Elsevier

Science of The Total Environment

Volume 443, 15 January 2013, Pages 681-691
Science of The Total Environment

Using a photochemical model to assess the horizontal, vertical and time distribution of PM2.5 in a complex area: Relationships between the regional and local sources and the meteorological conditions

https://doi.org/10.1016/j.scitotenv.2012.11.047Get rights and content

Abstract

A photochemical transport model has been implemented to assess the PM2.5 spatial and temporal distribution in Venice-Mestre. This is a large city of the eastern Po Valley, which is recognized having among the highest levels of many air pollutants in Europe. This study is a first attempt to evaluate PM2.5 distribution in such a complex ecosystem strongly affected by several different environments (the adjacent Alps, the lagoon and the sea) that create a spatial discontinuity of climate. Model performance was tested with experimental results. Samples have been collected in three sites representative of different emission characteristics. A second simulation was performed with clean boundary conditions to check the influence of the background concentrations on the study domain. Local and regional contributions were found to be strongly dependent on seasonal conditions and on local meteorology. A further analysis was conducted to predict the PM2.5 distribution with respect to air mass movements. The non-homogeneity of surfaces affects the Planetary Boundary Layer (PBL) behavior. This consequently influences the vertical distribution of PM2.5 especially during cold seasons and on occasion of particular meteorological events.

Highlights

► A first attempt of PM2.5 description in Venice with photochemical model was studied. ► A no-boundary simulation was performed to assess local and regional contributions. ► Good results was reached and discussed in relation to the complex meteorology. ► PM10 was well predicted for all seasons and in all the sampled sites. ► Model tendency to overestimate PM2.5 was discussed and related to PM2.5/PM10 ratio.

Introduction

The high levels of atmospheric pollution in Europe and the harmful health effects associated with the exposure to airborne particulate matter (Schwartz et al., 2002, Anderson et al., 2001, Klemm et al., 2000) have stimulated the attention toward the control of PM concentrations especially of the finer fraction (PM2.5, aerodynamic diameter < 2.5 μm, Council Directive 2008/50/EC). The primary emissions of PM2.5 are closely linked to the spatial distribution of sources. However, topography and meteorological conditions have a key role in transport, in secondary generation processes and also affect the residence time of particles. Small and large-scale meteorological phenomena may influence each other. As a consequence, a more complex composition of the particulate results and it requires sophisticated systems, such as chemical transport models, to be accurately studied. Advection, turbulent diffusion, gas phase reactions and heterogeneous transformations involving cloud formation and precipitation should be evaluated in these studies (Lonati et al., 2010, Zhang et al., 2006, Boylan and Russell, 2006, Hass et al., 2003, Seigneur et al., 2000).

Chemical models, integrated with experimental measurements, enable better understanding of the spatial and temporal PM2.5 distribution. They are particularly useful in heavily polluted areas such as the Po Valley. In this region the presence of large cities and the density of industries cause high PM concentrations (Barnaba and Gobbi, 2004, Gonzalez et al., 2000, Chu et al., 2003). In addition, the surrounding mountains (the Alps to the north and west and the Apennines to the south) favor the accumulation of particles in the valley. In this complex scenario Venice is a hot spot suffering from the advection of polluting aerosols originated both inside and outside the national boundaries. The contribution of local or long-range transports in Venice area has been assessed on the basis of considerations related to the movement of air masses (Masiol et al., 2010, Masiol et al., 2012a, Masiol et al., 2012b, Squizzato et al., 2012) or through the evaluation of the aerosol optical properties and the solar radiation (Barnaba et al., 2007). However the complexity of the phenomena requires more detailed data to be described. The first approach is based on the back trajectory technique that investigates the possible long-range origin of PM2.5, but provides no details on the air masses within the Planetary Boundary Layer. On the other hand, Lidar and sunphotometer observations are expensive and data, when available, are sparse in time and space. The missing information can be supplied by modeling especially in the Venice area, where winds, through a combination of large and local scale effects (Camuffo et al., 1979) vary the PM2.5 concentrations.

This work discusses the results of a photochemical transport model simulation to study the distribution of PM2.5 in the Venice area. The calculated data and measurements have been compared to assess the model performance. The PM2.5 origin and its spatial and temporal distribution have been investigated. A modeling approach with clean boundary conditions has also been tested to discriminate local and regional contributions to PM2.5. In order to better understand model capacity to estimate the secondary fine fraction, a PM10 and a PM2.5/PM10 ratio analysis has been developed. Finally the vertical PM2.5 distribution has been analyzed in relation to the Planetary Boundary Layer (PBL) behavior and to the air masses movements. The integration of calculated and experimental data has given additional interesting knowledge of PM2.5 distribution and its effect on human risk in this area.

Section snippets

Study area

The city of Venice is located on the coast of the Adriatic Sea at the margin of the Po Valley, in one of the most industrialized areas of Italy (Fig. 1). The study area covers almost 2500 km2, and includes dry land, with urban and industrial settlements, water, intertidal mud flats and marshes. Various emission sources are disseminated over the principal parts of the area including the historical center, the mainland urban settlement of Mestre (270,000 inhabitants) and the industrial district of

PM measurements and sampling

Three sites with different emission characteristics were selected in the Venice area: a semi-rural coastal site, Punta Sabbioni (SRC); an urban site in Mestre (URB) and an industrial site, near the Porto Marghera industrial area (IND) (Fig. 1). PM2.5 samples were collected during a one year period (2009–2010) according to EN 14907:2005 with a low-volume sampler (2.3 m3 h 1) on quartz fiber filters (Whatman QMA). PM2.5 mass was measured by gravimetric determination (microbalance with 1 μg

PM2.5 predictions

The general performance analysis has shown that the model well predicted the PM2.5 concentrations and their temporal and spatial distributions (Table 2). Values of correlation coefficient between observed and simulated daily average concentrations are in the range 0.7–0.8 for the spatial distribution and 0.6–0.8 for the seasonal one. The PM2.5 annual average concentrations in the domain are in the range 5–84 μg m 3 for predicted data and 3–120 μg m 3 for measurements (Table 2). The spatial

Conclusions

A photochemical model was implemented to study the PM2.5 dispersion dynamics in the Venice area. This site represents a very complex system where meteorology and emissions sources contribute to create very high pollutant scenarios. The model performance was estimated by comparing predictions with experimental data obtained during intensive field campaigns carried out in 2009. Measurements were collected from three different stations (semi-rural coastal, urban and industrial sites) located in

Acknowledgments

This work was supported by the Ente Zona Industriale, ENEL, ENI, EDISON and POLIMERI EUROPA, partners in the project, which provided data for the emissions inventory. The authors would also like to thank ARIANET S.p.A. and the Venice Water Authority. The authors are grateful to ARPAV managers and technicians for their analytical support and the Comando Zona Fari e Segnalamenti Marittimi di Venezia for logistics.

References (55)

  • J.F. Pankow

    An absorption model of gas/particle partitioning of organic compounds in the atmosphere

    Atmos Environ

    (1994)
  • A. Pederzoli et al.

    Quantification of Saharan dust contribution to PM10 concentrations over Italy during 2003–2005

    Atmos Environ

    (2010)
  • J.-P. Putaud et al.

    A European aerosol phenomenology — 3: physical and chemical characteristics of particulate matter from 60 rural, urban, and kerbside sites across Europe

    Atmos Environ

    (2010)
  • G. Rampazzo et al.

    Geochemical characterization of PM10 emitted by glass factories in Murano, Venice (Italy)

    Chemosphere

    (2008)
  • C. Silibello et al.

    Modelling of PM10 concentrations over Milano urban area using two aerosol modules

    Environ Model Softw

    (2008)
  • S. Squizzato et al.

    A procedure to assess local and long-range transport contributions to PM2.5 and secondary inorganic aerosol

    J Aerosol Sci

    (2012)
  • R. Vautard et al.

    On the contribution of natural Aeolian sources to particulate matter concentrations in Europe: testing hypothesis with a modeling approach

    Atmos Environ

    (2005)
  • K.T. Whitby

    The physical characteristics of sulfur aerosols

    Atmos Environ

    (1978)
  • Y. Zhang et al.

    A comprehensive performance evaluation of MM5-CMAQ for the summer 1999 southern oxidants study episode, part III: diagnostic and mechanistic evaluations

    Atmos Environ

    (2006)
  • H.R. Anderson et al.

    Particulate matter and daily mortality and hospital admissions in the UK's West Midlands conurbation: associations with fine and coarse particles, black smoke and sulphate

    Occup Environ Med

    (2001)
  • Aria Technologies

    MINERVE wind field models version 7.0

    General Design Manual ARIA Report

    (2001)
  • Arianet

    EMMA (EMGR/make) – user's guide – version 3.5

    Arianet R2005.08, Milano

    (2005)
  • F. Barnaba et al.

    Modeling the aerosol extinction versus backscatter relationship for lidar applications: maritime and continental conditions

    J Atmos Oceanic Technol

    (2004)
  • F. Barnaba et al.

    Aerosol stratification, optical properties and radiative forcing in Venice (Italy) during (ADRIEX)

    Q J R Meteorol Soc

    (2007)
  • A. Benassi et al.

    A one-year application of the Veneto air quality modelling system: regional concentrations and deposition on Venice lagoon

    Int J Environ Pollut

    (2011)
  • F.S. Binkowski

    The aerosol portion of models-3 CMAQ

  • F.S. Binkowski et al.

    The regional particulate matter model 1. Model description and preliminary results

    J Geophys Res

    (1995)
  • Cited by (30)

    View all citing articles on Scopus
    View full text