The influence of impactor size cut-off shift caused by hygroscopic growth on particulate matter loading and composition measurements

• Published in: Atmospheric Environment
• Main Authors: Chen, Ying, Wild, Oliver, Wang, Yu, Ran, Liang, Teich, Monique, Größ, Johannes, Wang, Lina, Spindler, Gerald, Herrmann, Hartmut, Van Pinxteren, Dominik, Mcfiggans, Gordon, Wiedensohler, Alfred
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2018
• Subjects: Arctic
• Online Access: https://research.manchester.ac.uk/en/publications/65214d91-617a-4773-ab82-4089497aaaed; https://doi.org/10.1016/j.atmosenv.2018.09.049; https://linkinghub.elsevier.com/retrieve/pii/S1352231018306496

The mass loading and composition of atmospheric particles are important in determining their climate and health effects, and are typically measured by filter sampling. However, particle sampling under ambient conditions can lead to a shift in the size cut-off threshold induced by hygroscopic growth, and the influence of this on measurement of particle loading and composition has not been adequately quantified. Here, we propose a method to assess this influence based on κ-Köhler theory. A global perspective is presented based on previously reported annual climatological values of hygroscopic properties, meteorological parameters and particle volume size distributions. Measurements at background sites in Europe may be more greatly influenced by the cut-off shift than those from other continents, with a median influence of 10–20% on the total mass of sampled particles. However, the influence is generally much smaller (<7%) at urban sites, and is negligible for dust and particles in the Arctic. Sea-salt particles experience the largest influence (median value ∼50%), resulting from their large size, high hygroscopicity and the high relative humidity (RH) in marine air-masses. We estimate a difference of ∼30% in this influence of sea-salt particle sampling between relatively dry (RH = 60%) and humid (RH = 90%) conditions. Given the variation in the cut-off shift in different locations and at different times, a consistent consideration of this influence using the approach we introduce here is critical for observational studies of the long-term and spatial distribution of particle loading and composition, and crucial for robust validation of aerosol modules in modelling studies.