Popularna meteorologija
Problems with Using Short Time Series Weather Data in CALPUFF Simulations
Dispersion numerical models, whether using CALPUFF, AEROMD or any other, require adequate weather data time series. Usually, one year time series data is used. Why? Simple, it is assumed that one year contains approximatively 95 % of all possible weather situations in particular area. Other 5 % occur very rarely and could be assumed as extremes. Short time series of weather data can produce wrong picture which can’t be used as relevant one. To show this problem I used 13 days period, from 10th to 22nd May 2024 time series data obtained by WRF ARW meteorological mezzo-model.
The models
The WRF ARW (Weather Research and Forecasting Advanced Research WRF) model is a state-of-the-art numerical weather prediction system used for simulating atmospheric processes. It is highly versatile, capable of producing high-resolution forecasts for a wide range of spatial and temporal scales, from local weather phenomena to global climate patterns. WRF ARW is widely utilized in both research and operational meteorology due to its ability to accurately simulate complex atmospheric dynamics and improve understanding of weather-related phenomena.
The dispersion model I used was CALPUFF, developed by the American company "ENVIRON International Corporation" in collaboration with the United States Environmental Protection Agency (https://www.epa.gov/) (EPA). The model emerged from the need for a more sophisticated tool for simulating air pollutant dispersion over long distances, capable of accounting for complex atmospheric conditions and terrain characteristics. It considers changing weather conditions, including variations in wind direction and speed, enabling more accurate predictions of pollutant dispersion. It can model pollutant spread in areas with complex terrains, such as mountains and valleys, as well as coastal zones and supports the analysis of multiple types of pollutants simultaneously, including gases and particulates. CALPUFF is used for regulatory purposes, environmental impact assessments, and air quality research, making it a crucial tool for understanding and managing air pollution.
CALPUFF output can be categorized as concentration, dry flux or deposition and wet flux or deposition. Dry flux or dry deposition refers to the process by which pollutants from the air deposit onto the surface of the ground or vegetation without the presence of rain or other forms of precipitation. Wet flux or wet deposition refers to the process by which pollutants from the atmosphere deposit onto the surface of the ground or water bodies through precipitation, such as rain, snow, fog, or ice.
Emission sources
For the testing purpose, I used emissions from the two HEP (Hrvatska elektroprivreda) stacks in Zagreb. First one (EL-TO) is settled in the west and second (TE-TO) in the south-east part of Zagreb. Both are cca 200 m height, EL-TO has 10 m stack diameter while TE-TO has 6,5 m. Exit gases velocities are around 18 m/s with temperatures between 380 and 390 °C. Five species were modelled (SO2, NO2, CO2, CO and PM10).
Results
Here I present concentrations, dry and wet depositions of PM10 particles. PM10 particles are a category of particulate matter (PM) with a diameter of 10 micrometers (µm) or less. These particles are small enough to be inhaled and can pose significant health risks, particularly affecting the respiratory and cardiovascular systems.
Figure 1. PM10 concentrations
Figure 2: PM 10 dry deposition
Concentrations and dry deposition images show the pattern which follow the wind direction frequency pattern for Zagreb Maksimir station shown on Figure 3.
Figure 3. Wind direction frequency, Zagreb Maksimir.
On the other side, wet deposition picture is completely different, what was expected. Wet deposition is defined and ordered by the precipitation, rain, snow etc. When a short time series meteorological data is used, you may detect which event caused wet deposition.
Figure 4. Wet deposition in the period 10. - 22. May 2024.
On the Figure 4. is clearly visible that most of the wet deposition happened in south-west Hungary. Now I looked for the event which caused such strong deposition and found it on EUMETSAT images:
Figure 5. Intensive convective activity which caused wet deposition.
To be sure about the precipitation, I found radar precipitation image published by DHMZ, Croatian Weather Service.
Figure 6. Radar precipitation image. Image source: DHMZ.
Conclusion
Interpreting wet deposition is difficult and unrewarding because it depends on precipitation, which generally lacks a repeating spatial and temporal pattern, unlike winds. So, the only solution is to use long enough time series of weather data, at least one year, because it will include cca 95 % of possible cases in this area.
