Neural network forecasting of air pollutants hourly concentrations using optimised temporal averages of meteorological variables and pollutant concentrations

The new method for the forecasting hourly concentrations of air pollutants is presented in the paper. The method was developed for a site in urban residential area in city of Zagreb, Croatia, for four air pollutants (NO₂, O₃, CO and PM₁₀). Meteorological variables and concentrations of the respective pollutant were taken as predictors. A novel approach, based on families of univariate regression models, was employed in selecting the averaging intervals for input variables. For each variable and each averaging period between 1 and 97 h, a separate model was built. By inspecting values of the coefficient of correlation between measured and modelled concentrations, optimal averaging periods for each variable were selected. A new dataset for building the forecasting model was then calculated as temporal moving averages (running means) of former variables. A multi-layer perceptron type of neural networks is used as the forecasting model. Index of agreement, calculated for the entire dataset including the data for model building, ranged from 0.91 to 0.97 for the respective pollutants. As suggested by the analysis of the relative importance of the input variables, different agreements for different pollutants are likely due to different sources and production mechanisms of investigated pollutants. A comparison of the new method with more traditional method, which takes hourly averages of the forecast hour as input variables, showed similar or better performance. The model was developed for the purpose of public-health-oriented air quality forecasting, aiming to use a numerical weather forecast model for the prediction of the part of input data yet unknown at the forecasting time. It is to expect that longer term averages used as inputs in the proposed method will contribute to smaller input errors and the greater accuracy of the model.     

北京空气重污染红色预警期间污染物与气象因子特征

揭示空气重污染红色预警期间污染物与气象因子的变化特征对空气质量预报和污染减排措施评估具有重要参考价值。利用大气污染和气象观测资料,研究了北京2015年11-12月空气重污染红色预警时期污染物浓度、气候特征及气象因子对空气质量影响。结果表明,PM2.5在大气颗粒物中占有较大比重,为首要空气污染物;在重污染峰值时段,城郊PM2.5与PM10比值（R）相差不大,可达0.9以上,空气呈均匀混合的高PM2.5浓度特征,而空气质量较好时城区R值明显高于郊区;研究时段气候特征与历史同期相比有明显差异,其中平均风速偏小19%,平均气温偏高0.23 ℃,气温日较差减小,而多次小型降水增加了空气湿度,导致相对湿度值偏高40%,垂直方向上的逆温层或等温层则加剧了空气重污染的形成和发展,重污染过程中的红色预警措施明显降低了颗粒物浓度;风速与污染物浓度呈指数相关,城郊风速分别低于2.0和2.5 m·s-1时,空气质量较差、污染物浓度随风速升高快速下降,而当城郊风速大于2.0和2.5 m·s-1时变化特征则相反;相对湿度与污染物浓度呈幂相关,相对湿度在65%左右为空气质量特征发生变化的转折点;由于气温日较差存在季节变化,其与空气质量相关关系不太显著。     

Impacts of some meteorological parameters on SO2 and TSP concentrations in Erzurum, Turkey

The air pollution is the one of the most important environmental problems in Erzurum, situated in the eastern of Turkey, during winter periods. The unfavorable climate as well as the city’s topography, and inappropriate urbanization cause serious air pollution problems. The air pollutant concentrations in a city have a close relationship with its meteorological parameters. In the present study, the relationship between daily average total suspended particulate (TSP) and sulphur dioxide (SO₂) concentrations with meteorological factors, such as wind speed, temperature, relative humidity, pressure and precipitation, in 1995–2002 winter seasons was statistically analyzed using the stepwise multiple linear regression analysis. According to the results obtained through analysis, higher TSP and SO₂ concentrations are strongly related to colder temperatures, lower wind speed, higher pressure system and weakly lower precipitation and higher relative humidity. The statistical models of SO₂ and TSP including meteorological parameters gave R² of 0.74 and 0.88, respectively. Furthermore, the correlation between the previous day’s SO₂, TSP concentrations and actual concentrations of these pollutants on that day was investigated and found as 0.84 and 0.53, respectively. In order to develop this model, previous day’s SO₂ and TSP concentrations were added to the equations. The new model for SO₂ enhanced considerably (R² = 0.92), but for TSP new model was not enhanced (R² = 0.89).     

湖北省能源消费CO2与大气污染物协同减排研究

以2015年为基准年,基于拓展的STIRPAT模型预测2025年湖北省能源消费CO2和主要大气污染物排放量.通过设置基准(记为BAU)情景、低碳(记为LC)情景和强化低碳(记为ELC)情景3种控制情景,测算CO2和主要大气污染物的减排量,并运用污染物减排量交叉弹性法评价了CO2减排对主要大气污染物的协同效应.结果表明,...     

Kosetice, Czech Republic--ten years of air pollution monitoring and four years of evaluating the origin of persistent organic pollutants

The regional observatory Kosetice is a central European background station. Unique continuous monitoring from 1988 on is held here. POP (persistent organic pollutant) concentration values of air samples from Kosetice taken between 1996 and 2005 were statistically processed. Values of Czech ambient air quality standards were not exceeded. Concentrations of polycyclic aromatic hydrocarbons reached two maxima, in 1996 and 2001-2002. Polychlorinated biphenyls concentrations reached the highest values in 1997 and 1998 and hexachlorocyclohexanes concentrations in 1998. DDTs, hexachlorobenzene and pentachlorobenzene were analysed as well. Long-range transport of pollutants between 2002 and 2005 was evaluated using the Potential Source Contribution Function hybrid receptor model. Indicated potential source areas of PCBs coincide with many well-known urban and industrialised areas, while the indicated potential source areas of HCHs and DDTs coincide with many agricultural and/or forested regions and the potential source areas of HCB comprise all land use types.     

Effects of low concentration biodiesel blend application on modern passenger cars. Part 1: Feedstock impact on regulated pollutants, fuel consumption and particle emissions

Five biodiesels from different feedstocks (rapeseed, soy, sunflower, palm, and used fried oils) blended with diesel at 10% vol. ratio (B10), were tested on a Euro 3 common-rail passenger car. Limited effects (−2% to +4%) were observed on CO₂ emissions. CO and HC emissions increased between 10% and 25% on average, except at high speed - high power where emissions were too low to draw conclusions. NOx emissions increased by up to 20% for two out of the five blends, decreased by up to 15% for two other blends, and remained unchanged for one blend. Particulate matter (PM) was reduced for all blends by up to 25% and the reductions were positively correlated with the extent of biodiesel saturation. PM reductions are associated with consistent reductions in non-volatile particle number. A variable behaviour in particle number is observed when volatile particles are also accounted.     

Soil burdens of persistent organic pollutants - Their levels, fate and risk. Part I. Variation of concentration ranges according to different soil uses and locations

Detailed soil screening data from the Czech Republic as a typical Central European country are presented here. Determination of a wide selection of organic and inorganic pollutants as well as an assessment of specific soil parameters allowed us to study the soil contamination in relation to the land use and soil properties. While HCHs and HCB were found at highest levels in arable soils, the higher concentrations of PCDDs/Fs, PCBs, PAHs and DDTs were observed in high altitude forest soils. Concentrations of these compounds strongly correlated with the soil organic carbon content. Several possible reasons have been suggested for the observed higher concentrations in mountain forest soils but the impact of each of these influencing factors remains to be identified. An inventory of the soil contamination is needed as a first step in our effort to estimate an extent to which the secondary sources contribute to the enhanced atmospheric levels of POPs.     

Correlations between mutagenic activity of organic extracts of airborne particulate matter, NOx and sulphur dioxide in southern Germany: results of a two-year study

Goals, Scope and Background Among other substances, sulphur dioxide (SO2), nitric oxide (NO) and nitrogen dioxide (NO2) are parameters which are routinely measured to describe basic air quality. Organic extracts of airborne particulate matter contain mutagenic chemical compounds of different origins. The aim of the study was to find correlations between routine monitoring data and mutagenic activity of organic extracts of simultaneously drawn samples.Methods Specimens were collected over a period of two years at 8 sampling sites in south-west Germany. Simultaneously, concentrations of NO, NO2, and SO2 were measured on-line within the framework of the official air monitoring network of Baden-Württemberg, Germany. Dust samples were collected for biotesting using high volume air samplers equipped with glass fibre filters. After sampling was completed, filters were extracted and samples were prepared for biological testing. Mutagenic activity was tested by means of the plate incorporation assay (Ames test) using S. typhimurium TA98 and TA100 tester strains. During the first year of the study, all tests have been performed with and without metabolic activation. Additionally, a series of tests has been performed in parallel with TA98 and TA98NR.Results and Discussion Comparison of Ames test data obtained with and without metabolic activation indicates no statistically significant difference between both methods. Therefore, during the second year of the study, all tests have been performed without metabolic activation. Average yearly activities at the sampling sites were between 1 und 27 Revertants per m³ (Rev/m³). High activities were preferably found at congested sites (Karlsruhe, up to 95 Rev/m³). However, peak values of over 100 Rev/m³ were found in other places where pollution by traffic is significantly lower. The reason for these high level values is not evident. Tests performed using TA98NR tester strain indicate a significant share (average 31%) of compounds requiring activation by nitroreductase for mutagenic activity. Average mutagenic activity can be correlated to routine monitoring parameters. Comparison of averaged data for particular sampling sites indicates significant correlation between nitric oxide and mutagenic activity in TA98 (r²=0.90), while correlation between nitrogen dioxide (0.84) or sulphur dioxide (0.52) and mutagenic activity is weaker. For TA100, correlations are generally weaker than for TA98. Comparison of data for mutagenic activity and routine monitoring data of distant sites being sampled simultaneously shows parallel behaviour.Conclusions Results from this study show that mutagenic activity can be compared to seasonal and local variations of gaseous indicator air pollutants. Tester strain TA98 generally shows the best correlations. Although pollution by particle-bound mutagenic substances is significantly higher during the cold season than during summer on average, mutagenic activity of airborne dust is not a continuous effect. During winter, peak levels as well as low pollution periods can occur. Even during winter time mutagenic activity can reach very low levels typical for summertime. Comparison of results for distant sampling sites where samples have been collected simultaneously indicate that “classical” indicators of air pollution and bacterial mutagenicity of organic extracts from airborne particulate matter are influenced by connected effects. Seasonal trend of mutagenic activity, in particular, is similar to the concentrations of nitrogen oxide. NO is a strong indicator for vehicle exhaust gases. It is concluded that the average mutagenic activity at particular sites can be estimated using NO concentrations as an indicator.     

The water-soluble fraction of potentially toxic elements in contaminated soils: relationships between ecotoxicity, solubility and geochemical reactivity

To better understand the impacts posed by soil contamination to aquatic ecosystems it is crucial to characterise the links between ecotoxicity, chemical availability and geochemical reactivity of potentially toxic elements (PTE’s) in soils. We evaluated the adverse effects of water extracts obtained from soils contaminated by chemical industry and mining, using a test battery including organisms from different trophic levels (bacteria, algae and daphnids). These tests provided a quick assessment of the ecotoxicity of soils with respect to possible adverse effects on aquatic organisms although the ecotoxicological responses could be related to the solubility of PTE’s only to a limited extent.The analysis of results of bioassays together with the chemical characterisation of water extracts provided additional relevant insight into the role of conductivity, pH, Al, Fe, and Mn of soil extracts on toxicity to organisms. Furthermore, an important conclusion of this study was that the toxicity of extracts to the aquatic organisms could also be related to the soil properties (pH, Org C and Fe_ox) and to the reactivity of PTE’s in soils which in fact control the soluble fraction of the contaminants.The combined assessment of ecotoxicity in water fractions, solubility and geochemical reactivity of PTE’s in soils provided a more comprehensive understanding of the bioavailability of inorganic contaminants than ecotoxicological or chemical studies alone and can therefore be most useful for environmental risks assessment of contaminated soils.