Elsevier

Atmospheric Research

Volume 167, 1 January 2016, Pages 216-223
Atmospheric Research

Characterization of hourly NOx atmospheric concentrations near the Venice International Airport with additive semi-parametric statistical models

https://doi.org/10.1016/j.atmosres.2015.07.023Get rights and content

Highlights

  • Hourly NOx near Venice Airport is characterized by additive semi-parametric models.

  • Four model specifications of increasing complexity are employed.

  • The road traffic is an important source of NOx in the monitored area.

  • Impact of aircraft ground operations on air quality is statistically significant.

Abstract

An additive modeling approach is employed to provide a statistical description of hourly variation in concentrations of NOx measured in proximity of the Venice “Marco Polo” International Airport, Italy. Differently from several previous studies on airport emissions based on daily time series, the paper analyzes hourly data because variations of NOx concentrations during the day are informative about the prevailing emission source. The statistical analysis is carried out using a one-year time series. Confounder effects due to seasonality, meteorology and airport traffic volume are accounted for by suitable covariates. Four different model specifications of increasing complexity are considered. The model with the aircraft source expressed as the NOx emitted near the airport is found to have the best predictive quality. Although the aircraft source is statistically significant, the comparison of model-based predictions suggests that the relative impact of aircraft emissions to ambient NOx concentrations is limited and the road traffic is the likely dominant source near the sampling point.

Introduction

In the last decade, several studies have been focused on the impact of rapidly growing aviation activities on the environment (e.g., Herndon et al., 2004, Yu et al., 2004, Carslaw et al., 2006, Schurmann et al., 2007, Agrawal et al., 2008, Westerdahl et al., 2008, Dodson et al., 2009, Hu et al., 2009, Lee et al., 2009, Amato et al., 2010, Kurniawan and Khardi, 2011, Mazaheri et al., 2011, Zhu et al., 2011, Hsu et al., 2012, Hsu et al., 2013, Hudda et al., 2014, Masiol and Harrison, 2014, Keuken et al., 2015). During the late 1980s and early 1990s, special attention was paid to the effects of nitrogen oxide (NOx: the sum of NO2 and NO normalized to the molecular weight) emissions on the formation of tropospheric O3 (Lee et al., 2009). Today, aviation activities are included into the Community Scheme of Allowances of Greenhouse Gases by the 2008/101/CE Directive of European Parliament and the Council of December 19th, 2008, amending the 2003/87/EC Directive. Although the aviation source provides a relatively small contribution to global air quality worsening and greenhouse gas emissions compared to other sources (FAA, Aviation and Emissions — A Primer, 2005), several studies indicate that the airport activities can have a serious impact on the air quality locally through both the emissions of aircrafts and ground support vehicles (Carslaw et al., 2006, Schurmann et al., 2007, Westerdahl et al., 2008, Dodson et al., 2009, Zhu et al., 2011, Hsu et al., 2012, Hsu et al., 2013, Hudda et al., 2014, Keuken et al., 2015).

The mean composition of plume emitted by aircraft engines is 70% CO2, something less than 30% H2O and the remainder consisting of NOx, CO, SOx, unburned or partially combusted hydrocarbons, particulates and other trace components (FAA, Aviation and Emissions — A Primer, 2005, Agrawal et al., 2008, Kinsey et al., 2011, Kurniawan and Khardi, 2011, Mazaheri et al., 2013). Although at ground level the aircraft spends most of the time at low power thrust setting, both the fuel consumed/minute and the NOx emissions greatly increase during takeoff (Herndon et al., 2004, EMEP/EEA, 2013). Specifically, NO2 is a significant fraction of NOx at idle condition for the turbofan engines that are usually employed in airliners. Vice versa, when the power thrust setting increases, the respective emission index decreases and the main constituent of NOx becomes NO (ACRP, 2008, Chapters 5–6).

From the point of view of human health, NO2 is the most troubling nitrogen oxide form (World Health Organization, 2000, World Health Organization, 2006). Short-term exposure to NO2 levels well above the typical atmospheric concentration (20–90 μg m−3) may cause defective pulmonary function, while long-term exposure to lower concentrations may cause abnormal effects in blood, spleen, liver and lungs. The negative health effects comprise cell modifications at tracheobronchial and pulmonary regions, variations in metabolism and reduction of immunological defenses against bacterial and viral infections.

Usually, road traffic was identified as an important source of NOx in urban areas (Colvile et al., 2001, Mavroidis and Chaloulakou, 2011, Lawson et al., 2011) and previous studies on NOx concentration close to Los Angeles International (Yu et al., 2004) and Heathrow International (Carslaw et al., 2006) airports highlighted that the dominant effect of the road traffic makes it difficult to detect and quantify the contribution of aircrafts to the local concentrations of NOx. Moreover, such a characterization is further complicated by the effect of weather conditions that play a key role in air pollutant dispersion. Indeed, the aircraft contributions vary in magnitude when moving a few hundred meters from the airstrip (Carslaw et al., 2006).

The aim of this study is twofold. Firstly is to illustrate statistical methods that can be fruitfully employed to evaluate the impact of aircraft NOx emissions on monitored hourly NOx concentrations in proximity to airports. Secondly is to forecast the impact associated with a potential increase of the number of flights. Although direct observations about the road traffic source are not available, the statistical approach considered in this paper is helpful to understand the different factors that contribute to the NOx emissions around an airport. The methodology is illustrated through data collected at the Venice International Airport, managed by company SAVE S.p.A. The data derive from a collaborative project of Ca' Foscari University of Venice, SAVE S.p.A and Ente Zona Industriale di Porto Marghera on monitoring the main atmospheric variables and the major pollutants near the airstrips of the Venice International Airport.

The atmospheric emission inventory of the Veneto region (Italy is subdivided into twenty administrative regions further subdivided in provinces; Venice is the capital of the Veneto region) for the years 2007/8 (INEMAR, 2013a), assessed as indicated by EMEP/EEA (2009), points out that the Venice International Airport is located in the area of the Venice province characterized by the higher NOx emissions. According to the document, the contribution of the macrosector “exhaust emissions from road transport”, that includes mopeds, motorcycles, passenger cars, buses, light and heavy duty vehicles, represents about 37% of the total emissions (INEMAR, 2013b). In the same document, the air traffic activities related to LTO cycle below about 1000 m belong to the macrosector “other mobile sources and machinery” together with military, railways, inland waterways and maritime activities that play an important role in the Venice Lagoon. The overall emission due to all these sources is quantified to be about 21%. The relative contributions for the remaining macrosectors are: Combustion in energy and transformation industries 24%; combustion in the manufacturing industry 7%; production processes 6%; non-industrial combustion plants 4%; other 1%.

The road traffic is expected to have a significant impact on the air quality worsening near the Venice International Airport given the presence of a large motorway and a series of busy roads that connect the airport to the nearby towns.

Section snippets

The sampling site

The Venice International Airport is located near the Venice Lagoon, a complex and sensitive ecosystem (Ros and Nesto, 2005, Pranovi et al., 2008, Coccioni et al., 2009) in North-Eastern Italy between the Po Valley, commonly considered the most industrialized district of Italy, and the Adriatic Sea. As discussed in Valotto et al. (2014), the area is influenced by several anthropogenic emission sources, such as the urban areas of Mestre and Venice, the industrial area of Porto Marghera, private

Selection of the best model specification

AIC statistics indicate that model specifications including a measure of air traffic volume significantly improve the quality of the model fit with respect to the baseline specification. The AIC for the baseline model specification (Eq. 2) is equal to − 3393.205. Although the first three panels of Fig. 2a are very similar, the second model specification (Eq. 5) with the additional aircraft variable yields a drop in AIC to value − 3407.665. A further improvement is obtained by the third model

Conclusions

This paper illustrates statistical methodology for the characterization of NOx concentrations in the proximity of an international airport. Differently from several papers focused on the study of pollutant concentrations around airports, we chose to work with hourly data because the variations throughout the day are informative on the concomitant sources.

In order to evaluate the airport contribution, we suggested employing additive semi-parametric statistical modeling with covariates accounting

Acknowledgments

The Authors are grateful to Professor Giancarlo Rampazzo for helpful comments. The kind permission of SAVE S.p.A. and Ente Zona Industriale del Porto di Marghera to collect the data analyzed in this paper is also acknowledged.

References (51)

  • H.H. Hsu et al.

    The relationship between aviation activities and ultrafine particulate matter concentrations near a mid-sized airport

    Atmos. Environ.

    (2012)
  • H.H. Hsu et al.

    Contributions of aircraft arrivals and departures to ultrafine particle counts near Los Angeles International Airport

    Sci. Total Environ.

    (2013)
  • M.P. Keuken et al.

    Total and size-resolved particle number and black carbon concentrations in urban areas near Schiphol airport

    Atmos. Environ.

    (2015)
  • J.S. Kurniawan et al.

    Comparison of methodologies estimating emissions of aircraft pollutants, environmental impact assessment around airports

    Environ. Impact Assess. Rev.

    (2011)
  • A.R. Lawson et al.

    Prediction of traffic-related nitrogen oxides concentrations using structural time-series models

    Atmos. Environ.

    (2011)
  • D.S. Lee et al.

    Aviation and global climate change in the 21st century

    Atmos. Environ.

    (2009)
  • M. Masiol et al.

    Aircraft engine exhaust emissions and other airport-related contributions to ambient air pollution: a review

    Atmos. Environ.

    (2014)
  • M. Masiol et al.

    Thirteen years of air pollution hourly monitoring in a large city: potential sources, trends, cycles and effects of car-free days

    Sci. Total Environ.

    (2014)
  • I. Mavroidis et al.

    Long-term trends of primary and secondary NO2 production in the Athens area. Variation of the NO2/NOx ratio

    Atmos. Environ.

    (2011)
  • I. Mavroidis et al.

    Trends of NOx, NO2 and O3 concentrations at three different types of air quality monitoring stations in Athens, Greece

    Atmos. Environ.

    (2012)
  • M. Mazaheri et al.

    An inventory of particle and gaseous emissions from large aircraft thrust engine operations at an airport

    Atmos. Environ.

    (2011)
  • A. Melkonyan et al.

    Long-term analysis of NO, NO2 and O3 concentrations in North Rhine-Westphalia, Germany

    Atmos. Environ.

    (2012)
  • A. Notario et al.

    Analysis of NO, NO2, NOx, O3 and oxidant (OX = O3 + NO2) levels measured in a metropolitan area in the southwest of Iberian Peninsula

    Atmos. Res.

    (2012)
  • F. Pranovi et al.

    Historical changes in the structure and functioning of the benthic community in the lagoon of Venice

    Estuar. Coast. Shelf Sci.

    (2008)
  • G. Schurmann et al.

    The impact of NOx, CO and VOC emissions on the air quality of Zurich airport

    Atmos. Environ.

    (2007)
  • Cited by (14)

    • Source emission and attribution of a large airport in Central China

      2022, Science of the Total Environment
      Citation Excerpt :

      GAM has been widely recognized in air pollution research for removing the effects of meteorological specificity and autocorrelation of variables (Schwartz, 1994; Schwartz, 1999). The selections of meteorological, pollutant and air traffic variables were referenced by a number of previous studies (Zhou et al., 2012; Valotto and Varin, 2016; Hua et al., 2021). Most previous applications of additive models to air pollution have considered daily average data (Carslaw et al., 2007).

    • Source apportionment of PAHs and n-alkanes bound to PM<inf>1</inf> collected near the Venice highway

      2017, Journal of Environmental Sciences (China)
      Citation Excerpt :

      The representative days for the last two factors are characterized by significantly higher concentrations of all monitored pollutants (CO, NO, NO2, PM10, PM1, TC, C350, ΣPAHs, Σalkanes). As in the case of the above mentioned PAHs, several works report that in urban areas the vehicular traffic is a significant source of these compounds (Ravindra et al., 2008; EMEP/EEA, 2013; Valotto and Varin, 2016; Wang et al., 2015; Hamra et al., 2015; Schiavon et al., 2015; Liu et al., 2015). It must be emphasized that in addition to the major role of the close highway, also the weather conditions contribute significantly to the increase of the pollutant concentrations.

    View all citing articles on Scopus
    View full text