The utilization of urban green areas has increased, but it is unclear whether urban green areas can decrease the concentration of particulate matter at an industrial complex city in Korea. We measured the extent of particulate matter (PM) reduction at a buffer green area in the Sihwa Industrial Complex. PM was measured at the industrial complex, the urban green area, and a nearby residential area from April to October 2019. PM reduction rates were highest at the urban green area in August and October, which is related to increased atmospheric mixing height and the active west wind blowing from the industrial complex to the residential area. Reduction rates of PM₁₀ and PM_2.5 at the urban green area showed the lowest values, namely 14.4% and 25.3%, respectively. The air temperature, wind speed, and humidity could affect the PM reduction rate by influencing the movement and dispersion of PM at the micro-spatiotemporal scale. These results indicated that PM concentration could be reduced by the structural change of a forest layer at a micro scale in urban green areas.

Oak is the dominant tree species in Korea. Oak pollen has the highest sensitivity rate among all allergenic tree species in Korea. A deep neural network (DNN)-based estimation model was developed to determine the concentration of oak pollen and overcome the shortcomings of conventional regression models. The DNN model proposed in this study utilized weather factors as the input and provided pollen concentrations as the output. Weather and pollen concentration data were used from 2007 to 2016 obtained from the Korea Meteorological Administration pollen observation network. Because it is difficult to prevent over-fitting and underestimation by using a DNN model alone, we developed a bootstrap aggregating-type ensemble model. Each of the 30 ensemble members was trained with random sampling at a fixed rate according to the pollen risk grade. To verify the effectiveness of the proposed model, we compared its performance with those of models of regression and support vector regression (SVR) under the same conditions, with respect to the prediction of pollen concentrations, risk levels, and season length. The mean absolute percentage error in the estimated pollen concentrations was 11.18%, 10.37%, and 5.04% for the regression, SVR and DNN models, respectively. The start of the pollen season was estimated to be 20, 22, and 6 days earlier than that predicted by the regression, SVR and DNN models, respectively. Similarly, the end of the pollen season was estimated to be 33, 20, and 9 days later that predicted by the regression, SVR and DNN models, respectively. Overall, the DNN model performed better than the other models. However, the prediction of peak pollen concentrations needs improvement. Improved observation quality with optimization of the DNN model will resolve this issue.

Green buffer (GB) zones are designed to prevent the spread of air pollutants and odors from industrial complexes (ICs) to residential areas (RAs). We analyzed changes in the concentration of particulate matter (PM) and the number of high PM pollution days for 10 years after the GB was implemented, using the National Atmospheric Environmental Research Stations 2001–2018 dataset. We also performed field measurements of PM₁₀ and PM_2.5 from February 2018 to January 2019 to analyze the PM concentrations at human breathing height throughout the GB. Before GB implementation (2001–2006), PM₁₀ in the RA was 9% higher than that in the IC. After GB zone implementation (2013–2018), PM₁₀ in the RA was 11% lower than that of the IC. Furthermore, the PM concentration in the RA (slope = ∆Concentration/∆Time, −2.09) rapidly decreased compared to that in the IC (slope = −1.02) and the western coastal area (WCA) (slope = −1.55) over the 10-year period. At PM concentrations at human breathing height, PM₁₀ and PM_2.5 in the RA were lower than those in the IC by 27% and 26%, respectively. After GB implementation, the wind speed was positively correlated but SOx was negatively correlated with the PM reduction rate at a local scale. These results show that there was a reduction of PM during and after GB implementation, implying the need for proper management of GBs and continuous measure of pollutant sources at the green buffers of industrial complexes.