An Approach to Locating Unknown Sources of Increased Air Emissions

We are considering determining the locations of sources with exceeding emissions of atmospheric pollutants in a given area. In the report, we present one of the options for formalizing such a problem. In our approach, we assume that there is a set of known sources within the considered area which are expected to emit a declared amount of pollutant. Additionally, there may be unidentified sources of this pollutant. Let’s suppose that the measurements at the monitoring points exceed the modeled cumulative emissions of the registered sources. In this case, we want to understand where the violation occurs. The violation could come from the registered sources that exceed their declared emissions. The violation can also be expressed in the appearance of new pollutant sources. The analysis considers the registered climatic data for the examined time intervals. Specifically, the weather stability class for different time intervals is determined using cloudiness information from an online service’s API and atmospheric pressure, air temperature, and wind velocity data in the measurements database. Furthermore, we propose identifying the suspected areas of violations through iterations of simulating increased emissions from various points in the studied area using a Gaussian plume model and comparing the results with actual data. The report presents a general scheme of this approach and calculations obtained in its first two stages. The accuracy of the implemented mathematical model is evaluated. Conclusions are drawn regarding the advantages and limitations of the Gaussian plume model for the considered tasks.