Emission Data Uncertainty in Urban Air Quality Modeling—Case Study

Abstract

Air pollution models are often used to support decisions in air quality management. Due to the complexity of the forecasting system and difficulty in acquiring precise enough input data, an environmental prognosis of air quality with an analytical model of the air pollution dispersion is burdened with a substantial share of uncertainty, especially as regards urban areas. To ignore the uncertainty in the modeling would lead to incorrect policy decisions, with further negative environmental and health consequences. This paper presents a case study which shows how emission uncertainty of air pollutants generated by the industry, traffic, and the municipal sector relates to concentrations measured at receptor points. The computational experiment was implemented in the Warsaw metropolitan area, Poland. The main source of this adverse environmental impact is the transportation system, including the transit traffic. The Monte Carlo technique was used for assessing the key uncertainty factors. Several types of pollution species that are characteristic for the urban atmospheric environment (e.g., PM 10 , PM 2.5 , NO x , SO 2 , Pb) were included in the analysis. The results show significant spatial variability of the modeled uncertainty. The reason of this variability is discussed in detail. It depends not only on the category of the emission source but also on the contributing emission sources and their quantity.

Air quality modeling for Warsaw agglomeration

“…The uncertainty of the model predictions, which is mainly related to the input data, such as emission inventory or meteorological forecast (Sax and Isakov 2003, Park et al 2006, Maxim and van der Sluijs 2011 is also an important factor in decision making. Quantifi cations of emission related uncertainty discussed in Holnicki and Nahorski (2015) show that high uncertainty values were associated with the cases of strongly dominating contribution of one individual source or one category of emission sources. Within this study such domination occurs in the central zone for NO X (domination of line sources), or in peripheral districts for B(a)P and PM 2.5 (domination of area sources).…”

“…Emission fi eld in an urban area usually represents concentration of a large number of sources in the study domain, which vary in technological parameters, emission characteristics, composition of emitted compounds, and also the assigned uncertainty (Holnicki and Nahorski 2015). To take into account specifi c technological characteristics of the different emission sources, the total emission fi eld was split down into the following categories: point (high/low), area, and line (mobile) sources.…”

“…Similar estimates can be found in (Trapp 2010) for PM 10 , NO 2 and SO 2 in Mazovian Voivodship or in (Elbir 2003) for SO 2 in Izmir. Table 5 shows the computed and calculated concentrations at nine monitoring sites, as well as the values of FAC2 index (see also Holnicki and Nahorski 2015, Juda-Rezler 2010, Trapp 2010 for the above six basic compounds. Similar assessments, for the limited number of the measurement stations, are shown in (Holnicki et al 2016 (Table 5), as compared with the limit value, 10,000 μg/m 3 .…”

A Tethered And Navigated Air Blimp (TANAB) for observing themicroclimate over a complex terrain

“…It can also provide boundary conditions and validation datasets to support ABL simulations using Computational Fluid Dynamics (CFD) or mesoscale modelling. Furthermore, the environmental data collected can be used to develop and validate microclimate and dispersion models in complex urban environments (Krayenhoff and Voogt, 2007;Bueno et al, 2012;Holnicki and Nahorski, 2015;Aliabadi et al, 2017).…”

Development of a high-resolution spatial inventory of greenhouse gas emissions for Poland from stationary and mobile sources

“…Therefore, it was Silesian province Warsaw assumed that the uncertainties in the proxy data values are the same as for the statistical data used. It was also assumed that relative uncertainties of disaggregated activity data at the level of emission sources are ffiffi ffi n p times higher than those in the coarse base area, where n is the number of emission sources of certain category during disaggregation of activity data, similar to Hogue et al (2018); see also Holnicki and Nahorski (2015) for an analysis in another problem. Based on these input uncertainties, we estimated the distributions of the emissions at the level of emission sources using the Monte Carlo method and calculated the mean values and the lower and upper limits of the 95% confidence intervals as uncertainty ranges (as suggested by IPCC (2001)).…”