Available online at www.sciencedirect.com
ScienceDirect Procedia Engineering 178 (2017) 42 – 52
16thConference on Reliability and Statistics in Transportation and Communication, RelStat’2016, 19-22 October, 2016, Riga, Latvia
Development of Interactive Monitoring System for Urban Environmental Impact Assessment of Transport System Anton Pashkevicha*, Marina Beliakovab, Alexander Ivanovb, Alari Purjua a
b
Tallinn University of Technology, Ehitajate tee 5, Tallinn 419086, Estonia Nizhny Novgorod State University of Architecture and Civil Engineering, Ilyinskaya Str. 65, Nizhny Novgorod603950, Russia
Abstract All countries worldwide have a global problem with environmental pollution. One of its main reasons is a negative impact of transport. In such situation, research, assessment and monitoring of this negative environmental impact are the best tools, which could support decision-making process concerning approaches and measures to reduce or to eliminate its consequences and, thereby, could help to address this challenge. This research paper considers emissions of the motor transport system as an air pollution source as well as describes an issue of its assessment and monitoring, especially, in the large cities. Unfortunately, traditional air monitoring systems are unable to present a complex spatial structure of pollution fields for different pollutants, especially, in real time. To solve this problem, an interactive monitoring system was proposed and tested on the example of Nizhny Novgorod (Russian Federation). This system contains a calculation module, which is based on the actual information about traffic flows and weather conditions. It allows to estimate concentrations of chosen pollutants and population health risk in the chosen monitoring points in online mode. The proposed system can be applied to any place with the similar initial situation. © 2017 2017The TheAuthors. Authors. Published by Elsevier © Published by Elsevier Ltd. Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of the 16th International Conference on Reliability and Statistics in Peer-review underand responsibility of the scientific committee of the International Conference on Reliability and Statistics in Transportation Communication. Transportation and Communication Keywords:interactive monitoring, environmental pollution, traffic congestion, Gaussian model
* Corresponding author. E-mail address:
[email protected]
1877-7058 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of the International Conference on Reliability and Statistics in Transportation and Communication
doi:10.1016/j.proeng.2017.01.058
Anton Pashkevich et al. / Procedia Engineering 178 (2017) 42 – 52
1. Introduction Road transport is one of the main sources of greenhouse gas (GHG) emissions as well as air and noise pollution in the cities today. To understand importance and actuality of this issue it is necessary to pay attention to the main transport strategic document in the European Union: The White Paper 2011 on Transport “Towards a competitive and resource efficient transport system” (White Paper, 2011). Between numbers of global and local problems as well as goals mentioned in this document it is possible to point out following of them concerning relations between the transport sector and the environment. First of all, unfortunately, the transport system is not still sustainable: its negative impacts on environment, economy and society are quite large. Secondly, transport sector is the biggest producer of greenhouse gas emissions (GHGs) comparing to other branches of economy. This sector must reduce 60% of GHGs by 2050 in comparison to the level of 1990.In the third place, it must be pointed out that urban transport is a reason of approximately 25% of CO2 emissions made by all transport. Fourthly, of course, during the last decades transport started to be cleaner, but because of increased traffic volume it is still the main source of air and noise pollution. Cities suffer a lot for this reason. That is why there are a number of suggestions and measures proposed and presented by White Paper 2011 to reduce level of their negative impact, especially, on urban areas. Although the Russian Federation is not a part of the European Union, it faces similar problems and challenges concerning an influence of transport on the environment. The reasons are also the same: growth of population transport demand and, as a consequence, increase of motorization rate, traffic intensity and traffic congestions. Thereby, it is clear that research, assessment and monitoring of negative transport impact on the environment, especially, in the cities play an important role. These tools could help to find right solutions for achieving the abovementioned goals and for solving the above-mentioned problems. They could also support a decision making process as well as a realization of measures and projects in the sphere of environmental management. As it was seen from above-declared challenges, air pollution occupies a special place in the negative environment impact. Traffic flows on the urban road network create local zones of toxicological risks, which could lead to both immediate and chronic effects. To avoid such negative consequences, it is necessary to organize a steady monitoring of air conditions in the city. Unfortunately, that traditional air monitoring systems are unable to present a complex spatial structure of pollution fields for the required list of pollutants, especially, in real time. This fact underlines the issue to create an interactive monitoring system, which allows to assess urban environmental impact of the transport system in online mode and which could be easily implemented in different towns with the similar traffic problems and conditions. 2. Traffic congestion as a source of environmental problems Currently, traffic congestion is almost normal condition on the road network in the large cities. Such situations are characterized by small driving speed, high intensity and density of traffic flow, usage of the maximum permitted road capacity and only in exceptional cases by blocking the traffic in one or more directions. Ecological consequences of traffic congestions were confirmed, for example, by research done in Sydney, Australia (Ferreira et al., 2003; Morawska et al., 2004). These long-term studies assessed degrees of air pollution and associated with them carcinogenic and non-carcinogenic risks in tunnel M5 East Motorway. Results showed that concentration of pollutant emissions during traffic congestion could increase in 100 times in comparison with the city average level. The negative impact of air pollution on public health caused by emissions of motor transport under the conditions of traffic congestions remains still not enough explored because of the extreme complexity of task assignment, especially, in metropolises. Also existing methodological materials and practical research works have not ready nature. At the same time, the experience of the USA as well as European countries during the last decades led to overestimate the contribution of different pollutants emitted by motor transport systems. In particular, an approach based on the distinction made between carcinogenic and non-carcinogenic risks was created. The second important methodological step was the expansion of the list of analysed carcinogens emitted by motor vehicles (Ministry of Health of the Russian Federation, 2004).
43
44
Anton Pashkevich et al. / Procedia Engineering 178 (2017) 42 – 52
The traditional list of pollutants in motor transport emissions, which are estimated, is quite small and incudes obligatorily only carbon monoxide CO, hydrocarbons CnHm, nitrogen oxides NOx, lead compounds, soot, sulfur dioxide SO2, formaldehyde and benzo[a]pyrene (State Environment Committee, 1999; Federal Road Department, 1995). Its expansion is expected together with the introduction of new Methodological Guideline for Population Health Risk Assessment (first of all, it will be an upgrade of the previous document). This new methodology will allow to identify the relative contributions of separate air pollutants to the established risk levels and, therefore, will give an opportunity to organize efficient and rational measures to manage risks of chronic effects. Unfortunately, issues of high sporadic concentrations and immediate toxicological outcomes do not get enough attention of researchers. In recent years 1,3-butadiene (divinyl) is included with the most toxic carcinogens produced by motor transport. For instance, 37.7% of its emissions in the USA is done by motor vehicles (National Center for Environmental Assessment; 2002). One vehicle emits from 0.007 till 0.057 g/km of 1,3-butadiene. The catalytic post-combustion eliminates efficiently this element from emissions. For this reason, its concentration is especially high in the emissions generated by old models of gasoline-engine cars as well as by diesel-engine cars. The United States Environmental Protection Agency uses in calculations the value 0.01 g/km per 1 vehicle. But the accuracy of such estimations is not particularly high. Also it is common practice to include acraldehyde in the list of the most dangerous non-carcinogens (US Environmental Protection Agency, 2003). All above-mentioned factors create a principally new conceptual situation concerning context and sense of monitoring systems. Road segments with traffic, which changes constantly by form and intensity and where traffic congestions are observed, start to be a source of dangers. These sources under the conditions of changing speed and direction of wind lead to create a complex dynamic structure of pollution fields for dozens of carcinogenic and noncarcinogenic pollutants. Existing air monitoring systems in the large cities have normally a very restricted amount of automated control stations: not more than few dozens of units per one city. Such small network are unable even approximately to show in real time a comprehensive spatial structure of pollution fields for different contaminants, which are connected with overloaded urban roads. For instance, the concentrations of pollutants directly on such streets is in several dozens of times more than in places, which are situated from these streets at about 100 meters. There is also a lot of programs and packages to estimate traffic-related emissions such as software package ZONE. For example, special study (Toşa et al., 2015) describes the integration as well as implementation of a computer program COPERT III for calculation of traffic emissions and software CUBE VOYAGER for transport modelling. The main disadvantage of all such programs and systems is their slow work. That is why it is suited for the calculation of average daily concentrations, but not for actual calculation in real time. Only monitoring system with a special calculation module could meet these challenges. Such module must allow to calculate pollution field and health risk based on information about real intensity of traffic flows and information about actual meteorological conditions. The general role of such monitoring system in this case is to check and to correct calculated fields in monitoring points. Precisely this kind of monitoring systems is considered in this research work. 3. Concept of interactive monitoring system 3.1. Concept background The main aim of the presented research is to create and to realize a concept of interactive monitoring system, which calculates actual and daily average concentrations of chosen pollutants to assess impact factors of traffic jams as well as population health risks related to them. The developed concept could be applied in the medium and large cities. To achieve the above-declared aim, it is necessary to have an amount of initial data mainly in the online mode. This required information could be divided into obligatory one and additional one. The obligatory initial information includes a speed of synchronized traffic flow as well as speed and direction of wind. To get the first required element of this list, there is a number of Internet resources which contain different information about traffic flows including their structure and speed and which data are available online in the free-of-
Anton Pashkevich et al. / Procedia Engineering 178 (2017) 42 – 52
charge form. It gives a revolutionary opportunity to use them for the calculation of emission masses, concentrations of pollutants, noise levels and proper population health risks. To develop the monitoring system presented in this paper, the data from GPS devices installed on municipal buses was used. In the Russian Federation such relevant information in the online mode exists, for example, on the website “www.doroga.tv” in digital form. This source provides also images from video cameras installed in the cities, which could help to assess additionally the online situation with traffic jams. The information concerning speed and direction of wind is available on different services gathering hydro-meteorological data, which are also available in real time, but, unfortunately, often for extra charge. Additional data required to create an interactive environmental monitoring system contains information about structure of traffic flow on certain street, about characteristics of certain street as well as about structure of vehicle fleet by production years and according to the compliance with standards EURO 0, …, 6. The first component of additional information could be taken from a special traffic measurements. The structure of traffic flow could be based on the classification of vehicles, for example, from Recommendations how to account the requirements of environmental protection for the design of roads and bridges (Federal Road Department, 1995). Characteristics of a road include such parameters as a number of lanes and traffic conditions, which could be gathered with the help of field observations. Additional traffic measurements together with a special assessment could help to get the age structure of vehicle fleet as well as compliance of vehicles with EURO standards. The general calculation scheme of the developed concept includes the following steps • Step 1: time intervals between vehicles in the traffic flow depending on speed; • Step 2: intensity of traffic flow, which is calculated based on traffic flow speed and time intervals between vehicles in the traffic jams; • Step 3: mass of emissions produced by traffic flow per unit time and per unit length of roads; • Step 4: concentrations of pollutants; • Step 5: noise level on the area adjacent to roads; • Step 6: health risks including risk of immediate toxicological effects, risk of chronic effects, risk of premature mortality and risk of non-specific chronic effects related to noise. The traffic flow theory and its simplest models to assess traffic flow intensity should help to realize Steps 1 and 2 (Silyanov, 1977; Ivanov et al., 2016). Within the scope of this research, the following pollutants will be considered: carbon monoxide – CO, hydrocarbons – CnHm, nitrogen oxides – Nox. These elements are produced by engines of road transport and were chosen, because they occupy together the largest volume in the composition of toxic substances in the exhaust gases. On the Step 3, the emission volume of each selected component is determined by using an approach to calculate the emission power (Federal Road Department, 1995). The Step 4 has an aim to estimate the concentration of each pollutant. It could be done using different models of atmospheric dispersion. Within the proposed concept of the interactive environmental monitoring system, the Gaussian model in the Pasquill-Gifford modification with 6 classes of atmospheric stability is used to calculate pollutant concentrations. This approach is based on the Gaussian plume distribution of emissions and is used as an engineering model. It is probably the oldest atmospheric dispersion model created in circa 1936. Despite its “age”, this approach is still applied by various national and international organization such as the United States Environmental Protection Agency and the International Atomic Energy Agency. Also the following assumption for this model is accepted: the emission source – traffic flow – is approximated to point sources. The main advantage of the Gaussian model is a possibility to make all calculations in an explicit form, i.e. without application of numerical methods to calculate Navier-Stokes equations. This fact reduces time expenses and, thus, gives an opportunity to use the model in online mode interactively. In addition, all calculations do not require any licensed software and could be realized in standard packages as Ms Excel or Open Office. The main disadvantage of the chosen model is its relatively low accuracy. Steps 4 and 5 concern the assessment of noise level and population health risks. These characteristics could be calculated also based on simple models. For example, the slope factor (SF) could be used to estimate carcinogenic risks (Kasyanenko, 2008)
45
46
Anton Pashkevich et al. / Procedia Engineering 178 (2017) 42 – 52
More detailed background for Steps 1, 2, 3 and 4 are described below as well as their implementation is presented. Instead of entry into the Steps 4 and 5 deeper, demo version of developed interactive environmental monitoring system will finish with the comparison of calculated actual concentrations for each pollutant with the maximum permissible concentrations (MPC) of the same substances. 3.2. Traffic flow theory The highest concentrations of pollutants appear in zones, where intensive traffic flows decelerate their movement. This is due to the fact that, in such case, an amount of vehicles per km of road increases as well as engines of vehicles work in non-optimal mode. These traffic flows can be considered as synchronized flows. It means that all cars move with the same speed chosen by driver for safety reasons. The Figures 1a and 1b show calculation results of changing the load of road lane depending on the speed of intensive synchronized flow (Silyanov, 1977; Ivanov et al., 2016).
Fig. 1. The intensity of traffic on road depending on the speed: (a) in normal conditions; (b) in conditions of clear ice.
Fig. 2. The coefficient m, which characterizes the dependence of emissions on the vehicle speed.
3.3. Air pollution dispersion model As it was mentioned above, the mass of emissions is estimated based on the power of emission, which is calculated separately for each chosen pollutant (CO, CnHm, NOx):
47
Anton Pashkevich et al. / Procedia Engineering 178 (2017) 42 – 52
⎡⎛ i ⎞ ⎛ i ⎞⎤ q = 2,06 ⋅ 10−4 ⋅ m ⋅ ⎢⎜⎜ ∑ Gig ⋅ N ig ⋅ c g ⎟⎟ + ⎜⎜ ∑ Gid ⋅ N id ⋅ cd ⎟⎟⎥ . ⎠ ⎝ 1 ⎠⎦⎥ ⎣⎢⎝ 1
(1)
Whereqis the emission power of particular component type from traffic flow on the particular road section, [g/m.sec.]; 2,06 ⋅ 10−4 is the coefficient to transform to the accepted measurement units; m is a coefficient, which allows to take into account traffic and road conditions and is accepted according to the diagram (Figure 2) depending on the average speed of traffic flow; Gig is the average operational fuel consumption for a given type or model of vehicles with gasoline engines, [l/km]; Gid is the same for vehicles with diesel engines, [l/km]; Nig is the estimated traffic intensity for each selected type of gasoline-powered vehicles, [vehicles/hour]; Nid is the same for diesel-powered vehicle, [vehicles/hour]; cg and cd are coefficients accepted for particular pollutant for gasoline and diesel engines, respectively (Table 1). Table 1. Values of coefficients cg and cd. Type of engine Type of emission gasoline-powered
diesel-powered
carbon monoxide
0.6
0.14
hydrocarbons
0.12
0.037
nitrogen oxides
0.06
0.015
As it was described above, the concentration of each pollutant is calculated based on the Gaussian model in the Pasquill-Gifford modification with 6 classes of atmospheric stability (Smith, 1995):
C=
⎡ ( y − y0 )2 ⎤ ⎧⎪ ⎡ (z − h )2 ⎤ ⎡ (z − h )2 ⎤ ⎫⎪ Q exp⎢− exp ⎢− , + − 2 ⎥⎨ 2 ⎥ ⎢ 2 ⎥⎬ 2 ⋅ π ⋅ U ⋅ σy ⋅ σz ⎣⎢ 2 ⋅ σy ⎦⎥ ⎪⎩ ⎣⎢ 2 ⋅ σz ⎦⎥ ⎣⎢ 2 ⋅ σz ⎦⎥ ⎪⎭
(2)
where C is the concentration of particular pollutant in the point with coordinates (x0,y0), [g/m3];Q is the power of continuous point source of pollution, [g/sec];Uis a wind speed at heights of h meters, [m/sec];his the effective height of pollution source, [m];σy and σz are coefficients representing horizontal and vertical dispersions, respectively; x, y, z are coordinates of pollution source;x0, y0are coordinates of calculation point. This model allows to calculate the ground level concentrations of pollutants for 6 classes of atmospheric stability. There are 2 approaches to estimate such stability: using synoptic information and using information about highaltitude changes of meteorological parameters. The first one was created by Pasquill and Meade and was based on processing of the large data amount concerning stack plumes (Slade, 1971). It was proposed to divide all possible kinds of weather conditions into 7 classes. These groups are presented in the Tables 2 and 3 and characterized by wind speed at heights of 10 meters as well as solar insolation (Romanov, 2006; Slade, 1971). The insolation degree for daytime (strong, moderate, slight) could be estimated with the help of solar altitude and size of sky part covered by clouds. If sky is clear and sun is high, the insolation is strong. If sky is clear and altitude of the sun is medium, the insolation is moderate. If sky is changeable and sun is high, the insolation is also moderate. In all another cases the insolation is slight. Within the scope of research, concept considers only 6 classes of atmospheric stability: class G will not be taken into account because of its rare weather conditions. To estimate the coefficients of dispersions σyand σz, the preferred means are the equations extracted from the empirical Pasquill-Gifford curves (Gifford, 1976). Final formulas to calculate them are gathered in the Table 4, where X is the distance downwind from pollution source to calculation point.
48
Anton Pashkevich et al. / Procedia Engineering 178 (2017) 42 – 52
Table 2. The Pasquill stability classes. Class
Stability level
Description
A
very unstable
very sunny summer weather
B
moderately unstable
sunny and warm
C
slightly unstable
partly cloudy
D
neutral
cloudy day or night
E
slightly stable
partly cloudy at night
F
moderately stable
clear night
G
very stable
clear cold night with light wind
Table 3. Meteorological conditions that define the Pasquill stability classes. Daytime Wind speed
V < =2
Nighttime
insolation strong
moderate
slight
cloudy
cloud cover > 50%
cloud cover = or < 50%
A
A-B
B
D
F
F
2
