TY - GEN
T1 - Forecasting PM2.5 Concentrations with uEMEP and EMEP4PL for Poland
AU - Kryza, Maciej
AU - Werner, Malgorzata
AU - Denby, Bruce Rolstad
AU - Mu, Qing
AU - Sawiński, Tymoteusz
AU - Remut, Arkadiusz
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2023/1/2
Y1 - 2023/1/2
N2 - In this work we use the EMEP4PL model in combination with the high-resolution uEMEP model. EMEP4PL is an application of the EMEP MSC-W chemistry transport model for the area of Poland. The model uses two nested domains. The parent domain covers Europe with 12 km × 12 km. The nested domain is centered over Poland with a 4 km × 4 km grid. uEMEP is based on Gaussian modelling principles and allows downscaling of the EMEP4PL model results to very high spatial resolution using the local fraction approach. In this work, we have applied the uEMEP model to forecast hourly PM2.5 concentrations for the area of SW Poland with a spatial mesh of 250 m × 250 m. Both the EMEP4PL and uEMEP results are compared with two measuring networks. First, we compare the results with data gathered at six reference stations operated by the Chief Inspectorate of Environmental Protection. Second, we use data from the newly established LIFE/MappingAir network. This network uses low-cost sensors and provides data from 25 sites located in SW Poland. This network provides more data from residential areas, which are often the hot-spots for PM2.5 concentrations. We have shown that the uEMEP forecasts have smaller bias and higher index of agreement if compared to EMEP for the test period from April to end of June 2021.
AB - In this work we use the EMEP4PL model in combination with the high-resolution uEMEP model. EMEP4PL is an application of the EMEP MSC-W chemistry transport model for the area of Poland. The model uses two nested domains. The parent domain covers Europe with 12 km × 12 km. The nested domain is centered over Poland with a 4 km × 4 km grid. uEMEP is based on Gaussian modelling principles and allows downscaling of the EMEP4PL model results to very high spatial resolution using the local fraction approach. In this work, we have applied the uEMEP model to forecast hourly PM2.5 concentrations for the area of SW Poland with a spatial mesh of 250 m × 250 m. Both the EMEP4PL and uEMEP results are compared with two measuring networks. First, we compare the results with data gathered at six reference stations operated by the Chief Inspectorate of Environmental Protection. Second, we use data from the newly established LIFE/MappingAir network. This network uses low-cost sensors and provides data from 25 sites located in SW Poland. This network provides more data from residential areas, which are often the hot-spots for PM2.5 concentrations. We have shown that the uEMEP forecasts have smaller bias and higher index of agreement if compared to EMEP for the test period from April to end of June 2021.
KW - Air quality modelling
KW - PM2.5 forecasting
KW - uEMEP model
UR - http://www.scopus.com/inward/record.url?scp=85148024937&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-12786-1_27
DO - 10.1007/978-3-031-12786-1_27
M3 - Conference Proceeding
AN - SCOPUS:85148024937
SN - 9783031127854
T3 - Springer Proceedings in Complexity
SP - 193
EP - 197
BT - Air Pollution Modeling and its Application XXVIII
A2 - Mensink, Clemens
A2 - Jorba, Oriol
PB - Springer Science and Business Media B.V.
T2 - 38th International Technical Meeting on Air Pollution Modeling and its Application, ITM 2021
Y2 - 18 October 2021 through 22 October 2021
ER -