Constructing a pollen proxy from low-cost Optical Particle Counter (OPC) data processed with Neural Networks and Random Forests

Pollen allergies affect a significant proportion of the global population, and this is expected to worsen in years to come. There is demand for the development of automated pollen monitoring systems to progress from conventional manual sampling techniques. Low-cost Optical Particle Counters (OPCs) are conventionally used for measuring particulate matter and have attractive advantages of real-time high time resolution data and affordable costs. Our study asks whether low-cost OPC sensors be used for meaningful monitoring airborne pollen and utilises data from the EUMETNET Autopollen ADOPT – COST Action Intercomparison Campaign (2021) in Munich, Germany.

We employ a variety of methods, including supervised machine learning techniques, to construct pollen proxies from hourly-average OPC data and evaluate their performance. The most successful methods are supervised machine learning Neural Network (NN) and Random Forest (RF) methods, trained from pollen concentrations collected from a Hirst-type sampler. These perform significantly better than using a simple particle size-filtered proxy or a Positive Matrix Factorisation (PMF) source apportionment pollen proxy. We developed a collection of models - varying by model type, input features (including particle size and meteorological data) and target variable (i.e. total pollen vs selected pollen taxa) - and evaluated their suitability for constructing a pollen proxy. The results show that such models can construct useful information on pollen from OPC particle size data with Spearman correlation coefficients up to 0.85 and coefficients of determination up to 0.67. Model-constructed proxies demonstrated an ability to distinguish high pollen events with promising results (F1 Scores up to 0.83) and to follow monthly and diurnal trends in pollen. We discuss the suitability of OPCs for monitoring pollen and offer advice for future progress. We demonstrate an attractive alternative for automated pollen monitoring that could provide valuable and timely information to the benefit of pollen allergy