274
M. Kryza et al.
respiratory and cardiovascular diseases and increased mortality [4], not mentioning
deterioration of life quality. The air quality forecasts are becoming more and more
frequently used to support local authorities in reducing the population exposure.
In this work we present the results of the air quality forecasting system developed
within the LIFE-APIS/PL project [6] for the area of Poland using the WRF-Chem
model. The study period covers 11–25 February 2017, when high PM2.5 and PM10
concentrations were observed, but not always accurately forecasted by the system.
The main aim of this work is to test if we can improve the performance of the forecasting system by assimilation of the conventional surface and radiosonde meteorological
data using the Grid point Statistical Interpolation (GSI) system.
43.2 Data and Methods
43.2.1 Study Period
In this study we analyse the period of 11–25 February 2017. During the first half
of this period, central Europe was under the influence of a high-pressure system, with the centre located over Poland and Belarus. Calm wind conditions were
favourable for accumulation of atmospheric pollutants. Moreover, low air temperatures, observed especially over the night, lead to increased emission from low elevated
residential sources. Hourly concentrations of PM2.5 and PM10 frequently exceeded
100 µg m
−3 at number of measuring sites. After 17th of February, advection of warm
air masses from the south of Europe was observed. This lead to higher wind speeds
and air temperatures exceeded the long term mean values for February. This resulted
in a rapid decrease of PM2.5 and PM10 concentrations in Poland.
43.2.2 WRF-Chem Model Configuration
The Weather Research and Forecasting (WRF) model with chemistry (WRF-Chem),
version 3.9, was used in this study [1]. The model was configured using two one-way
nested domains with spatial resolutions of 12 km × 12 km and 4 km × 4 km. Both
domains had 35 vertical levels. RADM2 chemical mechanism with the Kinetic PreProcessor (KPP) for gas chemistry and GOCART scheme for aerosols were applied.
For the outer domain covering Europe, the model uses the TNO MACC III emission inventory [3]. The same emission database was also used for the nested domain
except for Poland, where high-resolution emission inventory, developed by the Chief
Inspectorate of Environmental Protection (CIEP) was applied.
Two sets of forecasts were calculated. They both were initialised each day at
00 UTC. The base model runs were using the data from the Global Forecast System
(GFS) as meteorological initial and boundary conditions. For the second set of the
M. Kryza et al.
respiratory and cardiovascular diseases and increased mortality [4], not mentioning
deterioration of life quality. The air quality forecasts are becoming more and more
frequently used to support local authorities in reducing the population exposure.
In this work we present the results of the air quality forecasting system developed
within the LIFE-APIS/PL project [6] for the area of Poland using the WRF-Chem
model. The study period covers 11–25 February 2017, when high PM2.5 and PM10
concentrations were observed, but not always accurately forecasted by the system.
The main aim of this work is to test if we can improve the performance of the forecasting system by assimilation of the conventional surface and radiosonde meteorological
data using the Grid point Statistical Interpolation (GSI) system.
43.2 Data and Methods
43.2.1 Study Period
In this study we analyse the period of 11–25 February 2017. During the first half
of this period, central Europe was under the influence of a high-pressure system, with the centre located over Poland and Belarus. Calm wind conditions were
favourable for accumulation of atmospheric pollutants. Moreover, low air temperatures, observed especially over the night, lead to increased emission from low elevated
residential sources. Hourly concentrations of PM2.5 and PM10 frequently exceeded
100 µg m
−3 at number of measuring sites. After 17th of February, advection of warm
air masses from the south of Europe was observed. This lead to higher wind speeds
and air temperatures exceeded the long term mean values for February. This resulted
in a rapid decrease of PM2.5 and PM10 concentrations in Poland.
43.2.2 WRF-Chem Model Configuration
The Weather Research and Forecasting (WRF) model with chemistry (WRF-Chem),
version 3.9, was used in this study [1]. The model was configured using two one-way
nested domains with spatial resolutions of 12 km × 12 km and 4 km × 4 km. Both
domains had 35 vertical levels. RADM2 chemical mechanism with the Kinetic PreProcessor (KPP) for gas chemistry and GOCART scheme for aerosols were applied.
For the outer domain covering Europe, the model uses the TNO MACC III emission inventory [3]. The same emission database was also used for the nested domain
except for Poland, where high-resolution emission inventory, developed by the Chief
Inspectorate of Environmental Protection (CIEP) was applied.
Two sets of forecasts were calculated. They both were initialised each day at
00 UTC. The base model runs were using the data from the Global Forecast System
(GFS) as meteorological initial and boundary conditions. For the second set of the