Prediction of airborne pollen and sub-pollen particles for thunderstorm asthma outbreaks assessment.

When exposed to convective thunderstorm conditions, pollen grains can rupture and release large numbers of allergenic sub-pollen particles (SPPs). These sub-pollen particles easily enter deep into human lungs, causing an asthmatic response named thunderstorm asthma (TA). Up to now, efforts to numerically predict the airborne SPP process and to forecast the occurrence of TAs are unsatisfactory. To overcome this problem, we have developed a physically-based pollen model (DREAM-POLL) with parameterized formation of airborne SPPs caused by convective atmospheric conditions. We ran the model over the Southern Australian grass fields for 2010 and 2016 pollen seasons when four largest decadal TA epidemics happened in Melbourne. One of these TA events (in November 2016) was the worldwide most extreme one which resulted to nine deaths and hundreds of hospital patient presentations. By executing the model on a day-by-day basis in a hindcast real-time mode we predicted SPP peaks exclusively only when the four major TA outbreaks happened, thus achieving a high forecasting success rate. The proposed modelling system can be easily implemented for other geographical domains and for different pollen types.

Assessment of PM10 pollution level and required source emission reduction in Belgrade area.

The aim of this study was to assess PM10 pollution level and estimate required source emission reduction in Belgrade area, the second largest urban center in the Balkans. Daily mass concentrations and trace metal content (As, Cd, Cr, Mn, Ni, Pb) of PM10 were evaluated for three air quality monitoring sites of different types: urban-traffic (Slavija), suburban (Lazarevac) and rural (Grabovac) under the industrial influence, during the period of 2012-13. Noncompliance with current Air Quality Standards (AQS) was noticeable: annual means were higher than AQS at Slavija and Lazarevac, and daily frequency threshold was exceeded at all three locations. Annual means of As at Lazarevac were about four times higher than the target concentration, which could be attributed to the proximity of coal-fired power plants, and dust resuspension from coal basin and nearby ash landfills. Additionally, levels of Ni and Cr were significantly higher than in other European cities. Carcinogenic health risk of inhabitants' exposure to trace metals was assessed as well. Cumulative cancer risk exceeded the upper limit of acceptable US EPA range at two sites, with Cr and As as the major contributors. To estimate source emission reduction, required to meet AQS, lognormal, Weibull and Pearson 5 probability distribution, functions (PDF) were used to fit daily PM10 concentrations. Based on the rollback equation and best fitting PDF, estimated reduction was within the range of 28-98%. Finally, the required reduction obtained using two-parameter exponential distribution suggested that risks associated to accidental releases of pollutants should be of greater concern.

Forecasting of VOC emissions from traffic and industry using classification and regression multivariate methods.

In this study, advanced multivariate methods were applied for VOC source apportionment and subsequent short-term forecast of industrial- and vehicle exhaust-related contributions in Belgrade urban area (Serbia). The VOC concentrations were measured using PTR-MS, together with inorganic gaseous pollutants (NOx, NO, NO2, SO2, and CO), PM10, and meteorological parameters. US EPA Positive Matrix Factorization and Unmix receptor models were applied to the obtained dataset both resolving six source profiles. For the purpose of forecasting industrial- and vehicle exhaust-related source contributions, different multivariate methods were employed in two separate cases, relying on meteorological data, and on meteorological data and concentrations of inorganic gaseous pollutants, respectively. The results indicate that Boosted Decision Trees and Multi-Layer Perceptrons were the best performing methods. According to the results, forecasting accuracy was high (lowest relative error of only 6%), in particular when the forecast was based on both meteorological parameters and concentrations of inorganic gaseous pollutants.

Determination of O(3), NO (2), SO (2), CO and PM (10) measured in Belgrade urban area.

O(3), NO(2), SO(2), CO and PM(10) concentrations, simultaneously determined for the first time in Belgrade urban area in the autumnal period of 2005, are presented. The obtained results display similar behaviour of SO(2), NO(2), CO, PM(10) opposite from that of O(3). The weekend effect was also investigated showing diminution of average daily concentrations of SO(2), NO(2), PM(10) and CO for 72, 40, 37 and 42% respectively, and increase of the average daily concentration of O(3) for 56%. Influence of meteorological conditions on observed concentration levels was studied, too. The observed influence of wind speed on the O(3) nightly concentration levels was analyzed pointing to the phenomena of O(3) transport during episodic measurements. To make an identification of possible pollution sources and analyse the influence of meteorological parameters on pollution levels, air back trajectories for high level concentrations episodes were calculated and analysed. A multivariate receptor modelling (Principal Component Analysis, Cluster Analysis) has been applied to a set of data in order to determine the contribution of different sources. It was found that the main principal components, extracted from the air pollution data, were related to gasoline combustion, oil combustion and ozone transport.