Instrumental Odour Monitoring at a Metalworking Facility: Comparison of Two IOMS, Analysis with Chemical Concentration Data and Citizen Science

Industrial activities involving metal forming and heat treatment produce complex and highly variable emissions due to the diversity of processes, raw materials, and operating conditions involved. These emissions can be originate both from controlled (stack) and uncontrolled (fugitive) sources. As a result, these emissions can significantly compromise ambient air quality and generate odour nuisances that negatively impact the daily lives of people living in nearby residential areas. Persistent or intense odours may contribute not only to discomfort and complaints, but also to symptoms such as headaches, nausea, or respiratory irritation, especially among sensitive populations. Moreover, monitoring systems can contribute to building and maintaining trust between industrial operators, local authorities, and the communities affected by these type of emissions. This study presents the field implementation of two Instrumental Odor Monitoring Systems (IOMS)—WT1 (Ellona, France) and MSEM 3200 (Sensigent, USA)—at the perimeter of a metalworking facility that uses mineral oils. The systems were deployed in collaboration with ARPA Lombardia near a residential area considered sensitive to odour nuisances. Stack emissions were sampled and analysed using gas chromatography–mass spectrometry (GC/MS) and dynamic olfactometry following EN 13725:2022. The analyses revealed a complex chemical profile consisting of Volatile Organic Compounds (VOCs), ketones, alcohols, and sulphur-based compounds, with odour concentrations at the source reaching up to 6896 ouE/m³. These results were used to calibrate the IOMS units through a structured training protocol based on diluted odour samples. Temporal monitoring showed consistent weekday VOC peaks in the morning (08:30–10:00; 0.02–0.036 ppm) and evening (19:00–20:00; 0.032–0.074 ppm), with reduced concentrations on weekends. However, VOC levels alone did not fully account for odour episodes, highlighting the influence of specific odour-active compounds not captured by total VOC measurements. The WT1 unit (metal-oxide, electrochemical and PID sensors) recorded elevated hydrogen sulphide (H₂S) concentrations in the late afternoon (up to 0.036 ppm), while the MSEM 3200—with its 32 heterogeneous sensors (metal-oxide, electrochemical, polymer-based and PID)—detected more persistent odour signals, particularly during night shifts. Moderate odour levels (251–1000 ouE/m³) were frequently recorded throughout the day, while high-intensity peaks (>1000 ouE/m³) predominantly occurred in the evening and nighttime hours. To complement the instrumental data, a citizen science campaign was conducted to involve local residents in real-time odour reporting through a dedicated mobile application. This participatory approach allowed researchers to compare subjective human perception with objective sensor readings, thereby improving data interpretation and increasing community engagement in environmental monitoring. The results show the effectiveness of using multiple IOMS devices to distinguish background odours from transient, high-intensity odour events in complex industrial settings. Furthermore, the integration of citizen-reported data with instrumental monitoring strengthens the reliability of odour impact assessments and demonstrates the value of community participation in addressing environmental challenges.