HAL-INERIS
Not a member yet
8807 research outputs found
Sort by
Comment veiller les risques émergents dans des horizons de ruptures ? Leçons tirées du cadrage d’une démarche sur les risques technologiques, sanitaires et environnementaux
No full textInternational audienceLes méthodes prospectives présentent un indéniable intérêt pour améliorer la veille des risques technologiques, sanitaires et environnementaux émergents, en particulier dans un monde post-Covid 19, où les ruptures radicales rentrent dans le champ des futurs possibles. L’exercice de cadrage réalisé par l’Ineris sur sa veille stratégique peut constituer un cas pratique de l’apport de ce type d’approche collective et ouverte pour détecter des « signaux faibles » et renforcer la capacité d’anticipation de la surveillance. La dimension prospective permet notamment de prendre du recul sur les risques eux-mêmes, en redonnant son importance à tout ce dans quoi ces risques s’insèrent. Ce premier travail de « criblage » a fait émerger quelques points d’intérêt dans les trajectoires mises en lumière, en particulier dans l’évolution des méthodes d’évaluation et de maîtrise des risques, ainsi que dans les jeux d’acteurs concernés par les risques technologiques et les impacts sanitaires et environnementaux associés
Can AI Revolutionize QSPR Models for the Chemical Mixtures Hazards?
No full textInternational audienceThe physical hazards of chemical mixtures are typically characterized using experimental tools that could benefitto be prioritized by using predictive methods. Indeed, experimental tests can be costly, complex, timeconsuming,and potentially dangerous for the operator. In the last decades, particularly with the implementationof the REACH regulation, predictive methods such as QSAR/QSPR (Quantitative Structure-Activity/PropertyRelationships) have been encouraged and utilized as rapid and economical alternatives to experimental testingfor determining (eco)toxicological and physical hazards of chemical substances.Initially designed for pure compounds, adaptations of the QSPR approach were proposed to predict theproperties of mixtures even if their development, in particular for physical hazards, is still an emerging field.Indeed, existing QSPR models still present some limitations to complement mixing rules and experimentalapproaches, and there is a need for new and more reliable models to extend applicability and improve predictionaccuracy. A possible orientation could be using advanced machine learning approaches, taking advantage ofscientific progress in artificial intelligence beyond classical multilinear regressions. More complex non-linearapproaches (such as neural networks or random forests) have recently been used with the hope of betteraccounting for mixture complexity in QSPR models for mixtures.This research aims to investigate if integrating advanced AI analytical methods can enhance the performanceand applicability of QSPR models for predicting the physical hazards of chemical mixtures. To this end,applications of different machine learning methods were tested to evidence the advantages and limits of theseAdvanced AI algorithms compared to the more classical MLR approach when developing models for theflammability of liquid mixtures
Technical note: Reconstructing missing surface aerosol elemental carbon data in long-term series with ensemble learning
No full textInternational audienceGround-based measurements of elemental carbon (EC) – classified under thermal–optical methods and considered a surrogate for black carbon – are essential for assessing air quality and evaluating climate impacts. However, data gaps caused by technical challenges impede comprehensive analyses of long-term trends. This study proposed an ensemble learning modeling method to address these challenges. The model used readily accessible ground observation air pollutant data as proxies for EC-related tracers, along with meteorological parameters, to enhance prediction accuracy. It integrated outputs from Gradient Boosting Regression Trees, eXtreme Gradient Boosting, and random forest models, combining them through ridge regression to produce robust predictions. We applied this approach to reconstruct hourly EC concentrations from 2013–2023 for four cities in eastern China, filling 45 %–79 % of missing data and improving prediction performance by 8 %–17 % compared to individual models. Over the 11-year period, EC exhibited an overall decline (−0.20 to -0.14µgm-3a-1), with a more significant decrease from 2013–2020 (−0.24 to -0.15µgm-3a-1). During this time, the average EC concentration in the four cities dropped from 3.26 to 1.59 µg m−3, followed by a noticeable slowdown in the rate of decline from 2020–2023 (−0.12 to -0.04µgm-3a-1). Additionally, a fixed emission approximation method based on ensemble learning was proposed to quantitatively analyze the drivers of long-term EC trends. The analysis revealed that anthropogenic emission controls were the predominant contributors, accounting for approximately 92 % of the changes in EC trends from 2013–2020. However, their influence weakened post-2020, contributing approximately 80 %. These findings highlighted that while China's Clean Air Actions implemented since 2013 have substantially reduced black carbon concentrations, sustained and enhanced strategies are still necessary to further mitigate black carbon pollution
CFD modeling of hydrogen release and dispersion in a congested container
No full textInternational audienceHydrogen plays an important role in driving decarbonization within the current global energy landscape. As hydrogen infrastructures rapidly expand beyond their traditional applications, there is a need for comprehensive safety practices, solutions, and regulations. Within this framework, the dispersion of hydrogen in enclosed facilities presents a significant safety concern due to its potential for explosive accidents. In this study, hydrogen dispersion in a confined and congested environment (37 m³ container) is studied using computational fluid dynamics simulations. The experimental setup mirrors previous INERIS investigations, featuring a centrally located hydrogen injection point on the floor with a 20 mm diameter injector and a release rate of 35 g/s, resulting in a Froude number of 650. This yields an inertial jet and a challenging dispersion scenario for numerical prediction. Concentration mapping is carried out by 3 oxygen analyzers distributed throughout the 37 m3 chamber. Comparisons are made between the measured and numerical data to validate the solver used under such conditions. Best practice guidelines are followed, and sensitivity studies involving grid refinement and boundary conditions are conducted to ensure robust simulation results. The findings highlight the model's ability to reproduce the hydrogen concentration distribution for both empty and congested containers and underline the role of accounting for leakages in such scenarios
Source apportionment of transport-related air pollution in the Port of Rotterdam
No full textInternational audienceThe Port of Rotterdam is the largest harbor in Europe, and thus a very dense and complex region of industrial pollution and transport-related emissions. As one of the City Pilots for the Mitigating Transport Related Air Pollution (MI-TRAP) project, a suite of high time resolution measurements of aerosol composition, size distribution, and volatility are deployed in summer 2025 at a monitoring site in the port to test whether this additional information enables improved source apportionment among various industrial, shipping, and intermodal transport emissions in Rotterdam. Preliminary results from the deployment of an Aerosol Chemical Speciation Monitor (ACSM), an x-ray fluorescence spectrometer for elemental analysis (X-Act), an aethalometer for equivalent black carbon measurement, and a mobility particle size spectrometer (MPSS) will be presented. The size distribution and eBC measurements alternate between sampling ambient air and via a catalytic stripper, to remove semivolatile components, giving measurements of both total and solid particle size distributions and eBC concentrations. We assess the benefits of these additional measurements for near-real-time source apportionment in the Port of Rotterdam
Dust pollution substantially weakens the impact of ammonia emission reduction on particulate nitrate formation
No full textInternational audienceDust emissions significantly influence air quality and contribute to nitrate aerosol pollution by altering aerosol acidity. Understanding how dust interacts with ammonia emission controls is crucial for managing particulate nitrate pollution, especially in urban environments. In this study, we conducted field measurements of aerosol chemical components and gases across three cities in eastern China during the spring of 2023. By combining an aerosol thermodynamic model with machine learning, we assessed the relative contribution of dust to aerosol pH and its impact on nitrate formation. Our results show that changes in ammonia, in both the gas and particle phases, were the main factors affecting aerosol pH, with dust particles contributing to about 7 % of the total pH variation. During dust events, high concentrations of non-volatile ions increased aerosol pH, leading to higher nitrate levels in the particle phase. Machine learning analysis revealed that extreme dust storms caused a significant change in aerosol pH, enhancing nitrate partitioning. Further simulations indicated that while reducing ammonia emissions is effective in lowering nitrate levels under normal conditions, this effect is significantly reduced in dust-affected environments. Dust particles act as a buffer, reducing the sensitivity of nitrate formation to ammonia emission reductions. These findings emphasize the need to consider dust pollution when designing strategies for controlling particulate nitrate levels and highlight the complex interactions between dust and anthropogenic emissions
How short-term change in temperature or salinity affect cellular immune parameters of three-spined stickleback, Gasterosteus aculeatus?
No full textInternational audienceReference values for the non-specific immune response of stickleback have been developed to better understand the natural variability of the immunomarkers and to increase their relevance for the detection of environmental perturbations. However, under field conditions, temperature and salinity can vary from station to station and their influence on the reference ranges of the immunomarkers should therefore be quantified. To this end, adult sticklebacks were exposed either to different temperatures (from 12 to 18 °C) or to different salinities (from 0 to 30 g/L) for 21 days after 10 days of acclimatization. The results were then projected onto reference ranges to better determine the effect of temperature and salinity on the innate immune response. With the exception of leucocyte necrosis at higher temperature and respiratory burst at lower temperature, previously established reference ranges for immunomarkers of sticklebacks were suitable when variations in temperature and salinity were tested. Finally, this study argues for the possibility of using stickleback and its immune reference range in the field regardless of temperature and salinity, due to its relatively temperature and salinity independent innate immune response. Reference ranges for immunomarkers in stickleback could be a real added value to water quality diagnosis in biomonitoring programs in variable seasonal and geographical environmental contexts. Furthermore, these results confirm the rapid adaptability of sticklebacks to different variations in temperature and salinity without affecting their immunological parameters
Comparison of modelled and experimental PM10 source contributions for mapping source-specific oxidative potential
No full textInternational audienc
Evaluation of the Performances of a Biofilm-Based Passive Sampler to Monitor Micropollutants in Wastewater
No full textInternational audiencePassive samplers (PS) are powerful tools for monitoring micropollutants due to their ability to accumulatethem over weeks or months, providing great time-representativeness. Additionally, they enable to achievelower detection limits compared to traditional methods like grab or composite sampling.However, in wastewater this sampling method faces the constraint of biofouling, which hampers thetransfer of micropollutants from water to the receiving phase. This work aims to investigate an innovativePS, the Prebio cell STEP, that uses the biofilm as a receiving phase, and thus turns biofouling into anadvantage.The operating conditions of the Prebio cell STEP were defined, through the kinetic study of biofilmgrowth, and accumulation of targeted micropollutants over several months in influents and effluents of anurban wastewater treatment plant (WWTP), to determine the optimal deployment duration of this sampler.In parallel, the performances of the Prebio cell STEP were compared with conventional 24h watercomposite sampling regarding the detection of the same micropollutants targeted for the kinetic study, andnumber of compounds detected by suspect and non -target screenings using LC/HRMS analysis.First results showed major differences in appearance of biofilm, and growth kinetics between WWTPinfluents and effluents which could be explained in relation with the various complexity of these watermatrices.Chemical analysis of the biofilm revealed the ability of the biofilm to capture micropollutants present inthe wastewater samples. The kinetic of accumulation of some of these micropollutants in the biofilm willbe presented (metals, alkylphenols, polycyclic aromatic hydrocarbons). Moreover, LC/HRMS suspect andnon-target screening highligthed the presence of a variety of micropollutants in the biofilm, includingsome not detected in wastewater composite samples