A New Technique for the Passive Monitoring of Particulate Matter: Olive Pollen Grains as Bioindicators of Air Quality in Urban and Industrial Areas

Featured Application

Our new, simple, and low-cost technique of particulate matter passive monitoring, which uses olive pollen as a bioindicator, can be employed to efficiently evaluate the atmospheric quality of urban and industrial ecosystems; to estimate the effects of pollen interactions with pollutants on pollen reproductive functions; and to predict the potential increase in pollen allergenic potency.

Abstract

A new technique for the passive monitoring of particulate matter was developed, exploiting olive pollen as a bioindicator. We tested the pollen bioaccumulation efficiency when exposed to atmospheric particulate at three different sites in the Umbria region (Central Italy). Pollen grains, placed into sampling holders, were exposed in Perugia, a polluted town impacted by traffic emissions; in Terni, an industrial hotspot; and at Monte Martano, a regional rural site. At the end of the exposure period, the daily deposition fluxes of the soluble and insoluble elements and soluble molecular ions present in particulate were determined, and the bioaccumulation factor (BAF) and bioaccumulation index over time (BAIt) were derived to validate the passive monitoring system, distinguish the deposition contribute from natural pollen composition, and interpret the temporal dependence of the pollen exposure to pollutants. We observed BAFs greater than 1, which means that bioaccumulation occurs, and pollen can be considered a good passive sampler for several crustal and anthropic ions and toxic elements at all sites. BAIt values greater than 1 were detected only for some of the ions and metals previously present in the pollen, like Ca, Cr, and Mn at Terni; and nitrate, Ca, and Mn at Monte Martano and Perugia.

1. Introduction

The air quality in urban and industrialized areas is currently assessed by observing the spatiotemporal distribution of atmospheric pollutants [1] and comparing the obtained values with the Environmental Regulatory references. However, a more realistic estimate of the air quality and the effects on human health and ecosystems can be achieved only if the quantification of the atmospheric pollutants is associated with the detection of the alterations induced on living organisms. Only in this way is it possible to obtain a global and objective view of the quality status of the ecosystems these organisms are part of.

A handy approach to assessing the level and impact of atmospheric pollution on ecosystems consists of using bioindication techniques capable of quantifying the accumulation of pollutants and detecting morphological, physiological, and/or genetic changes induced by pollutants on sensitive animal and plant organisms [2,3,4].

Airborne pollen is one of the significant constituents of bioaerosol and is considered a good bioindicator thanks to its constant presence in the atmosphere and its sensitivity toward pollutants. The interaction with contaminants has direct effects on its biological and reproductive functions, on the chemical-physical properties of its surface, and on pollinosis, which has continuously increased worldwide in recent years [5,6,7,8].

Pollen grains are organisms that are constantly influenced by the environment. The mature pollen, ready to be released from the plant, is dehydrated and therefore highly hygroscopic; consequently, it is possible that through the humidity, the pollutants of the atmosphere adhere to its surface. The two walls of the pollen (Esine and Intine) represent very effective protection, making the pollen grain one of the most resistant structures in nature. However, substances may penetrate inside the granule, especially in the granule with more germination openings. Olive pollen has three sheep-shaped openings, named colpus, of about 10 × 4 µm in dimension. Each colpus, although protected by a membrane, represents an area of greater permeability and vulnerability; so, in polluted environments absorbing water, the interactions with pollutants can compromise vitality [9] and its ability to perform post-pollination events by carrying out fertilization.

The vitality, germination, and development of the pollen tube can therefore be considered a reliable index of environmental conditions [10]. Indeed, Renzoni et al. reported that Pimis pinea pollen collected from a polluted urban area (Pisa town center, Central Italy) showed a reduction in germination and morphological anomalies, such as an increasing number of anomalous grains and decreasing tube growth with respect to pollen collected in an unpolluted area [11].

The chemical composition of pollen is also modified by air pollution; in particular, changes in inorganic ions and elemental composition have been observed [7].

An increase in concentrations of chemical compounds adsorbed on the surface of grains is observable when comparing pollen from polluted and unpolluted sites. Wang et collaborators have shown that in the Kanto region (Greater Tokyo Area, Japan), the concentrations of ionic species (NO₃⁻, SO₄²⁻, and NH₄⁺) and gaseous pollutants (NO₂, SO₂, and NH₃) deposited on pollen grains from Cryptomeria japonica were higher in polluted urban zone than in mountainous areas [12]. In addition, the susceptibility of pollen to pollutants depends on the plant species. Visez et al. showed that the uptake of NO₂ depends on the analyzed pollen and decreases in this order: cypress, timothy grass, and birch [13]. Ribeiro et al. demonstrated that the pollen modifications due to O₃ are species-dependent [14].

On the other hand, the chemical modifications of the elemental composition of pollen grains are not yet well understood. The pollen of anemophilous plants harvested from the anthers contains mainly K, Cl, S, P, Ca, Mg, Fe, Si, Al, Na, and Br [15,16]. Potassium is the dominant element in freshly collected pollen, but several studies have shown that environmental pollution influences the concentration of this element, which is considered a co-inductor or adjuvant of allergenicity. When the pollen is exposed to a polluted atmosphere, the inorganic elements of the exine are modified so that a predominance of Cl⁻ ions, an increased amount of P (Dactylis glomerata), and a predominance of or increased S with a slight increase in Cl⁻ (Betula verrucosa) can be observed [15]. Several other elements, such as Zn and Pb, can also be found in pollen, and their relative amounts can change in polluted environments. An increase in Zn concentration was observed for the polluted Betula pollen at various sampling zones in Stockholm [17], and a differential concentration of Pb in Compositae pollen was detected according to its presence in zones where the pollen was collected [18]. Nevertheless, the literature data do not allow us to identify an apparent convergence in the modification of the elemental composition of polluted versus unpolluted pollen [19,20,21,22,23].

Particulate matter (PM) is one of the principal air pollutants and is potentially dangerous to human health. In fact, up to 96% of the European Union’s urban population is exposed to fine PM concentrations above World Health Organization guidelines [24]. The assessment of PM concentration and its chemical composition is mainly carried out at monitoring sites in fixed locations, equipped with active air-sampling systems and specific instruments for the continuous measurement of pollutants [25]. Air-quality management typically requires a high number of monitoring stations. Unfortunately, fixed stations have high installation and maintenance costs and have disadvantages such as the need for electricity to operate, the difficulty in finding an appropriate location, and the risk of damage. For all these reasons, the number of monitoring stations, measuring PM, is often insufficient and does not allow for a realistic population risk assessment. Bioindicators can be used as passive samplers alternatively to active sampling stations, to extend PM monitoring networks. Lichens [2,26] and spider webs [27,28,29,30,31] have been recently used as passive samplers for air pollution monitoring. Both bioindicators show good properties for collecting PM. However, there are also some drawbacks: lichens are sensitive to temperature, humidity, elevation, and substrate chemistry [32,33], while for the other bioindicator, the non-contaminated web from laboratory-reared spiders needs to be transplanted to the monitoring site [27,34].

In this work, olive pollen grains (Olea europaea L.) have been used as a passive particulate matter sampler. To validate the new technique, we exposed pollen samples to air in industrial, urban, and rural areas in Umbria (Central Italy) and quantified the bioaccumulation levels of particulate constituents. Samples of olive pollen (Cultivar San Felice and Leccino), placed into passive sampling holders, were exposed in two urban sites and one rural site in the same period. Vehicular traffic is dominant in the town of Perugia [1]; stainless steel and chemical industrial pollution distinguish the basin valley where the town of Terni is located [34,35,36,37,38]; and regional background conditions are characteristics of the Monte Martano EMEP network station [39,40]. At the end of the exposure period, the pollen samples were treated according to specific protocols, and the daily deposition surface fluxes of the main soluble and insoluble metals and ionic species deposited on pollen were determined to calculate the bioaccumulation factors (BFA) and bioaccumulation indexes over time (BAIts), which allowed us to: (I) estimate the pollutant accumulation level in pollen grains; (II) distinguish the deposition contribution from natural pollen composition; (III) interpret the temporal dependence of the pollen exposure to pollutants; and (IV) evaluate the site atmospheric quality and highlight the presence of different anthropic pressures in the study areas.

2. Materials and Methods

2.1. Study Areas

The monitoring tests were carried out in 2019 and 2021 during the spring-summer period (from June to August) in the towns of Perugia and Terni and at the Monte Martano EMEP network station. Perugia and Terni are the main cities of Umbria, a region of central Italy. Umbria includes the upper and middle valley of the Tiber River, bordered on the west and east by low hills that gradually rise in the east to the Umbria-Marche Apennine chain. The geographical characteristic of the region is the presence of wide tectonic basins, some of which host lacustrine (Trasimeno Lake) and riverine (Tiber River) basins. Perugia is the capital city of Umbria and is crossed by the river Tiber. It is located on the top of a hill that reaches a maximum altitude of 493 m a.s.l. The city presents a humid subtropical climate due to its inland location and the hilly topography of Umbria. Typically, summers are warm to hot and humid, while winters are cold with occasional snowfall. Terni is a densely populated and industrialized city located at the margins of the Central Apennines. It lies in a vast plain area (170 m a.s.l.) surrounded by hills (800–1200 m a.s.l.). The local climate is classified as mid-latitude temperate, with warm, humid summers and cold rainy winters [41]. Monte Martano is a regional-background mid-altitude station for air-quality monitoring; it is distant from any local pollution source and influenced by the boundary layer only during summertime [39]. The Monte Martano site was included in the European Monitoring and Evaluation Programme (EMEP) in 2017 (Figure 1 and Figure 2).

2.2. Pollen Sampling Campaigns and Air Exposure Tests

Pollen was collected in 2019 and 2021 from selected olive trees (Olea europaea L., Cultivar San Felice and Leccino) located in the suburbs of Giano dell’Umbria, a rural area close to Monte Martano, and in the town of Perugia. During the full pollination period, the inflorescences were enveloped by pollination bags constituted of lightweight paper, which provided protection from wind dispersion, foreign pollen, and air pollutants. Homogeneous layers of pollen (200 mg) were deposited in a set of Petri dishes (8 cm diameter) fitted inside six passive deposition chambers. Each chamber, constituted by four gridded side walls, contained three Petri dishes with pollen (POL in the following) and one clean Petri dish as a control of the net deposition fluxes (CNT in the following). In this configuration, each Petri dish acts as a collector plate where atmospheric pollutants are accumulated with a specific flux. We placed two deposition chambers at each monitoring site. The chambers were installed on the roof of air-quality monitoring cabins (ARPA-Umbria Regional Environmental Protection Agency) at about 3 m above the surrounding ground (Figure 1). In Perugia (PG), we selected a monitoring station installed in a residential and commercial district (Fontivegge Station headquarter) characterized mainly by traffic pollution [1], while in Terni (TR), we chose an industrial monitoring station located near the AST (Acciai Special Terni) factory, one of the oldest steel plants in Europe and one of the largest in Italy [38]. The background Monte Martano (1094 m a.s.l.) station (MM) is located about 50 km south of Perugia and 45 km north of Terni. All the selected air-quality monitoring stations are equipped with instruments to measure meteorological parameters and air pollutant concentrations (gases and aerosol), including sampling systems for total (wet and dry) depositions. The starting day of exposure was the same for all the chambers at the three sites. The first set of chambers was removed after 7 days, and the others after 15 days. Two exposure experiments were performed in the periods 3–17 July 2019 and 21 July–4 August 2021, respectively. During the 2019 experiment, pollen grains were exposed to air for 15 days only. After sampling, the POL and CNT collector plates were refrigerated at 5°, transported to the laboratory, and stored at −18 °C until chemical analyses.

2.3. Meteorological Conditions

During the two study periods (summers of 2019 and 2021), the meteorological conditions at the monitoring sites were similar. In general, TR and PG featured warmer temperatures and lower winds with respect to MM, due to the altitude of the latter site. TR was also slightly dryer. During the 2019 exposure test, poor precipitations were registered in the 10–15 August period, while in 2021, light rainfalls were detected on 1 August at all sites.

2.4. Air-Quality Parameters

Gaseous pollutants and particulate matter were also monitored. Daily aerosol filters were collected at all the sites, and PM₁₀ and PM_2.5 concentrations were determined automatically by using dual-channel low-volume samplers (SWAM5A, Fai Instruments, Roma, Italy) equipped with a beta attenuation detector. Concentrations of gaseous pollutants O₃ and NO₂ were also routinely monitored by photometric (Teledyne T400, San Diego, CA, USA) and chemiluminescent (Teledyne T200, San Diego, CA, USA) analyzers, respectively.

2.5. Samples Processing and Analysis

The soluble fractions were extracted in ultrapure water (18.2 MΩ/cm) and quantified, respectively, by ion chromatography and ICP–QQQMS. The freshly harvested pollen and POL and CNT samples were extracted twice by shaking with 14 mL of water for 2 min. The resulting extracts were centrifuged at 10,000× g for 20 min at 17 °C, and the supernatants were divided into two aliquots: one was filtered (0.2 µm) and used to assess the soluble cationic and anionic species by suppressed ionic chromatography (AQUION and ICS-2100 Dionex supplied by ThermoFisher Scientific, Waltham, MA USA) and the other, after acid digestion in an HNO₃:H₂O₂ mixture (4:2) assisted by a microwave oven (MARS 6 Microwave Digestion System, CEM Corporation, Matthews, NC, USA), was used to determine the soluble metals present by using the triple quadrupole Agilent 8900 ICP-MS-QQQ instrument (Agilent Technologies, Santa Clara, CA USA)equipped with a collision/reaction cell (CRC) and connected to an SP4 autosampler. In addition, the total metal concentrations were quantified by ICP–QQQMS spectroscopy, analyzing microwave digestions in HNO₃ and H₂O₂ (4:2) of aqueous dispersions (20 mL) of pollen and CNT samples shaken for 2 min. Details of the acid digestion procedure can be found in Selvaggi et al., 2023 [42].

Element analysis. Commercially produced standard solutions (ICP multielement standard solution CertiPUR^®, 1000 ppm, Merck Chemicals, and Reagents GmbH, Darmstadt, Germany) in nitric acid and internal standard solution (Internal standard mix for ICP-MS systems, 100 ppm, 6-Li, Sc, Ge, Rh, In, Tb, Lu, Bi, Agilent Technologies, Santa Clara, CA USA) were used to prepare appropriate elemental calibration standards. An ordinary least-squares regression model (OLSR) was used for calibrations. Linearity between intensity and concentration was observed in the range of 1–10,000 ppt (R² > 0.999). Instrument detection limits, based on the calibration curve method, were estimated to be 12 for Cr, 26 for Mn, 41 for Fe, 2 for Co, 15 for Ni, 1 for Cu, 175 for Zn, 3 for Pb, 5 for Cd 12 for Ba, 24 for Sr, 293 for Si, and 611 ppt for Al. Experimental repeatability was calculated by performing three replicate analyses of two multi-element standards solutions (0.5 and 1 ppb). The metal RSDs obtained by the repeatability test were 2.6–13.0%. The accuracy of the analytical method was obtained using standard reference material (ERM^®-BB422 fish muscle, European Commission, JRC, Geel, Belgium). The metal concentrations agreed with the certified value, and recovery fell in the range of 80–120%. Analytical quality control included the analysis of a digestion reagent blank with each batch of 12 samples.

Ion analysis. Multi-ion calibration standards were prepared from single-component standard solutions for IC (Fluka-TraceCERT^TM, 1000 ± 4 ppm, Honeywell International Inc, Charlotte, NC, USA). Linearity between peak area and concentration was observed in the range of 0.05–1 ppm for the soluble cationic and anionic species (R² > 0.999) using a 500 µL loop. Instrument detection limits, based on the calibration curve method, were estimated to be 0.01 for MSA, nitrite, and sulfate; 0.02 for Li⁺, Cl⁻, Br⁻, nitrate, and phosphate; 0.03 for Na⁺ and Mg²⁺; 0.04 for K⁺, oxalate, and F⁻; 0.05 for formate; 0.07 for ammonium; and 0.4 ppm for Ca²⁺. Experimental repeatability and accuracy of the analytical method were calculated by performing three replicate analyses of a multi-element standards solution (1 ppm). The ion RSDs obtained by the repeatability test were 2.6–19.8% for anionic and 2.2–8–4% for cationic species, and recovery fell in the range of 90–106% for anions and 90–102% for cations.

All reagents were of analytical- or, whenever available, suprapure-grade quality and were from VWR Chemicals (Radnor, PA, USA). Ultrapure water was supplied by the Milli-Q Direct-3 system from Merck-Millipore (Darmstadt, Germany). The laboratory plasticware was soaked in 10 % HNO₃ for 24 h and then rinsed before use with ultrapure water [43].

2.6. Bioaccumulation Factor (BAF) and Bioaccumulation Index over Time (BAIt)

The bioaccumulation factor (BAF) is used to evaluate the partitioning of natural or anthropogenic substances from the environment to the biota [44] and is calculated by the normalization of the concentration of the generic element or compound, x, with respect to a reference element, r. The reference element must be stable in the environment (soil, ambient air, and water) and possibly not associated with specific pollution sources. Examples of reference elements include Al, Fe, Mn, La, and Rb. In the present study, Al, a common terrigenous element, was chosen since it represents one of the predominant elements in natural atmospheric aerosols [35,39].

We evaluated the ion and metal accumulation levels by defining the bioaccumulation factor (BAF) as:

BAF_x = [(F_x/F_r)_POL/(F_x/F_r)_CNT]

where F_x and F_r are the deposition fluxes (µg·m⁻²·day⁻¹) of the x species and the r reference element on the pollen covered (POL) and control plates (CNT), respectively. Furthermore, to properly evaluate the accumulation of the elements naturally present in the pollen tissues and to interpret the temporal dependence of the exposure to pollutants, we calculated a bioaccumulation index over time (BAIt) [44] for the major species x originally present in the pollen grains, defined as follows:

BAIt_x = (P_tx/P_tr)/(P_0x/P_0r) − 1

where P_tx and P_tr are the amounts (µg) of the x and r species as a function of the exposure time t (t = 7 and 15 days). P_0x and P_0r (t = 0) have been obtained using the freshly harvested pollen, and Al is the reference, as above.

2.7. Scanning Electron Microscopy Analysis (SEM)

Olive pollen grains picked up from the Petri dishes exposed at the three sites were morphologically characterized by SEM coupled with energy-dispersive X-ray microanalysis (SEM-EDS) at the LUNA LAB of the University of Perugia. The samples were prepared by mounting a small aliquot of the pollen directly onto SEM aluminum stubs using double-sided carbon tape. The samples were finally coated with a 100–150-Å carbon film to provide electrical conductivity and prevent charge buildup during exposure to the electron beam. SEM imaging was performed using a ZEISS Supra 25 microscope equipped with a field emission gun and a GEMINI column employed at 15 kV and variable magnification to distinguish the textural details of particle types. The stubs preparation method did not allow us to evaluate any variations in the size and shape of the granules.

2.8. Statistical Analysis

Bioaccumulation factors (BAFs) and bioaccumulation indexes over time (BAIts) were compared using the one-way analysis of variance (ANOVA) test. Before ANOVA tests, the normality and homoscedasticity of the data were examined using the Shapiro–Wilk test and Levene’s test (p < 0.05), respectively. Tests were performed in the Origin(Pro) environment, Version 2018, OriginLab Corporation, Northampton, MA, USA.

3. Results and Discussion

The average chemical composition of the olive pollen used in the study is reported in

Table 1. Mean chemical composition of pollen olive grains (Cultivar San Felice and Leccino) collected in 2019 and 2021 in Umbria (Central Italy) and used in the study.

µg/g	Olive Pollen
Soluble Ionic Fraction	Mean	Standard Deviation
Na+	133	110
NH4+	227	178
K+	4075	3321
Mg2+	115	57
Ca2+	1525	1392
F−	<LOD *
HCOO−	48	45
MSA	<LOD *
Cl−	419	345
NO2−	<LOD *
SO42−	559	455
C2O42−	155	121
Br−	<LOD *
NO3−	117	49
PO43−	3392	2875
Total metallic fraction
Cr	0.15	0.15
Mn	5	5
Fe	57	57
Co	<LOD *
Ni	<LOD *
Cu	1
Zn	<LOD *
Pb	<LOD *
Cd	<LOD *

* Limit of detection of the analytical procedure.

. Olive pollen showed considerable amounts of potassium, calcium, sodium, and magnesium; minor quantities of manganese, iron, copper, and chromium; and high ion concentrations of PO₄³⁻, SO₄²⁻, Cl⁻, NH4⁺, C₂O₄²⁻, and NO₃⁻.

The meteorological and air-quality parameter values detected in the exposure periods are shown in

Table 2. Average values of meteorological and air-quality parameters detected during the study periods.

Air Exposition Test		2019		2021
		Mean	Standard Deviation	Mean	Standard Deviation
Temperature (°C)
	TR	26.8	2.2	27.7	1.4
	PG	24.2	2.7	26.6	1.5
	MM	20.9	2.8	22.4	1.6
Wind speed (km h−1)
	TR	0.4	0.2	2.1	0.1
	PG	1.1	0.2	1.1	0.5
	MM	4.5	1.5	6.3	3.1
Relative humidity %
	TR	50.2	6.4	43	5.5
	PG	64.6	6.9	56	3.3
	MM	66.7	13.5	50	11.7
Precipitations (mm)
	TR	0.3	0.7	0.01	0.05
	PG	1.4	3.4	0.1	0.4
	MM	1.3	3.1	0.2	0.6
PM10 (µg m−3)
	TR	31.9	14.2	43.5	9.3
	PG	16.7	6.8	18.8	0.9
	MM	14.1	6.1	17.9	5.1
PM2.5 (µg m−3)
	TR	19.2	8.0	20.7	9.1
	PG	9.4	3.1	8.6	2.3
	MM	9.4	3.1	7.5	3.0
Ozone (µg m−3)
	TR	83.5	12.1	137.1	18.6
	PG	70.2	12.5	79.5	7.4
	MM	98.4	14.3	116.6	13.5
NO2 (µg m−3)
	TR	34.4	15.9	50.6	17.0
	PG	27.7	10.6	32.3	19.0
	MM	6.1	2.5	4.6	2.1

TR, Terni; PG, Perugia; MM, Monte Martano.

. Meteorological factors, such as wind speed, wind direction, humidity, rainfall, and mean temperature, affect PM deposition. However, in field studies, it is difficult to isolate the effects of single meteorological parameters because monitoring periods typically last for days or weeks, whereas the weather conditions vary continuously [45]. During the two study periods (summers of 2019 and 2021), the average meteorological conditions registered at the monitoring sites, although belonging to different climatic zones, were similar, and we considered that the influence of meteorological parameters on deposition fluxes was comparable in all sites. PM concentrations also affect the deposition fluxes. Similar concentration values were detected for particles with diameter less than 10 μm (PM₁₀) and for particles with diameter less than 2.5 μm (PM_2.5) in the exposure periods at Perugia and Monte Martano sites. Higher values were detected at the Terni industrial site.

During the summer of 2019, a first exploratory experiment was carried out, and pollen grains were exposed to air in the three sites for 15 days. The daily deposition fluxes of the main soluble and insoluble metals and major ionic species were determined. In Table S1, the POL and CNT fluxes expressed in µg·m⁻²·d⁻¹ are reported. At the MM regional background site (a mid-altitude station for air-quality monitoring distant from any pollution source) and at the PG urban site (located in a residential and commercial district characterized mainly by traffic pollution), the deposition fluxes of soluble ionic and total metallic fractions on POL samples were always greater or similar to those of the CNTs. However, the TR industrial site, situated near the AST (Acciai Special Terni) factory, one of the oldest steel plants in Europe and one of the largest in Italy, registered the highest fluxes. At TR, the fluxes in POL and CNT samples were similar except for some pollutants associated with steel production, namely calcium, fluoride, formate, and sulfate, which were higher in the CNT collector dish [36].

At the end of the exposure time, olive pollen grains selected from the passive samplers exposed in the three sites were observed by SEM. The bioaccumulation and interaction of the constituents of the atmospheric particulate with olive pollen were revealed by the microscope images, which showed, in some cases, significant morphological anomalies on the grain surfaces. In particular, the granules exposed at the industrial TR site, and, to a lesser extent, those exposed at the urban PG site, presented an anomalous thinning of the walls of the reticulum of the exine (Figure 3), suggesting chemical damage was induced by the interaction with the pollutants. The morphological damage was more evident at TR probably due to the presence of specific pollutants generated by the steel plant and due to the site’s geographical location. The town of Terni is located in a vast plain area surrounded by medium-range geographical elevations (average elevation 800–1200 m a.s.l.), which limit the dispersion of locally generated air pollutants [37].

In light of the promising results obtained, and in order to validate the new system of passive sampling, we replicated the experiment in 2021, exposing pollen samples to atmospheric pollutants for 7 and 15 days. In Table S2 are shown the POL and CNT fluxes calculated for the soluble ionic and metallic fractions, expressed in µg·m⁻²·d⁻¹. After 7 days, the POL fluxes of soluble ionic and metallic fractions were always similar or greater than those of the CNT, except for Na, Ca, and Fe at MM; Ca at PG; and Cr, Fe, Co, Ni, Pb, Ba, Al at TR. Furthermore, for the 15-day exposure time, we observed higher controls than sample fluxes for Na and Br at MM and for Ca, F, Cr, Fe, Ni, and Al at TR. The different deposition fluxes detected in 2019 and 2021 at the three sites were probably due to the circulation of air masses characterized by different chemical composition, and in the rural background MM site also to the perturbation of aerosol mass concentrations induced by increasing height of the boundary layer, which is observable in the summertime [39].

The daily surface deposition fluxes of the main soluble and insoluble metals and ionic species deposited on pollen were used to calculate the bioaccumulation factors (BFAs) and bioaccumulation indexes over time (BAIts). The BAF index characterizes the bioaccumulation of natural or anthropogenic elements in organisms living in natural ecosystems. When an organism is born into a given natural matrix, bioaccumulation takes place from birth to death or to the moment when the organism is sampled for measurement [44]. In these conditions, the concentration of a given element in the tissues of the organism is related only to the concentration in the natural matrix. In this respect, the BFAs consent to correctly evaluate the element accumulation level. When the concentration of an element in an organism is higher than in the natural matrix, the BAF value is greater than 1, which means that bioaccumulation occurs, while the opposite occurs if BAF < 1. Finally, if BAF = 1, the concentration of an element in the organism corresponds to the concentration in its environment, and the organism is called a bioindicator [44].

Table 3. BAF indexes (values, mean, and standard deviation) for ions and elements present on the pollen grains surface in 2019 and 2021 sampling.

	Monte Martano					Perugia					Terni
	2019	2021				2019	2021				2019	2021
Days	15	7	15	Mean	SD	15	7	15	Mean	SD	15	7	15	Mean	SD
Na+	1.6	1.0	0.9	1.2	0.3	1.3	0.1	0.8	0.7	0.5	0.7	2.6	1.3	1.5	0.8
NH4+	1.2	5.1		3.2	2.0	9.0		3.5	6.3	2.8	3.9
K+	1.8	10.1	1.0	4.3	4.1	6.5	1.0	6.4	4.6	2.6	3.2	21.2	10.1	11.5	7.4
Mg2+	0.6	6.2	2.1	3.0	2.3	3.4	1.2	4.4	3.0	1.3	1.7	39.9	8.9	16.8	16.6
Ca2+	1.3	0.6	1.1	1.0	0.3	6.8	0.1	1.0	2.6	3.0	0.7	2.1	1.0	1.3	0.6
F−											0.4	2.2	0.8	1.2	0.8
HCOO−			0.9				1.0	1.1	1.1	0.1	0.6	3.5	3.4	2.5	1.4
MSA											2.0		1.8	1.9	0.1
Cl−	0.7	1.5	1.0	1.1	0.4	1.5	0.1	1.1	0.9	0.6	1.3	3.2	1.5	2.0	0.9
NO2
SO42−	0.5	1.2	1.0	0.9	0.3	1.3	0.12	1.0	0.8	0.5	0.7	3.0	1.2	1.6	1.0
C2O42−											1.9
Br−		1.5	0.4	1.0	0.6		0.2	1.0	0.6	0.4	.	6.0	2.6	4.3	1.7
NO3−	0.4	2.0	1.1	1.2	0.7	0.6	0.1	1.0	0.6	0.3	0.4	2.5	1.3	1.4	0.9
PO43−			7.4					19.6
Cr	1.2	1.4	0.7	1.1	0.3	1.6	0.7	1.5	1.2	0.4	1.0	1.4	1.1	1.2	0.2
Mn	1.6	2.6	1.1	1.8	0.6	2.0	1.9	1.4	1.8	0.2	1.0	1.5	1.3	1.3	0.2
Fe	1.4	0.9	1.0	1.1	0.2	1.2	0.9	1.1	1.0	0.1	0.8	1.1	1.1	1.0	0.1
Co	0.6	1.8	1.1	1.1	0.5	1.9	0.7	1.2	1.3	0.5	0.7	1.7	1.4	1.3	0.4
Ni	1.2	31.1	1.7	11.3	14.0	1.4	0.1	.	0.7	0.6	1.1	1.9	0.8	1.3	0.4
Cu	1.8	7.1	1.4	3.4	2.6	1.8	0.9	1.0	1.2	0.4	1.3	2.9	2.6	2.3	0.7
Zn	2.4	7.0	1.1	3.5	2.6	1.3	0.2	0.8	0.8	0.5	1.2	2.1	1.6	1.6	0.4
Pb	0.9	3.9	0.6	1.8	1.5	2.4	0.6	0.6	1.2	0.9	0.8	1.8	1.6	1.4	0.4
Cd	0.9					1.3					0.9
Ba	1.2	1.7	0.9	1.3	0.4	1.6	0.1	0.8	0.9	0.6	1.0	1.5	1.2	1.2	0.2
Sr	1.5	2.0	1.0	1.5	0.4	1.6	0.1	0.6	0.8	0.6	1.0	2.2	1.6	1.6	0.5

shows the BFA values for analytes detected on the surface of pollen grains in the 2019 and 2021 experiments. No significant differences were found between the mean BAF values of pollutants detected at each site and between those detected at the three sites (ANOVA test, p > 0.05), which could highlight the presence in the sites of specific and relevant contamination sources. In 2019, the BAF values were greater than 1, which indicates that bioaccumulation occurred for some soluble ions (crustal and anthropic) and toxic metallic elements at all sites. Specifically, this occurred at the MM site for NH4⁺, Na, K, and Ca soluble ions and for Cr, Mn, Fe Ni, Cu, Zn, Ba, and Sr; and at the PG site for NH4⁺, Na, K, Mg, and Ca soluble cations and for chlorides, sulfates, Cr, Co, Mn, Fe, Ni, Cu, Zn, Pb, Cd, Ba, and Sr. Finally, at the TR site, we obtained BAF values greater than 1 for NH4⁺, K, Mg, MSA, chloride, and oxalate soluble ions and for Ni, Cu, and Zn. The highest bioaccumulation values were observed for Zn (BAF = 2.4) at the MM site and for NH4⁺ at the PG (BAF = 9) and TR sites (BAF = 3.9) (Table 3 and Figure S1).

In 2021, BAF values after 7 days were greater than 1 for ammonium, chlorides, sulfates, bromide, nitrate, K, Mg, Cr, Mn, Co, Ni, Cu, Zn, Pb, Ba, and Sr at the MM site; for Mg and Mn at the PG site; and for all ions and elements characterized by flux values higher than the detection limit at the TR site. The highest bioaccumulation values were observed for Ni (BAF = 31.1) at the MM site, for Mn at the PG site (BAF = 1.9), and for Mg at the TR site (BAF = 39.9) (Table 3 and Figure S2). In the collector plates exposed for 15 days, we observed BAF values greater than 1 for nitrate, phosphate, Mg, Ca, Mn, Co, Ni, Cu, and Zn at the MM site; for ammonium, formate, chlorides, nitrate, phosphate, K, Mg, Mn, Co, Ni, Cu, and Zn at the PG site; and finally, for formate, MSA, chlorides, sulfates, bromide, nitrate, phosphate, Na, Mg, K, Cr, Mn, Fe, Co, Cu, Zn, Pb, Ba, and Sr at the TR site. In this case, the highest bioaccumulation values were noted for phosphate (BAF = 7.4) at the MM and PG sites (BAF = 19.6) and for K at the TR site (BAF = 10.1) (Table 3, Figure S2).

If the organism is taken from external sources, the constituent elements previously accumulated in its tissues will be added to those that are present in the ecosystem investigated. Therefore, the use of the BAF index could lead to mistakes in the interpretation of the results and a different index should be applied [44]. Since olive pollen presents some non-zero elements and ion concentrations (Table 1), in 2021, we also estimated the bioaccumulation index over time (BAIt) [44]. The BAIT index allowed us to determine the relative increase in the concentration of a given element or ion with respect to its initial concentration in pollen and interpret the temporal dependence of the pollen exposure to pollutants (

Table 4. BAIt indexes (values, mean, and standard deviation) for ions and elements present on the pollen grains surface in 2021 sampling.

	Monte Martano				Perugia				Terni
Days	7	15	Mean	SD	7	15	Mean	SD	7	15	Mean	SD
Na+	−0.2	0.5	0.1	0.3	0.1	0.6	0.3	0.3	−0.2	−0.04	−0.1	0.1
NH4+	−0.01	−0.8	−0.4	0.4	0.1	−0.4	−0.2	0.2	−0.8	−0.8	−0.8	0.0
K+	−0.7	−0.9	−0.8	0.1	−0.5	−0.7	−0.6	0.1	−0.8	−0.8	−0.8	0.0
Mg2+	−0.7	−0.8	−0.73	0.03	−0.5	−0.5	−0.52	0.02	−0.6	−0.6	−0.61	0.02
Ca2+	0.7	1.8	1.3	0.5	1.4	2.0	1.7	0.3	14.8	16.2	15.5	0.7
Cl−	−0.3	−0.04	−0.2	0.2	−0.1	0.2	0.1	0.2	−0.3	−0.1	−0.2	0.1
SO42−	0.1	−0.2	−0.1	0.2	0.5	0.1	0.3	0.2	0.3	0.7	0.5	0.2
C2O42−					−0.9
NO3−	3.5	1.67	2.6	0.9	1.4	1.26	1.4	0.1	0.03	0.3	0.2	0.1
PO43−	−0.6	−0.9	−0.8	0.1	−0.4	−0.6	−0.5	0.1	−0.99	−1.0	−1.0	0.0
Cr	0.2	−0.1	0.1	0.1	0.5	0.7	0.6	0.1	16.3	18.5	17.4	1.1
Mn	61.5	46.3	53.9	7.6	67.7	67.0	67.4	0.4	123.9	149.7	136.8	12.9
Fe	−0.3	−0.4	−0.3	0.0	−0.2	−0.2	−0.17	0.02	−0.6	−0.5	−0.52	0.04

, Figure S3).

BAIt values between 0 and 1, which indicate a low bioaccumulation level, were detected after 7 days for Ca, sulphate, and Cr at the MM site; for ammonium, Ca, sulphate, and Cr at the PG site; and sulphate and nitrate at the TR site. Instead, after 15 days, we observed BAIt values between 0 and 1 for Na at the MM and PG sites; for chloride at the PG site; for sulphate at the PG and TR sites; for nitrate at the TR site; and finally, for Cr at the PG site. BAIt values greater than 1 indicate that the relative concentration of the given element in pollen grains increased during the experiment and therefore bioaccumulation occurred. We observed BAIt > 1 for nitrate and Mn at the MM site; for Ca, nitrate, and Mn at the PG site; and Ca, Cr, and Mn at the TR site after 7 days of exposure. On the other hand, after 15 days of exposure, we found BAIT values greater than 1 for Ca and nitrate and Mn at the MM and PG sites and for Ca, Cr, and Mn at the TR site. We also observed negative BAIt values for all the other analytes, which indicated that the final concentration of a given element or ion in the pollen was lower than the initial value, meaning that a given element or ion was transferred from the pollen to the atmosphere.

In summary, we observed BAIt values between 0 and 1 or greater than 1, which indicated bioaccumulation in pollen samples, only for some ions and element among those already present in pollen. These were sulphates, nitrates, ammonium (typical urban pollutants), Cr and Mn, (industrial and steel mill contaminants), and the crustal elements Na and Ca. We did not detect any bioaccumulation of potassium—the main element in freshly collected pollen that is considered a co-inductor or adjuvant of allergenicity [11,12], nor chloride, which is different from the results of previous studies [12].

Furthermore, significant differences were found between the mean BAIT values observed in the three sites for Ca (F = 231.6, 2, p < 0.001), Cr (F = 227.1, 2, p < 0,001), and Mn (F = 26.5, 2, p = 0.012). The highest BAIt values were observed in both tests (7 and 15 days) for Mn (123.9 and 149.7), Cr (16.3 and 18.5), and Ca (14.8 and 16.2) at the Terni industrial site, and the BAIt values for these elements were always significantly higher (p < 0.05) than those detected at the Perugia urban site and the Monte Martano control site in both tests. The values were often greatest after 15 days of exposure to ambient air.

In conclusion, the bioaccumulation indexes calculated in 2021 confirmed that olive pollen is an efficient passive sampler and that Terni, the second-biggest town in the Umbria region (≃110,000 inhabitants) and one of the largest stainless steel production sites in Europe, was the investigated area subjected to greater anthropic pressures.

Our results were also in good agreement with those reported by Moroni et al. (2013) [36]. The authors investigated the airborne dust over Terni and the morphochemical characterization of dust samples by scanning electron microscopy revealed that the local steel plants supply a regular contribution of calcite, metal particles (mainly Fe, Cr, Mn, Ni, and Zn), metal carbonate, and sulphate compounds to the airborne dust. Since the tests were carried out during the spring-summer period (from June to August), the high bioaccumulation values observed for Mn (61.5 and 46.3, respectively after 7 and 15 days) at the Monte Martano site can be explained by the perturbation of aerosol mass concentrations induced by the increasing height of the boundary layer in the summertime [39]. Although the average bioaccumulation index for fluoride was low and not significantly different (average BFA = 1.2 ± 0.8) from the others, the morphological anomalies of the surface of the grains observed at the Terni site could also have been induced by the fluorides emitted from the steel mill. Indeed, fluoride is one of the most harmful air pollutants, affecting agricultural and natural vegetation. Abnormalities and effects on the shapes and sizes of pollen grains exposed to ambient airs affected by fluoride pollution have been evidenced by observations using scanning electron microscopy [46].

In addition to good efficiency, our pollen sampler presents several advantages over the other passive systems reported in the literature, such as those using lichens [32,33] and spider webs [31,34]. The deposition chambers are very easy to construct and have low costs for installation and maintenance. The olive pollen can be collected indifferently from both cultivars investigated (San Felice and Leccino), since we did not observe variety-specific bioaccumulation of ions and/or elements. Pollen samplers can be installed in different climatic areas if similar average meteorological conditions are observed at the control and sample sites, and they can be positioned in zones with different altitudes. Finally, after exposure to air, the pollen samples can be used to estimate the effects of their interaction with pollutants on reproductive functions, and on pollinosis, to predict the possible increase in allergenic potency.

4. Conclusions

We developed a simple, efficient, and low-cost technique for the passive sampling of particulate matter, which allows us to evaluate the atmospheric quality of urban, industrial, and rural ecosystems using olive pollen as a bioindicator.

To validate the sampling system, we assessed the air quality of three sites located in the Umbria region (Central Italy) and exploited the BAF and BAIt indexes. The BAF is a valid index when organisms living in natural ecosystems from birth to death are used as passive samplers (e.g., lichens). If the bioindicating organisms are taken from external sources, like the pollens, the BAIT index allows us to better characterize the phenomenon of bioaccumulation. The BAIT index has allowed us to identify the pollutants mainly accumulated in pollen at the three investigated sites. Ca, Cr, and Mn, from industrial and steel plant sources, were prevalent at the Terni site; nitrate, Ca, and Mn were detected at the Monte Martano control site and at urban Perugia site. Finally, we identified the site affected by the strongest anthropic pressures. The BAIt values observed at the Terni site for Ca, Cr, and Mn were always higher than those detected at the Monte Martano control site and at the Perugia site and were often greatest after 15 days of exposure to air.