Ozone Content Variability in the Ground-level Atmosphere Layer in the Mazowieckie Voivodeship, Central Poland

This publication presents an analysis of the variability of ground-level ozone in the Mazowieckie Voivodeship, Central Poland, in 2005−2012 and the forecast of ground-level ozone for the next day using artificial neural network models. The content of ground-level ozone in a given area is mainly determined by meteorological conditions and the presence of appropriate chemical compounds, i.e. ozone precursors. The average ozone mixing ratio is from 20 to 100 ppb, depending on the location of the measurement site.
   Despite its low concentration, ozone in the ground layer has a significant impact on natural environment, through the production of free radicals, shaping the greenhouse effect on the Earth and the formation of photochemical smog, to mention just a few effects. High contents of ground-level ozone may result in the occurrence of episodes that are harmful to human health.
     The analysis of ground-level ozone measurements in various time scales resulted in the following findings: (1) on an annual scale − the occurrence of a characteristic maximum in the spring-summer period and the minimum in the autumn-winter period, (2) on a daily scale − the occurrence of the highest ground-level ozone content in the afternoon and the lowest just before the sunrise, and (3) on a weekly scale − the existence of a weekend ozone phenomenon.
    The analysis of the long-term (1995-2016) ozone measurement series at Belsk gave grounds for distinguishing the three periods, representing an increase, decrease and re-increase of ozone content in the ground-level atmosphere in this locality.
     Models for forecasting the maximum 1-hour daily ozone concentration for the next day over the period of April–September 2015 were constructed using the Statistica 10 “Automatic Neural Networks” program package. The quality of the forecast based on neural models was verified for data from the same months of 2014. The results testify to the ability of the network to generalize the training-acquired knowledge for new, previously unexamined cases. A comparison of neural network modeling results with those of the Global Environmental Multiscale-Air Quality (GEM-AQ) troposphere chemistry model shows that the Unidirectional Multi-Layer Perceptron (MLP) type models applied in this study are an effective tool for the next-day ground-level ozone forecasting.

MONOGRAPHIC VOLUME

C O N T E N T S

Editorial note, ...4
Abstract, ...5
Streszczenie, ...6
1. Introduction, ...7
   1.1 How the ozone was discovered, ...7
   1.2 Occurrence and role of ozone in the atmosphere, ...7
   1.3 Time-space variability of ground-level ozone on a global scale, ...9
   1.4 Aim and scope of the work, ...11
2. Processes establishing the ozone content in the ground layer of the atmosphere, ... 13
   2.1 Chemical processes, ...13
         2.1.1 Chemical reaction cycles producing the ground-level ozone, ...13
         2.1.2 Chemical reactions leading to ozone destruction at the earth’s surface, ...16
         2.1.3 The role of NOx and VOC in the formation of ground-level ozone, ...17
         2.1.4 Ozone as a component of photochemical smog, ...19
   2.2 Physical processes, ...20
         2.2.1 Dry deposition on the earth’s surface, ...20
         2.2.2 Air transport from the stratosphere, ...21
3. Analysis of time-space variability of ozone in the ground layer of the atmosphere in the Mazowieckie Voivodeship
    over the years 2005−2012, ...22
    3.1 Variability of ozone concentration in the ground layer of the atmosphere over the years 2005−2012, ...22
    3.2 Annual variability of ozone concentration in the ground layer of the atmosphere, ...23
    3.3 Weekly variability of ozone concentration in the ground layer of the atmosphere, ...25
          3.3.1 Causes of weekend ozone phenomenon, ...25
          3.3.2 Analysis of the variability of ground-level ozone and nitrogen oxides on individual days of the week, ...26
          3.3.3 Comparative analysis of Ox content on workdays and on weekends, ...36
    3.4 Daily variability of ozone concentration in the ground layer of the atmosphere, ...38
    3.5 Long-term changes in ozone concentration in the ground layer of the atmosphere
          based on the 1995−2016 measurement series in Belsk, ...42
4. Impact of selected meteorological parameters on the variability of ground-level ozone content, ...44
    4.1 Air temperature, ....44
    4.2 Global solar radiation, ...47
    4.3 Relative humidity, ...48
    4.4 Wind speed, ...50
5. Artificial neural networks, ...51
    5.1 Structure of the artificial neural network, ...51
    5.2 How do neural networks work?, ...52
    5.3 Efficiency and usability of artificial neural networks, ...53
    5.4 Artificial neural network training, ...53
6. Artificial neural networks as a tool for forecasting the ground-level ozone concentration at selected monitoring stations
    in the Mazowieckie Voivodeship, ...54
    6.1 Research methodology, ...54
    6.2 Construction of artificial neural networks, ...56
    6.3 Architecture of artificial neural networks for predicting the maximum 1-hour ozone concentration, ...58
          6.3.1 Belsk, ...59
          6.3.2 Granica, ...59
          6.3.3 Legionowo, ...59
          6.3.4 Radom, ...60
          6.3.5 Warszawa-Ursynów, ...60
    6.4 Global sensitivity analysis, ...60
          6.4.1 Belsk, ...61
          6.4.2 Granica, ...61
          6.4.3 Legionowo, ...61
          6.4.4 Radom, ...61
          6.4.5 Warszawa-Ursynów, ...62
    6.5 Local sensitivity analysis, ...62
    6.6 Quality assessment of neural prognostic models, ...64
          6.6.1 Belsk, ...65
          6.6.2 Granica, ...67
          6.6.3 Legionowo, ...68
          6.6.4 Radom, ...69
          6.6.5 Warszawa-Ursynów, ...70
7. Forecast of the maximum 1-hour ozone concentration for the year 2014 (April−September), ...72
    7.1 Presentation of results, ...72
          7.1.1 Belsk, ...73
          7.1.2 Granica, ...74
          7.1.3 Legionowo, ...75
          7.1.4 Radom, ...77
          7.1.5 Warszawa-Ursynów, ...78
    7.2 Comparison of the results of the forecast by artificial neural networks and the GEM-AQ model, ...80
          7.2.1 Description of the GEM-AQ model, ...80
          7.2.2 Comparison results, ...82
                   7.2.2.1 Belsk, ...82
                   7.2.2.2 Granica, ...82
                   7.2.2.3 Legionowo, ...83
                   7.2.2.4 Radom, ...84
                   7.2.2.5 Warszawa-Ursynów, ...85
          7.2.3 Forecast analysis with relative error value above 50%, ...85
8. Summary and conclusions, ...88
References, ...90