Mutiparametric Characterization of Atmospheric Particulate in a Heavy-Polluted Area of South Italy

To obtain a real-time image of atmospheric particulate matter (PM) in highly polluted areas and to understand how the anthropogenic component linked to urban activities (industrial activities, domestic heating, road traffic, waste disposal) can locally affect near-surface measurement of PM, several measurement campaigns were achieved in the Campania region (Southern Italy) using both Lidar and in-situ instruments. A comparison between the obtained results highlights a good correlation between the data and the potential of remote sensing instruments for air quality monitoring. Data analysis was performed in terms of particle backscattering coefficient profile at 355 nm, PM mass concentration, and size distribution. Wind profiles, which covered a range of altitudes from 40 mestic heating systems. Bimodal size distribution in number concentration was measured, indicative of two types of atmospheric particles sources: gas and liquid combustion (particles with sizes below 80 nm), including vehicular traffic and domestic gas-heating, and biomass combustion (particles with sizes of the order of 200 - 500 nm). Finally, data collected in a highly populated and industrialized area highlights the presence of particles having a high level of spherical geometry (aerosol depolarization below 5%) pointing towards the industrial area. Conversely, the measurements performed pointing toward other directions highlighted a diffused source of aspherical particles (depola-rization values of about 3%) spreading throughout all city territory. The work showed as the co-location of remote sensing and near surface instruments is a promising approach to studying aerosol properties in the atmospheric layers and has more accurate information on atmospheric dynamics. Moreover, the correlation between the obtained results highlighted the potential of remote sensing instruments for air quality monitoring.


Abstract
To obtain a real-time image of atmospheric particulate matter (PM) in highly polluted areas and to understand how the anthropogenic component linked to urban activities (industrial activities, domestic heating, road traffic, waste disposal) can locally affect near-surface measurement of PM, several measurement campaigns were achieved in the Campania region (Southern Italy) using both Lidar and in-situ instruments. A comparison between the obtained results highlights a good correlation between the data and the potential of remote sensing instruments for air quality monitoring. Data analysis was performed in terms of particle backscattering coefficient profile at 355 nm, PM mass concentration, and size distribution. Wind profiles, which covered a range of altitudes from 40 m to 290 m, were also used to study sources and physical processes involved. Measurement carried out in a rural area with a landfill site highlighted the presence of a homogeneous particulate layer throughout the sounded area due to winds driving aerosol from the landfill to the surrounding areas. The size distribution in mass concentration, highlighted a modal diameter moving towards 0.9 and 2 µm with a larger mass concentration of particles in the morning, before noon and in the afternoon when a large number of trucks delivered solid wastes. Moreover, large concentrations of particulate matter were measured in a small urban town with few industrial activities which peak (211 ± 33 µg·m −3 ) was measured in the direction of the most urbanized area, probably due to the lighting of the do-

Introduction
Particle matter (PM) has a strong impact on the chemical and physical processes occurring in the atmosphere, affecting air quality and climate change [1]- [6].
Many studies highlighted PM's direct effect on environment, climate, natural ecosystems and human health [7] [8] [9] [10] [11] due mainly to their increased concentration. PM is generated by natural sources (i.e., wind-borne dust, sea spray, volcanic ashes, biogenic aerosol), anthropogenic sources (i.e., fossil fuel combustion, waste and biomass burning, industrial emissions), and photooxidation transformation, under sunlight irradiation, sulphates, nitrates and organic precursors that are emitted in the atmosphere from human and agricultural activities (secondary aerosol) [12]. PM is rarely homogeneous: particles vary in size and shape, chemical composition as well as successive modifications in the atmosphere that are related to the specific source and emission location. Particle size can vary from few nanometers to several tens of microns whereas the chemical composition is a complex mixture of organic and inorganic substances [13] [14]. The complexity of PM chemical composition, its wide range of size and shape, along with variations in the time, make the airborne particulate matter measurement very difficult and determining the sources a real challenge. In recent years, due to high levels of atmospheric pollution, and the considerable number of exceedances occurring for particulate matter concentration measured on the ground, several studies [15] [16] [17] were performed in a crowded area of the Campania region in Southern Italy. These studies were initiated to understand the link between atmospheric fine particulate high concentrations on the able to provide information about particle sources and how these particles, once emitted and dispersed in the atmosphere, are transformed and/or transported far from the sources over long and medium distances. Differently, Lidar remote sensing devices allow the study of fine particle dynamics and observe related processes where they occur [18]. Consequently, to correlate PM time variability, measured on the ground with diffusion processes, and air masses transport phenomena, a comparison between the results derived from remote sensing and near surface instruments becomes important [19].

Methods
Data collected during the measurement campaigns have been obtained with sensors based on different physical principles and mode operation. Daily mass concentration of PM10 has been measured using traditional gravimetric sensors.
The distribution of the particle size was obtained with an Electrical Low-Pressure Impactor (ELPI) able to detect particles from 7 nm up to 10 µm. A Doppler Lidar device allowed to measure wind profiles up to 200 m and a portable Lidar system was used to study the evolution of the aerosol geometrical and optical properties with high spatial (meters) and temporal (seconds) resolutions. Conventional meteorological parameters (temperature, pressure, relative humidity, wind speed, and direction) and main pollutants concentrations are routinely measured by the Campania Regional Environmental Protection Agency (ARPAC) (https://www.arpacampania.it/).

Size Distributions
The particle size distributions on the ground were measured by an Electrical Low Pressure Impactor (ELPI, Dekati Ltd., Kansagala, Finland). ELPI is a particle size spectrometer for real-time monitoring of aerosols; the instrument is constituted by a corona charger, a 12-stage cascade low-pressure impactor and a multichannel electrometer. The particle size range is from 30 nm to 10 μm. An extra filter stage is added to enlarge the measurement size range down to 7 nm.
Thirteen collection plates are used to sample particles as divided by the impactor and a calibration procedure is used to associate particles collected on the plates to their aerodynamic size. Data are processed by the ELPI XLS4.05 software (Dekati Ltd.). To avoid particle coagulation, an overall dilution ratio of 5 is used.

Atmospheric Aerosol Monitoring
Laser remote sensing devices are known as Lidar, acronym of Light Detection and Ranging, represent one of the most promising survey methods for the meas- high spatial (meters) and temporal (seconds) resolution [20]. In the measurement campaigns, we used a prototype version of a Lidar device, as developed in the frame of the I-AMICA project (http://www.i-amica.it/i-amica). The system, called Microjuole POrtable LIdar System (μ-POLIS), was designed and developed by ALA Advanced Lidar Applications s.r.l., spin-off company of the University of Naples Federico II and the CNR. μ-POLIS is a compact Lidar system combining good accuracy, safety, portability, autonomy in remote use and ease of operation. All of these features make the system suitable for atmospheric pollutant observations in a more convenient way than the traditional Lidar instruments. The Lidar device works in the UV region of the electromagnetic spectrum, swivels from the vertical to the horizontal plane, and can measure the linear depolarization to identify particulate shape. The system uses a laser source in the UV range (355 nm) with 0.04 W mean optical power at 1 kHz repetition rate. The laser beam is expanded to make the system eye-safe for distances greater than 100 m. The light reflected by the target is received with a 20 cm Ritchey-Chrétien telescope and sent to the spectral selection system that is designed to detect the elastically diffused light with both parallel and perpendicular polarization, i.e. the P and S components of the backscattered radiation. The device is equipped with a complete software suite allowing automatic continuous measurement and data analysis. It can analyse the atmosphere from 0.2 km to 10 km. Lidar measurements were reported in terms of particles backscatter coefficient and particle depolarization profiles [20]. From the backscattering coefficients, it is possible to evaluate the particle mass concentration. Knowing the density of the particles and their effective radius, the particle concentration can be obtained, through the equation is the conversion factor (in meters); eff r is the effective radius of the particles; β is the backscattering coefficient (m −1 ·sr −1 ); LR is the Lidar Ratio (sr), for urban particles equal to 50 sr [21]; ρ the density in the range (1 -2) g·cm −3 and for urban particle evaluated as 1.5 This procedure follows the methodology already applied in [22] and [23] for volcanic ash cases.

Wind Profiles Measurements
To measure wind profile, we used the Leosphere WindCube

Measurements in a Rural Area with a Landfill Site
To study the dynamics of atmospheric processes that strongly influence the in-    December, 2015 only few data were acquired simultaneously with µ-Polis and with the ELPI system; however, the data resulted in good agreement (γ = 0.69 with N = 20 and p = 5% where γ is the Bravais-Pearson correlation coefficient, N is the number of samples and p is the statistical significance). By extrapolating backscattering coefficient value to the ground (β = 1.21 × 10 −5 m −1 ·sr −1 ) effective radius resulted r eff = 0.28 µm. Figure 3 shows that the backscattering coefficient       (Figure 8(a)) with respect to data corresponding to other directions (Figure 8(b)). Results reported in the figure showed that the total mass concentration did not depend on the wind speed for wind coming from south direction; therefore, no aerosol emission come from the storage area.

Measuring Campaign in a Small Urban Town with Few Industrial Activities
On The scanning capability of the Lidar μ-POLIS allowed us to perform measurements at the zenith and test the atmosphere in the near horizontal direction.
In this way, we were able to achieve information on particulate spatial distribution and identify possible sources. Figure           From these measurements, we were able to obtain information about PBL daily dynamic in order to have a direct indication on the atmospheric layer height where pollutants are confined. Figure 16(a) and Figure 16    coming from the two systems and determine particle effective dimensions, consistent with those deriving from combustion processes. Moreover, particle size distribution in mass and in number allowed to identify different sources of the particles at ground level. Results showed the presence of airborne pollutants homogeneously spread out in the city centers (most likely due to vehicular traffic and domestic heating) and at the same time, they allowed reducing the impact that the nearby industrial area was supposed to have. Finally, this work highlighted the capability of the measurement combined approach and showed as remote sensing instruments once tested and validated, can help to increase the knowledge about PM on the ground, paving the way for a new air quality monitoring system.