Estimation of required PM 10 emission source reduction on the basis of a 10-year period data Mirjana Perišić & Andreja Stojić & Svetlana Stanišić Stojić & Andrej Šoštarić & Zoran Mijić & Slavica Rajšić Received: 2 June 2014 /Accepted: 21 August 2014 # Springer Science+Business Media Dordrecht 2014 Abstract The main purpose of this study was to assess the exceedance frequency of the EU (50 μgm -3 ) and US EPA (150 μgm -3 ) daily PM 10 limit values, as well as the emission source reduction required to comply with air quality stan- dards. The analysis was based on a 10-year (2003–2013) data set obtained for the Belgrade (Serbia) urban area, which is seriously affected by pollution. General probability (Pearson 5, lognormal and Weibull) and extreme value (two-parameter exponential and Gumbel’ s) distributions were used to fit data and to estimate the number of exceedances. With the use of best fitting general probability distribution and rollback equa- tion, the required reduction was determined to be in the range of 12 to 98 % for some years, although for others the total reduction of moderate sources would be insufficient to meet air quality regulations. The required emission reduction of strong sources for the whole monitoring period was estimated by means of extreme value distributions at about 45 %. The aforementioned methods can be successfully applied for prediction of pollutant loads, as well as for estimation of required reductions. As pollution in Belgrade is largely seen as the result of outdated technology, emission reduction mea- sures must primarily rely on technological improvements and advances. Keywords PM 10 . Probability density function . Extreme values . Emission source reduction . Serbia Introduction Particulate matter is seen as the pollutant attracting the greatest concern in terms of adverse effects on human health. The relationship between short- and long-term exposure to air- borne coarse particles (particulate matter with aerodynamic diameter of less than 10 μm, PM 10 ) and detrimental health effects measured as the daily number of deaths, hospital admissions and emergency room visits for cardiovascular and respiratory diseases has been widely studied (Pisoni and Volta 2009; Sanhueza et al. 2009; Araujo 2011). A compre- hensive evaluation of these research findings provides consis- tent evidence that chronic exposure to dangerous levels of PM 10 is associated with premature human mortality and in- creased prevalence of respiratory illnesses such as asthma (Cakmak et al. 2012; Rohr et al. 2014). Additionally, acute short-term exposure to PM 10 pollution causes oxidative stress and systematic and pulmonary inflammatory response in the airways of healthy subjects (Behndig et al. 2011). In developed countries, PM 10 emissions have various sources including the following: diesel exhaust, industrial activities, heat and power generation plants, agricultural pro- cesses, brake and tyre wear on unpaved road surfaces, burning activities and crushing and grinding operations (Juda-Rezler et al. 2011). Whereas the emphasis was initially placed on local sources, the influence of long-range transport and local M. Perišić (*) : A. Stojić : Z. Mijić : S. Rajšić Institute of Physics Belgrade, University of Belgrade, Pregrevica 118, Belgrade 11080, Serbia e-mail: mirjana@ipb.ac.rs A. Stojić e-mail: andreja.stojic@ipb.ac.rs Z. Mijić e-mail: zoran.mijic@ipb.ac.rs S. Rajšić e-mail: slavica.rajsic@ipb.ac.rs S. Stanišić Stojić Singidunum University, Danijelova 32, Belgrade 11010, Serbia e-mail: sstanisic@singidunum.ac.rs A. Šoštarić Institute of Public Health Belgrade, Bulevar Despota Stefana 54, Belgrade 11000, Serbia e-mail: andrej.sostaric@zdravlje.org.rs Air Qual Atmos Health DOI 10.1007/s11869-014-0292-5