Skip to main content
BioAirMet logo BioAirMet logo

WP1: Traceable calibration of automatic bioaerosol monitors with respect to particle number concentration

The counting efficiency of automatic bioaerosol monitors is expected to depend on particle size, instrument design and particle detection method. In general, the counting efficiency decreases with increasing particle size due to higher particle losses in the sampling line or detection chamber. Moreover, instruments that are equipped with an aerosol concentrator will exhibit a much higher counting efficiency than those without one. To make measurements reliable and comparable, bioaerosol monitors need to be calibrated against traceable methods to establish suitable correction factors.
The aim of this work package is to further develop and validate traceable methods for the measurement of airborne bioaerosol particle number concentration in order to determine the counting efficiency of automatic bioaerosol monitors with a target expanded uncertainty of ≤15 %. The methods to further develop and validate are: two laboratory-based methods (light scattering and particle flow visualisation) and a third (field standard) will also be suitable for field measurements.

  • Task 1.1 aims to further develop the first method (light scattering), which is based on a reference optical particle counter. Particles cross a beam of laser light and scatter light, which is detected by a photodiode.

  • Task 1.2 aims to validate the second method (particle flow visualisation), which is based on 3D particle imaging and tracking, where particles are illuminated with a sheet of light. Using tomographic reconstruction, the 3D location of each particle can be accurately reconstructed and the number of particles in the illuminated volume determined.

  • Task 1.3 aims to further develop the third method (field standard), which is a combination of a portable pollen generator with a custom-made optical particle counter. This will be developed as a portable field standard.

  • Task 1.4 aims to perform an inter-comparison of the above-mentioned methods and to provide a document to CEN/TC 264/WG 39 for consideration as an input into the new documentary standard on real-time bioaerosol monitoring.

WP2: Creation of pollen and fungal spore datasets and development of good practices for machine learning algorithms applied to pollen classification

The aim of this work package is to develop metrics and actionable approaches (training and validation) to give quantifiable confidence (via uncertainty evaluation and sensitivity analysis) in the results produced by machine learning (ML) classification algorithms for bioaerosols. This will include consideration of the specific challenges associated with the field deployment of these algorithms and an evaluation of the uncertainties associated with the ML algorithm outputs. This information will be presented to CEN/TC 264/WG 39 (sampling and analysis of airborne pollen grains and fungal spores).

  • Task 2.1 aims to create datasets for ML with well-defined pollen and fungal spore aerosols.

  • Task 2.2 aims to develop a set of metrics for the assessment of the quality of pollen and spore data.

  • Task 2.3 aims to investigate the use of transfer learning for the rapid development of region-specific classification ML algorithms.

  • Task 2.4 aims to use outputs from Tasks 2.1 to 2.3, to produce guidance for the assessment of ML pollen taxa classification algorithms. The guidance will also be presented to CEN/TC 264/WG 39 (sampling and analysis of airborne pollen grains and fungal spores).

WP3: Standardisation of the data output, interface and metadata of the automatic bioaerosol monitors; development of guidelines on data storage, handling and distribution

The aim of this work package is to establish a comprehensive standardisation framework for the data output, interface and metadata generated by automatic bioaerosol monitors. The successful implementation of standardised practices (e.g., data storage, handling and distribution) will ensure consistency, compatibility and seamless integration of data across various monitoring systems.

  • Task 3.1 aims to review the solutions, and proposals, from the bioaerosol monitoring community for the formatting of Level 0 data (raw data) generated by automatic bioaerosol monitors.

  • Task 3.2 aims to review the solutions, and proposals, from the bioaerosol monitoring community for the formatting of Level 1 (processed and calibrated) data, and with the addition of results from Task 3.1 (Level 0 – raw data), to formulate recommendations regarding specific data fields, metadata descriptors, units of measurement, uncertainty reporting guidelines and the overall structure of processed data.

  • Task 3.3 aims to review practices related to distributed data, database management (data storage), data accessibility and version control (handling) within the context of automatic bioaerosol monitors to develop a clear and effective set of recommendations/guidelines for better organisation, accessibility and traceability of data across stakeholders (e.g., network operators).