National Technical University of Athens (NTUA)
Επιπτώσεις λειτουργίας αεροδρομίων στο ατμοσφαιρικό περιβάλλον
Abstract
dc:descriptionAviation is a key element of the modern lifestyle, but its impact on air quality is important, especially in the vicinity of large airports. In this work, these impacts are determined through detailed computational approaches around the three major airports of Greece, namely Athens International Airport, Thessaloniki Airport and Heraklion Airport. The introductory part of the thesis presents coherent bibliographic information of the aviation sector at the international, European and national levels. In Greece, the highest number of air traffic was recorded in 2009 and approached 460 thousand flights. In the same year, the highest traffic was recorded in each of the airports, with 210,147 flights at Athens International Airport, 50,238 flights at Thessaloniki airport and 44,842 flights at Heraklion airport. The basic concepts of air pollution, including its impacts on public health and ecosystems, are presented in the introductory part, along with information on the state of air quality at four levels. At the global level, problems arise mainly in China, India and some parts of Africa, while concentrations of air pollutants at European level keep reducing over time. At the level of the Greek territory, carbon monoxide is very low, nitrogen oxides and sulfur oxides exceed limit values mainly in western Attica and Ptolemaida, while in the case of the particulate matter, there are extended exceedances in most of the southern and western part of the country as a result of both pollution transport from sources outside Greece and the physical emission of particles from unvegetated soil and from the sea. At the level of the three major cities of Greece, namely Athens, Piraeus and Thessaloniki, where about half of the country's population lives, air quality has been improving over the years since the 1980s. Around the big airports, the only available data come from the air quality monitoring system in Mesogia, and show, with regard to NO2 which is the main air pollutant, mixed trends, with significant improvements over time but also visible deteriorations during last decade. Aviation emissions of air pollutants relate to the overall course of a flight, but those with the most pronounced impacts are the emissions that occur near the ground, justifying the focus of interest of this work on airport regions. The main sources of air pollutant emissions at aerodromes are aircraft movements, which are to some extent standardized by means of the "landing-take-off cycle", as well as a series of ground sources such as auxiliary power units, ground service equipment and brake/tire residues from aircraft. These emissions affect ecosystems and other aspects of the environment, but the main concern comes from their potential impact on public health. A recent review article shows that this impact is likely, but it is very difficult to draw general conclusions, given the great diversity of aerodromes and the dependence of their air quality impacts from the specificities of their air traffic, aircraft mix and ground activities, but also from the location, geography and meteorology of the area. The interest of several modern studies focuses on the pollutants considered the most dangerous for human health, namely fine particles, and dangerous gaseous substances such as formaldehyde and tropospheric ozone. The main air pollutants of aviation, i.e. CO, NOx, PM10 and SOx, show concentrations distinctly higher than the background at short distances, but within more distant residential areas they tend to exhibit concentrations similar to background. Because of this similarity, it is difficult to derive quantitative conclusions about the effect of pollutants emitted by the airport on the public health of the population at residential areas. In the proximity of aerodromes, the concentration of tropospheric ozone is decreasing because, due to its photochemical features, part of O3 is destroyed; however, at a greater radius, the contribution of aerodromes to the formation of O3 leads to an increase in the effects of this pollutant on public health. The mechanism between the emissions of air pollutants by the operation of an airport and the impacts of these substances on public health, as well as on the environment, is the dispersion of these pollutants in the air of the region; the concentrations of air pollutants which occur via this dispersion are the core issue of this work. The way of studying the concentrations was selected following an assessment of the methods in this field. These methods are either based on measurements or on computational approaches. The first category is based on performing field measurements of concentrations, using either fixed or mobile equipment; those methods excel in accuracy but lack spatial coverage, and have difficulties illuminating the causal path between emissions from different sources (airport activities, road traffic, heating, industry etc.) and overall measured concentrations. Computational methods rely on a variety of computational simulation systems that interconnect the emissions from each particular source with its portion in the final concentration; those methods contain uncertainties related to both the internal assumptions of the models used and the lack of detailed input data. However, the use of computational methods is clearly encouraged by modern European and national legislation. In this context, we selected the most acknowledged system of computational simulation of air pollutant emissions from airport activities and the concentrations that occur by the dispersal of these emissions in the affected area. This system is the EDMS, which is the regulatory model proposed by the US EPA, i.e. to be used for compliance checking of existing or planned airports with air quality limits. The special character of the approach adopted in this work is the implementation of the EDMS in the most realistic way, i.e. using real meteorological data of hourly frequency, a real mix of aircraft and real time distributions of their movements. To ensure a sufficient margin of safety into the results of the calculations, the aircraft movements of 2009 were used as input data; during this year, the number of aircraft movements was the largest ever recorded, both in Greece as a total and in each of the three airports of this work. The main computational part of this work is complemented by two additional parts, with a view to extending the scope beyond the air pollution assessments around airports and air quality assessments in the affected areas. The first one is the comparison between computational estimates and measurements in the Athens Airport case where detailed air quality measurements are available. The second extension is the sensitivity analysis of the EDMS applied to the system as a whole, in order to examine its response under varied inputs. The results of applying this approach to the three largest airports in Greece (Athens, Thessaloniki and Heraklion) include first the calculation of emissions and then the determination of concentrations of air pollutants in the vicinity of each of these airports. In terms of emissions, the results can be used in monitoring the impact of airports on the micro- and macro-scale of air quality. In addition, emissions calculated by EDMS are found to be in good agreement with the monitoring data of the actual operation of Athens Airport. There is also some convergence with the results of other work, resulting in an expected average of NOx of 4.55 kg per aircraft movement. EDMS results are also consistent with the national emission inventory figures concerning aviation. In terms of concentrations, the results of their computational determination for each of the pollutants studied construct a realistic detailed air quality map around each of the three largest airports in the country. These maps can be useful to determine the extent and the levels of the airport's impact on the air quality of its vicinity, examining whether limit values are exceeded and pinpointing the locations of high pollution levels. The maps are also useful for assessing the importance of air quality impacts from airport operations against other environmental threats in the area. Finally, the calculation and mapping of air pollutant concentrations around major Greek airports can be used to design appropriate measures of reducing them, where necessary, based on objective and detailed images of atmospheric pollution, as well as to document the lack of urgent need for action where such action is not expected to have any significant benefit. In addition, the analysis of computational results, via approaches such as comparison with field measurements and sensitivity analysis, has the potential to indicate both airport participation in the overall levels of air quality and to reveal aspects of the behavior of the model that are not well known yet. The results of the computational determination of concentrations around Athens International Airport show that CO, PM2.5, PM10 and SO2 are expected to stay well below the limit values. In contrast, NO2 concentrations indicate potential exceedances of the limit value for average hourly concentration (200 μg/m3), but not of the limit value for annual mean concentration. These exceedances are expected to occur in areas of significant size but with a limited frequency, since their occurrence requires a high rate of air movements along with specific meteorological parameters which lead to a high stability of the surface layer and the entrapment of NOx at low altitudes. Although these conditions are not quite uncommon during the July-August period, they lay far from being the norm; so, the eighteen times that legislation allows to exceed 200 μg/m3 NO2 seem sufficient, even marginally, to cover the likely occurrences of high NOx concentrations. At Thessaloniki Airport, the concentrations calculated with EDMS for CO, PM2.5, PM10 and SO2 are also well below the limit values. NO2 concentrations, however, include potential exceedances of the limit value for average hourly concentration, expected to occur in small areas and at a very limited frequency, since their occurrence requires specific conditions for high air traffic and special NOX entrapment meteorology, which are not expected to occur more than eighteen times during the July-August period. However, it is worth noting that, despite the fact that air traffic at Thessaloniki airport is about four times lower than that of AIA, exceedances of the NOx limit are not avoided, they simply extend to much smaller areas. At the Heraklion airport, EDMS estimates that concentrations of CO, PM2.5, PM10 and SO2 are also expected to move well below limit values. NO2 concentrations are high, but the potential exceedances of the limit value for average hourly concentration are rare, much fewer than at Thessaloniki airport and within the eighteen times allowed for such exceedances. Despite the fact that air traffic between Heraklion and Thessaloniki airports is the same, the frequency and extent of NO2 exceedances in Heraklion are significantly lower, reflecting the influence of the other parameters - apart from flight numbers - shaping the final concentrations. As the analysis of air pollution around the three airports has been focused on the most adverse situations, i.e. those with the higher concentrations, there is a point in asking how often these situations are expected to occur. Processing the results for NOx concentration for each of the 1,488 hours of July and August, we found that the relative frequency of very high levels (such as the ones close or even above the limit values) are very rare, despite the intense use of the airports and the consequent intense emissions. Medium or high levels (but well below the limit values) of NOx have been found to occur more frequently, but levels from 0 to 150 μg/m3 cumulatively hold for more than 90% of the processed hours. Comparing the distribution of relative frequencies of the three airports, we found that NOx levels at Thessaloniki are considerably higher than at Athens, despite of the higher emissions of the latter airport. This is attributed to the difference in dispersion conditions: At Thessaloniki, the higher frequency of low-speed winds and of calm weather, compared to Athens, delays the dispersion of emitted NOx and leads to higher concentrations from lower emissions. This explanation, i.e. the dominance of dispersion conditions over the emission rates in defining the final concentrations, is confirmed by the case of Heraklion. There, similar emission rates as at Thessaloniki lead to considerably lower concentrations because of the higher frequency of medium and high wind speeds. Turning to the comparison of EDMS results with concentration measurements, it is first noted that these concern the Mesogia region around AIA and include measurements from four stations. As far as CO is concerned, the general finding is that measurements at all stations, although in absolute numbers very small (0,1 – 1 μg/m3) are quite higher than the results of the calculations, which for the vast majority of times are close to zero. This means that throughout the Mesogia region there are concentrations of CO background that cannot be attributed to the operation of Athens International Airport. A similar image is derived from the SOx figures. Background concentrations from sources of the local anthropogenic environment unrelated to the airport, although small in absolute numbers (e.g. 5 – 9 μg/m3 for the Spata station) lead to measurements that are significantly larger than the results of the calculations, which are well below 1 μg/m3. Even more pronounced is the similar phenomenon of high background values in the case of PM10 concentrations, where a range of phenomena irrelevant from the airport lead to measurements of 20 – 50 μg/m3, while the results of EDMS are very low and the majority of them are zero. In regard with NOx concentrations, the findings from the comparison of measurements and computational results are more complex. The first conclusion is that the measurements are systematically different from the EDMS results, both in the time series of the hourly values and in their statistical distribution. The divergence picture at the concentration level is in contrast with the good convergence observed at the emission level; this happens due to a number of reasons, such as the model's stochastic features, some small differences in hourly and daily flight distribution patterns, and especially the significant distance between the source (airport) and the monitoring stations (4 9,35 km), which allows for significant variability of the dispersion conditions within this distance, while the model operates on the assumption that the dispersion conditions are constant throughout the study area. Finally, the most significant part of the deviations is attributed to the influence of sources of the local anthropogenic environment, irrespective of the airport, such as road traffic close by the stations. The second conclusion is that EDMS results on NOx concentrations are generally smaller than the measurements. This is not due to a systematic tendency of the model to underestimate air pollution levels; to a greater extent it is attributed to the contribution of other NOx sources beyond the airport’s operation. And, this is not an unusual finding; underestimations of computational results in relation to field measurements are quite common, even in recent works where, by means of reverse modeling, it is ascertained that in order for the calculated concentrations to rise to the level of the measured ones, it is necessary to seriously increase the emission rate introduced in the dispersion model from 14 to 55%. The third conclusion relates to the role of NOx as the main pollutant at the airport and, consequently, to the origin of a significant portion of the field measured concentrations from the airport. Because of this origin, there is a prospect of better correlation between measurements and EDMS results than other pollutants, which is confirmed by the statistical analysis. The fourth conclusion on NOx relates to the finding that all approximations between computational results and hourly measurements are detected at low wind speeds of less than 3 m/s. This is because, for higher wind speeds, the EDMS calculated dispersion is quite intense and leads to near zero NOx concentrations at the long distance (~ 4 km) of the Spata station, so whatever the station records are, they come from local sources and not from the airport. The final part of the thesis is the sensitivity analysis of the model, i.e. the study of how changes in input conditions cause changes in the results. The parameters that affect the results in a complex way are meteorological, which affect the AERMOD module, while parameters affecting emissions such as flights number and aircraft types are expected to have a more linear influence. In this context, the sensitivity of EDMS to the five main meteorological parameters was analyzed. The parameter with the strongest effect on the results is wind speed. Small changes of this parameter cause significant changes in peak concentrations and the extent of pollution levels. From an environmental point of view, this means that even a light breeze is sufficient to disperse the air pollutants of an airport and reduce ground concentrations to levels well below limit values; on the contrary, apnea or very low wind speeds (less than 1 m/s) are a necessary and often sufficient condition for the appearance of high concentrations. At the same time, the identification of the high sensitivity of the concentration results in wind speeds demonstrates the critical role of this parameter in the modeling exercises used in environmental impact assessment studies. The time of the day strongly influences the effects of concentration calculations. This is a remarkable finding, in the sense that this parameter is rarely included amongst those that influence the results of dispersion models. Variation in the results, depending on whether they occur for morning, afternoon or evening hours, is due to the transition of the atmospheric surface boundary layer from convective to stable stratified state over the 24-hour period. For this reason, during most hours with sunshine, the calculated concentrations are very low, almost zero, due to the dispersion caused by the vertical movements in the convective surface layer, despite the low shear wind intensity. A practical consequence of this finding is, when conducting environmental impact assessments, where emphasis is placed to the worst-case scenarios, this part of the 24-hour period when convective layer appears, should be accounted for with a smaller weighting factor. Ambient temperature only affects the calculation results of the concentrations under certain conditions. In particular, the temperature change cause visible change in the concentration results calculated by the EDMS only during the transition times from a convective to a stratified surface layer and vice versa; during the remaining hours, temperature changes lead to a very small change in the results. During the transition time of the surface layer, the effect of temperature is not simple: it is characterized by a general tendency to decrease the concentration with the temperature increase, interrupted by fluctuations. The cloud cover percentage affects the concentration results in differentiated ways during the afternoon hours in relation to the morning hours. In the afternoon, as cloud cover increases, both the maximum value and the spatial spread of the concentration decrease, while in the morning cloud cover change does not cause visible changes in the calculation results of the of the concentrations. The cloud ceiling height has little effect on the results of the calculation of concentrations. This is a positive coincidence because cloud ceiling height measurements are rarely available; however, they can be substituted by a rough empirical value, with no significant impact on the results of simulated concentrations with EDMS.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Koulidis, Alexandros
- Κουλίδης, Αλέξανδρος
Subjects
dc:subject × 17- Ποιότητα αέρα
- Ατμοσφαιρική ρύπανση
- Περιβαλλοντικές επιπτώσεις
- Αεροδρόμια
- Μοντελοποίηση
- EDMS
- Air quality
- Atmospheric pollution
- Environmental impacts
- Airports
- Modelling
- Περιβαλλοντική Μηχανική
- Επιστήμες Μηχανικού και Τεχνολογία
- Πράσινη και Αειφόρα επιστήμη και Τεχνολογία
- Environmental Engineering
- Engineering and Technology
- Green and Sustainable Science and Technology
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/57593