TY - JOUR
T1 - Influence of meteorological variables and air pollutants on measurements from automatic pollen sampling devices
AU - González-Alonso, M.
AU - Oteros, J.
AU - Widmann, M.
AU - Maya-Manzano, J. M.
AU - Skjøth, C.
AU - Grewling, L.
AU - O'Connor, D.
AU - Sofiev, M.
AU - Tummon, F.
AU - Crouzy, B.
AU - Clot, B.
AU - Buters, J.
AU - Kadantsev, E.
AU - Palamarchuk, Y.
AU - Martinez-Bracero, M.
AU - Pope, F. D.
AU - Mills, S.
AU - Šikoparija, B.
AU - Matavulj, P.
AU - Schmidt-Weber, C. B.
AU - Ørby, P. V.
N1 - Publisher Copyright:
© 2024 The Authors
PY - 2024/6/25
Y1 - 2024/6/25
N2 - This study examines the influence of meteorological factors and air pollutants on the performance of automatic pollen monitoring devices, as part of the EUMETNET Autopollen COST ADOPT-intercomparison campaign held in Munich, Germany, during the 2021 pollen season. The campaign offered a unique opportunity to compare all automatic monitors available at the time, a Plair Rapid-E, a Hund-Wetzlar BAA500, an OPC Alphasense, a KH-3000 Yamatronics, three Swisens Polenos, a PollenSense APS, a FLIR IBAC2, a DMT WIBS-5, an Aerotape Sextant, to the average of four manual Hirst traps, under the same environmental conditions. The investigation aimed to elucidate how meteorological factors and air pollution impact particle capture and identification efficiency. The analysis showed coherent results for most devices regarding the correlation between environmental conditions and pollen concentrations. This reflects on one hand, a significant correlation between weather and airborne pollen concentration, and on the other hand the capability of devices to provide meaningful data under the conditions under which measurements were taken. However, correlation strength varied among devices, reflecting differences in design, algorithms, or sensors used. Additionally, it was observed that different algorithms applied to the same dataset resulted in different concentration outputs, highlighting the role of algorithm design in these systems (monitor + algorithm). Notably, no significant influence from air pollutants on the pollen concentrations was observed, suggesting that any potential difference in effect on the systems might require higher air pollution concentrations or more complex interactions. However, results from some monitors were affected to a minor degree by specific weather variables. Our findings suggest that the application of real-time devices in urban environments should focus on the associated algorithm that classifies pollen taxa. The impact of air pollution, although not to be excluded, is of secondary concern as long as the pollution levels are similar to a large European city like Munich.
AB - This study examines the influence of meteorological factors and air pollutants on the performance of automatic pollen monitoring devices, as part of the EUMETNET Autopollen COST ADOPT-intercomparison campaign held in Munich, Germany, during the 2021 pollen season. The campaign offered a unique opportunity to compare all automatic monitors available at the time, a Plair Rapid-E, a Hund-Wetzlar BAA500, an OPC Alphasense, a KH-3000 Yamatronics, three Swisens Polenos, a PollenSense APS, a FLIR IBAC2, a DMT WIBS-5, an Aerotape Sextant, to the average of four manual Hirst traps, under the same environmental conditions. The investigation aimed to elucidate how meteorological factors and air pollution impact particle capture and identification efficiency. The analysis showed coherent results for most devices regarding the correlation between environmental conditions and pollen concentrations. This reflects on one hand, a significant correlation between weather and airborne pollen concentration, and on the other hand the capability of devices to provide meaningful data under the conditions under which measurements were taken. However, correlation strength varied among devices, reflecting differences in design, algorithms, or sensors used. Additionally, it was observed that different algorithms applied to the same dataset resulted in different concentration outputs, highlighting the role of algorithm design in these systems (monitor + algorithm). Notably, no significant influence from air pollutants on the pollen concentrations was observed, suggesting that any potential difference in effect on the systems might require higher air pollution concentrations or more complex interactions. However, results from some monitors were affected to a minor degree by specific weather variables. Our findings suggest that the application of real-time devices in urban environments should focus on the associated algorithm that classifies pollen taxa. The impact of air pollution, although not to be excluded, is of secondary concern as long as the pollution levels are similar to a large European city like Munich.
KW - Air pollutants
KW - Automatic pollen sampling
KW - Environmental conditions
KW - Weather
UR - http://www.scopus.com/inward/record.url?scp=85192238215&partnerID=8YFLogxK
U2 - 10.1016/j.scitotenv.2024.172913
DO - 10.1016/j.scitotenv.2024.172913
M3 - Article
C2 - 38697521
AN - SCOPUS:85192238215
SN - 0048-9697
VL - 931
JO - Science of the Total Environment
JF - Science of the Total Environment
M1 - 172913
ER -