studies were conducted under resting or non-stressful conditions and involved only healthy participants. Therefore, our conclusions cannot be generalized to stressed states or clinical populations. Moreover, methodological harmonization is needed to improve comparability across future studies. The main limitation of the current evidence being the heterogeneity of experimental protocols, highlighting the need for methodological standardization in future studies. To address current heterogeneity, we propose specific methodological recommendations, including systematic blinding, accurate exposure measurement and detailed exposure, to improve comparability and reproducibility in future studies. Bioelectromagnetics. 00:00–00, 2025. © 2025 Bioelectromagnetics Society
Opinion and report of the French Agency for Food, Environmental and Occupational Health & Safety (ANSES) relating to an analysis of the health risks associated with exposure to caterpillars with stinging hairs and the development of management recommendations
No full textXavier Bailly et Elsa Jourdain ne sont pas auteurs mais relecteurs du rapportInternational audienceSeveral species of Lepidoptera with caterpillars bearing urticating hairs are increasing their range in metropolitan France from year to year, such as the pine processionary (Thaumetopoea pityocampa Denis et Schiffermüller 1775) whose presence was historically limited to the south of France, and which is gradually being observed in the north and at higher altitudes. The procession period also seems to be appearing earlier. Climate change could have an impact on the range and phenology of this species, as well as other species of Lepidoptera with caterpillar bearing urticating hairs. Urticating caterpillars, which can trigger diffuse urticariform papules and allergic reactions for human people through contact with the skin, mucous membranes or inhalation, represent a health issue. These caterpillars also cause a health risk for animals and can have an impact on plants due to the resulting defoliation. This risk assessment therefore focused on the risk posed by these urticating caterpillars in areas where they have been established for a long time as well as in areas where they currently do not occur but could be colonized in the short or medium terms. One of the outputs of the evaluation consisted in identifying 55 different species of Lepidoptera with urticating caterpillars, for metropolitan France, but also for the overseas regions, based on the review by Kawamoto and Kumada (1984) and the GBIF database. In addition, detailed sheets for the well‐known species were drawn up, describing in particular their range, their phenology and the period when the urticating stages occur. A literature review as well as data from poison control centres and the analysis of the results of two surveys (of three associations of dermatologists/allergists and of occupational populations at risk) allowed to characterize the contexts of exposure and the human populations at risk. Concerning animal health, in addition to an epidemioclinical description based on the literature, data from veterinary poison control centres were analyzed in the specific case of dog exposure to the pine processionary between 1999 and 2021 in metropolitan french territory. Given the impact of these species of Lepidoptera on ecosystems, due to the defoliation they induce, this scientific opinion investigated the impact of these species on ecosystem services. The socio‐economic impacts of defoliations and risk management measures implemented in the event of an outbreak were studied based on a literature review and simulations for forest recreation. To represent risk for the whole metropolitan France territory, three risk maps of the human population exposure to the pine processionary, the oak processionary and the brown‐tail moth were designed at municipality level. A multi‐criteria analysis was undertaken to characterize danger (urticating caterpillars) and vulnerability of human populations. Then, the municipalities were ranked according to their level of danger and vulnerability. As a result, municipalities have been classified into 5 danger classes and 5 vulnerability classes. Based on a risk matrix with levels of hazard (abscissa axis) and levels of vulnerability (ordinate axis), 5 levels of risk were finally defined and the 34,816 municipalities of metropolitan France were classified according to their level of risk
Alternative H2O2 Production Processes: An Outlook on Candidate Technologies Beyond the Anthraquinone Process
No full textInternational audienceHydrogen peroxide is currently made on a largescale via the auto-oxidation (AO) process, an indirect process viathe hydrogenation of anthraquinone over a Pd catalyst, followed byreaction with O2. The AO process has a significant climatefootprint, since H2 is usually sourced from methane steamreforming. A more sustainable H2O2 production process is,therefore, highly relevant. Potential alternative routes include thedirect reaction of H2 and O2 and processes that use electricity,either directly via electrochemical or bioelectrochemical systems orindirectly via plasmas. Despite research on these routes, there arecurrently no real challengers to the established AO process. An overview of these alternative processes is provided before consideringthe energetic, techno-economic, life cycle, and safety challenges facing them. By highlighting these challenges through comparisonwith the AO process, this review aims to inspire further work for scaling up these alternative processes for truly determining theirindustrial viability. As these processes may ultimately prove more advantageous than the AO process for on-demand production,their industrialization might not need to occur on the same production scale as that of the AO process