Peplospheric Influences on Local Greenhouse Gas and Aerosol Variability at the Lamezia Terme WMO/GAW Regional Station in Calabria, Southern Italy: A Multiparameter Investigation

1. Introduction

The Peplosphere, also referred to in literature as Planetary Boundary Layer (PBL) or Atmospheric Boundary Layer (ABL), is the lowest part of the atmosphere. The PBL is heavily influenced in its dynamics by constant contact with Earth’s surface [1,2,3,4,5,6]; it is also characterized by turbulences, vertical currents, distinct temperature profiles, and air flows which are not parallel to either the surface or isobars [7,8,9,10]. Only in the free atmosphere above the PBL air flows are mostly geostrophic [11,12].

The PBL has a well-defined diurnal-nocturnal cycle [13,14,15,16,17,18,19], which also results in peculiar dynamics directly affected by this cycle [20,21,22,23,24]. PBLH (Planetary Boundary Layer Height) is among the key parameters used in research to characterize peplospheric variability over time [25,26,27]. The PBLH is normally described as the height of the inversion level separation between the free troposphere above from the boundary layer [14]. In addition to the physical boundary of the PBL, it is worth mentioning that it possesses peculiar characteristics in terms of microbial ecology [28,29].

Research demonstrated correlations between PBLH and the concentration of several greenhouse gases and aerosols: therefore, PBL control over the surface concentration of these compounds is among the factors that can help predicting their variability and diffusion [30,31,32,33,34,35,36,37]. This also has a number of implications in terms of human health and environmental protection, as higher concentrations can lead to hazards [38,39,40]. For example, in the case of black carbon (BC) an inverse correlation with PBL height was demonstrated in areas affected by anthropogenic pollution [41]. In a world that is constantly dealing with climate change, air quality issues and the need for sustainable policies and regulations [42], additional knowledge on the PBL may be required.

In research, two main methods for PBLH measurement are used: the Gradient Method, and the Threshold Method. The former estimates PBLH by pinpointing the height of the minimum backscatter gradient [43,44,45,46,47], while the latter estimates PBLH as the threshold at which the backscatter signal is below a certain value [45,48].

In addition to the complexities of surface-PBL-free atmosphere interactions that normally occur on Earth, a significant role in PBL dynamics is played by the presence of sea/ocean masses, continents, and coastal boundaries below [8,49,50,51,52,53,54,55,56]. These extra complexities require ad hoc methodologies to assess PBL variability in these environments [57,58,59].

A better understanding of peplospheric influences over the concentration of GHGs and aerosols is necessary for provide regulators and policymakers with additional tools to better manage air quality issues in urban and rural areas. Various parameters affect air quality and local air pollution; this research paper will focus on three gases and two aerosol types.

Carbon dioxide (CO₂) is the main driver of present-day anthropogenic climate change [60,61,62] and is therefore subject to monitoring on local to global scales [63,64]. Although CO₂ does not pose the same health hazards as other atmospheric pollutants, it can impact the environment [65,66] and trigger long-term effects on human health [67,68,69]. Fossil fuel burning is the main source of anthropogenic CO₂ in the atmosphere [70].

Carbon monoxide (CO) is an effective tracer of combustion and can be both natural and anthropogenic in origin. Wildfires are a prominent source of CO in the atmosphere [71] and the multi-year variability of this gas has shown a generally downward trend [72] that followed years of constant increases [73]. The decrease rate of CO has reduced in the past few years, indicating changes in the total global budget [74]. Although carbon monoxide is short-lived, it is known to play a role in the increase of methane [75] and surface ozone (O₃) [76].

Methane (CH₄) is heterogeneous in origin, with anthropogenic [77,78] and natural [79] sources contributing to its global budget. It is also a byproduct of fuel combustion [80,81]. Due to its high GWP (global warming potential) compared to CO₂, CH₄ emission reductions are one of the main challenges of present-day climate change mitigation [82].

Black carbon (BC) is a notable byproduct of combustion processes [71] and acts as both a driver of climate change [83,84,85] and a factor of health hazard [86]. Its effects are partially counterbalanced by a short persistence rate in the atmosphere, as BC has been observed to last for days [87,88].

Particulate matter (PM) is a common byproduct of vehicular emissions in urban areas [89,90] but can also be originated by several natural processes [91,92,93]. PM constitutes a significant health hazard due to its small size and the consequent capacity to affect lungs directly [94,95]; for this reason, it is subject to constant monitoring at various levels [96]. Several research studies have reported significant correlations between PBL and particulate patterns, thus showing that research on PBL variability also has implications for air quality and the sustainable development of urban and rural areas [97,98,99,100,101].

Previous research aimed at the WMO/GAW (World Meteorological Organization – Global Atmosphere Watch) observation site of Lamezia Terme (code: LMT) in the southern Italian region of Calabria was limited to data gathered during a summer 2009 campaign (specifically, from July 12^th to August 6^th) [102]. The campaign allowed the characterization of the main features of peplospheric variability at LMT, however, at the time there were no measurements of greenhouse gases and aerosols available. A further characterization of PBL influences over the concentration of pollutants was focused on the 2015 solar eclipse [103], but the study could not provide additional details on variability over time.

This study is therefore aimed at an unprecedented analysis of PBL variability at LMT, as well as its correlations with the patterns of key greenhouse gases and aerosols. This research is divided as follows: section 2 describes the LMT site, its instruments and datasets; section 3 will show the results of this campaign; sections 4 and 5 cover the discussion and results, respectively.

2. The Station, Methods, Instruments, and Datasets

2.1. The Observation Site and Its Characteristics

Fully operated by the National Research Council of Italy – Institute of Atmospheric Sciences and Climate (CNR-ISAC), the observation site (code: LMT; Lat: 38.88 N°; Lon: 16.23 E°; Elev: 6 m.a.s.l.) is located in the southern Italian region of Calabria, in the municipality of Lamezia Terme (Figure 1). The station is located 600 meters from the Tyrrhenian coast of the region (Figure 1), in an area known to be the narrowest point in the entire Italian peninsula, as the distance between the Tyrrhenian and Ionian coasts is ≈30 kilometers. This area, known as the Catanzaro isthmus, effectively separates the coastal chain (Catena Costiera) and Sila Massif in the north, from the Serre Massif in the south. The presence of two seas, with the Calabria region in between, results in increased meteorological instability, and oftentimes results in the occurrence of floods in the area [104,105,106].

Local wind circulation patterns in the western part of the isthmus were characterized in two 2010 studies by Federico and collaborators, which demonstrated the presence of a well-defined west-east local circulation [107,108]. In fact, the Lamezia Terme International Airport (IATA: SUF; ICAO: LICA), located 2 kilometers north of the LMT observatory, has a 10/28 (100/200 °N) runway orientation and local air traffic is subject to the same wind regime. Vertical wind profiles were characterized in a consequent study [109], which integrated previously gathered data on preliminary PBL characterization at the site [102]. A cross-study on multiple southern Italian stations also performed a PBL characterization via WRF modeling and related methodologies, however, the study was limited to one month of data from the 2009 summer season and did not evaluate GHGs or aerosols [110].

Measurements of greenhouse gases and other key parameters at the site started in 2015. The first study evaluating the results, Cristofanelli et al. (2017) [111], provided new insights on the characterization of LMT as what we would be later defined as a “multisource” site. The study also indicated that local wind circulation patterns, characterized in earlier studies [107,108], have a direct influence on LMT observations: western-seaside winds generally yield low concentrations, while northeastern-continental winds are linked with higher mole fractions of GHGs. With respect to aerosols, a more detailed description was performed in Donateo et al. (2018) [112].

Seven years (2016-2022) of methane data at LMT have been evaluated in D’Amico et al. (2024a) [113] and confirmed the influence of wind regimes on CH₄ mole fractions with additional detail and also provided new insights on seasonal cycles: the highest mole fractions The study also found evidence of a HBP (Hyperbola Branch Pattern), with the highest mole fractions linked to low wind speeds and, vice versa, the lowest fractions linked to high speeds. The study D’Amico et al. (2024d) [114] on nine years (2015-2023) of surface ozone (O₃) data however demonstrated that these patterns do not apply to all parameters, as ozone maxima are linked to spring/summer diurnal winds from the western sector. These opposite patterns observed at LMT underlined the complexity of parameters and how they combine with local wind circulation.

Figure 1. A: Modified Copernicus Digital Elevation Model [115]. B: Modified EMODnet [116] highlighting LMT’s location in Southern Italy. C: Google Earth map, tilted by 70°, showing the observation site and key infrastructural/emission hotspots in the area. The “Highway” label indicates a point where the distance between LMT and the highway is ≈4.2 kilometers. The “Lamezia Terme” label points to the town center. The “Station” label points to the busiest train station in the municipality of Lamezia Terme, the central one (Lamezia Terme Centrale).

Figure 1. A: Modified Copernicus Digital Elevation Model [115]. B: Modified EMODnet [116] highlighting LMT’s location in Southern Italy. C: Google Earth map, tilted by 70°, showing the observation site and key infrastructural/emission hotspots in the area. The “Highway” label indicates a point where the distance between LMT and the highway is ≈4.2 kilometers. The “Lamezia Terme” label points to the town center. The “Station” label points to the busiest train station in the municipality of Lamezia Terme, the central one (Lamezia Terme Centrale).

Research studies on LMT data also focused on weekly trends under the assumption that they can only be anthropic in nature, unlike natural trends such as the daily, seasonal, and yearly cycles. In D’Amico et al. (2024c) [117], the first COVID-19 lockdown period of 2020 was used to further assess local sources in a context of exceptionally limited anthropic activities. In fact, the restrictions introduced by the Italian government at the time [118] preceded similar measures issued by other European countries by days or even weeks, thus allowing changes in LMT data to be linked to local changes in emission sources. A cross study on aerosol data from multiple southern Italian stations, LMT included, was performed by Donateo et al. (2020) [119] and provided new insights on local vehicular traffic influences on aerosol diffusion. A more detailed local characterization of weekly patterns was performed in D’Amico et al. (2024b) [120] and demonstrated different behaviors of GHGs and aerosols, possibly linked to changes in anthropogenic activities throughout the week.

Due to LMT’s location in the central Mediterranean sector, observations are also known to be affected by Saharan dust events [121] and open fires [122].

2.2. Instruments, Methodologies, and Datasets

To retrieve the aerosol backscatter profiles at the monitoring site, a Lufft CHM 15k Nimbus ceilometer (Fellbach, Germany) operating within ALICEnet (Italian Automated LIdar–CEilometer network) was used. The network is coordinated by CNR-ISAC in partnership with other Italian research institutions and environmental agencies. ALICEnet measurements are currently usefully employed to detect the altitude and temporal evolution of cloud layers and track the transport of polluted or mineral dust aerosol plumes at different sites along the Italian peninsula. This ceilometer is a ground-based, monostatic, active remote sensing instrument based on the LiDAR (Light Detection and Ranging) technique and principle [123].

The Nimbus ceilometer observes backscattered profiles with a vertical resolution of 15 meters over 24 continuous hours per day. Data are averaged every 15 seconds. The operating range is between the surface (~15 meters) to 15000 meters. Technical details of the instrument are shown in Table 1.

Specifically, the LiDAR technique is used by the Nimbus to emit short light pulses facing directly upward. Cloud layers, precipitations, and aerosols present in the column scatter back the pulses. By analyzing flight time, the intensity of backscattering, and counted pulses, the main features of the vertical column can be determined. The output is used to evaluate PBLH influences on tropospheric greenhouse gas and aerosol transport and dispersion [124]. In particular, the topmost aerosol layer detected by ceilometer pulses and consequent backscatter is used to represent the PBLH itself [125]. The expression used to receive the normalized backscattered signal power is given below (Eq. 1) and is provided by the manufacturer [126]:

P _{c( r)} = (P_raw -b)/(C_s*O(r)*P_calc)

(1)

where P_c(r) is normalized backscattered signal power, P_raw is raw backscattered profile (photon counts), b is the baseline, C_s is calibration constant, O(r) is the overlap function, and P_calc is average test pulse intensity. The output is calibrated in order to attenuate backscattered pulse results. In this research study, negative attenuated signals have been filtered out.

During diurnal time ranges, dense (and/or multiple) cloud or aerosol layers may alter the backscattering signal and generate noise due to attenuation. In order to optimize the accuracy of results, backscattering profiles data are 5 minutes averaged to improve the signal-to-noise ratio.

In addition to ceilometer outputs, additional data have been gathered and processed to assess correlations between PBLH, greenhouse gases, pollutants, and key meteorological parameters. All data have been aggregated on an hourly and daily basis to allow direct comparison between different parameters.

Particulate matter (PM) data in micrograms per cubic meter (μg/m³ or μg PCM) have been gathered by a Palas Fidas 200 S (Karlsruhe, Germany). The instruments provides both Cn (particle numbers) and PM concentrations, however in this study only the latter have been used. A Sigma-2 sampling head draws in ambient air with a flow rate of approximately 0.3 m³/h. Aerosol gathered from ambient air passes through the sampling tube, equipped with a drying section meant to prevent measurement distortion attributable to moisture particle, and is ultimately drawn by the aerosol sensor. Via a Lorenz-Mie light analysis, particle size is determined by the instrument. Particles move through an optically differentiated volume that is illuminated by a polychromatic LED source. In response, each particle emits an impulse of scattered light that is detected by the instrument at angles of 85 to 95 degrees. Measurements are performed every ≈5 seconds and the results have been aggregated on hourly and daily bases, differentiated by particle size.

Data on downward solar radiation in watts per square meter (W/m²) have been gathered by a Kipp & Zonen radiometer, model CNR4. The radiometer used at LMT relies on two pyrgeometers and two pyranometers to measure upward (LW, 4.5–42 μm) and downward (SW, 0.31–2.8 μm) irradiance. Uncertainty in these measurements, traced to the BSRN (Baseline Surface Radiation Network) standard, is approximately 1% [127]. Additional details are available in Lo Feudo et al. (2015) [128] and Romano et al. (2017) [103].

Carbon monoxide (CO), carbon dioxide (CO₂), and methane (CH₄) mole fractions in ppm (parts per million) have been gathered by a Picarro G2401 (Santa Clara, California, USA) CRDS (Cavity Ring-Down Spectrometry) analyzer [129]. Via the principle of CRDS, these carbon compounds in the atmosphere are measured with high degrees of precision. At LMT, the G2401 gathers continuous data and is subject to WMO-compliant calibration cycles. Additional details on G2401 data gathering, calibration procedures and quality assurance are available in Cristofanelli et al. (2017) [111], D’Amico et al. (2024a) [113], D’Amico et al. (2024b) [120], and D’Amico et al. (2024c) [117].

Key meteorological surface data have been gathered by a Vaisala WXT520 (Vantaa, Finland). The instrument relies on ultrasounds to monitor wind direction and speeds, and a transducer to gather temperature data. Additional details on WXT520 measurements at Lamezia Terme are available in D’Amico et al. (2024a) [113]

Equivalent black carbon (eBC) micrograms per cubic meter (μg/m³ or μg PCM) have been measured by a Thermo Scientific 5012 (Franklin, Massachusetts, USA). The instrument operates as a MAAP (Multi-Angle Absorption Photometer), measuring the short-wave absorption parameters of aerosol [130,131,132]. Specifically, the aerosol absorption coefficient (sa), and equivalent black carbon are measured at 637 nm. Additional details are available in Calidonna et al. (2020) [121], D’Amico et al. (2024b) [120], and D’Amico et al. (2024c) [117].

Particle scattering and hemispheric backscattering coefficients at 450, 525, and 635 nanometers (nm) were measured by a LED-based integrating nephelometer (model Aurora 3000, ECOTECH, Knoxfield, Australia) at a temporal resolution of one minute. Air sampling was obtained from the top of a stainless steel tube having a 15 mm internal diameter and a length of ≈1.5m. The inlet was fitted with a funnel covered by a screen to prevent rain drops and arthropods from reaching the sample line. No aerosol size cut-off was applied to the sampled air, and a relative humidity threshold of 60% was set by a processor controlled automatic heater inside the nephelometer to prevent the hygroscopic effects that enhance particle scattering.

Each instrument and its respective dataset has been processed and quality-checked. Data coverage rates (%) compared to the actual number of days (153) and hours (3672) elapsed between May 1^st 2024 and September 30^th 2024 are shown in Table 2.

Using meteorological data, days have been divided into four categories (plus a fifth “NA” category not matching any of the requirements, which is hereby excluded). The “Breeze” regime has been defined as alternating wind directions (WD) between 270 and 70 degrees N, and wind speeds (WS) in the 2-6 m/s interval. The “Eastern Synoptic” regime is defined with a constant WD of 70°, and WS ≥6 m/s. The “Western Synoptic” has the same WS threshold as its eastern counterpart, but the WD is 270°. The “Not Complete” breeze regime is applied to days with a growing breeze regime which was later overcome by a regime leaning to western synoptic characteristics. In this case, the WS threshold is within 4-8 m/s. All reported WD have an applicability range of ±15°. Table 3 shows how days whose wind regime falls in one of the above-mentioned categories are distributed over the observation period.

Analyses have been performed in R 4.4.0 using ggplot2, ggpubr, tidyverse, openair [133,134], and dplyr packages/libraries. Ceilometer data have been processed using MATLAB 2016a to get 2D backscattered profiles [102]. Quantile-quantile plots have also been computed in MATLAB.

3. Results

3.1. Daily Variability During the Observation Period

The campaign evaluated in this study accounts for five months of data gathering at LMT, from May 1^st to September 30^th 2024. As described in section 2.1, the Lamezia Terme regional WMO/GAW station is affected by cyclic wind patterns. Figure 2 shows a wind rose based on hourly-aggregated wind speeds and directions.

Data on CO, CO₂, CH₄, eBC, and PM have been gathered on a daily basis to highlight possible trends during the campaign. As reported in Table 2, the coverage rates of these findings vary depending on instrument maintenance and quality assurance. Figure 3 shows hourly aggregated data of all five parameters, with PM data being further divided into multiple sub-categories.

The same analysis has been applied to meteorological and environmental data. Figure 4 shows the daily averages of primary data gathered at LMT during the campaign.

In addition to daily averages, hourly aggregations have also been computed to show variations between May and September 2024 with enhanced details. Figure 5 and Figure 6 show these aggregates for GHG/aerosol and environmental parameters, respectively. Y axes scales have been adjusted to account for observed peaks.

3.2. Daily Cycles

Research studies on LMT data highlighted the existence of daily cycles [111,113,114], which are affected by local wind circulation patterns. Up until the campaign presented in this research, these cycles were not shown considering the influence of wind regimes. Figure 7 therefore shows an enhanced evaluation of the LMT daily cycle of greenhouse gases and aerosol accounting for the four categories described in section 2.2. Figure 8 shows the daily cycle of key environmental and meteorological parameters.

3.3. Percentile Roses

In D’Amico et al. (2024d) [114], percentile roses have been used to correlate surface ozone concentrations at LMT with specific wind directions. The same method has been used in this research for CO, CO₂, CH₄, eBC, and PM, with the results being reported in Figure 9. PM_2.5 and PM₁₀ have been plotted separately.

3.4. PBL Variability and Cycles

Following the evaluation seen in the previous sections with respect to GHGs, aerosols, meteorological, and environmental parameters, the PBLH at Lamezia Terme has been also characterized using daily and hourly averages. The results are shown in Figure 10.

Vertical profiles of select days deemed representative of all wind regimes have also been plotted using Nimbus data. The results are shown in Figure 11.

3.5. Correlations Between PBLH, Gases and Aerosols

Scatter and quantile–quantile plot (qplots) of observed parameters (x axis) have been correlated with PBLH (y axis) in Figure 12, Figure 13, Figure 14, Figure 15, Figure 16 and Figure 17. Colored squares indicate the amount of data in each 0.1x0.1 bin. Green lines represent the quantile–quantile plots, and red lines the least-square best fits. For each parameter, the top plots show breeze and not complete breeze regimes, while the bottom ones show western and eastern synoptic flow data, respectively.

4. Discussion

A campaign focused on Planetary Boundary Layer (PBL) characterization has been performed at the Lamezia Terme regional WMO/GAW observation site in Calabria, Southern Italy. The campaign relied on a new ceilometer (Table 1), a longer observation span compared to previous comparable research, and the integration of additional data (specifically, greenhouse gases and aerosols, see Table 2) meant to test correlations that were not applicable during the summer 2009 campaign described in Lo Feudo et al. (2020) [102].

The preliminary evaluation of environmental parameters allowed to define the fundamentals upon which the findings of this campaign would be based on. Local wind circulation patterns are oriented on a W/NE axis [107,108], as shown in Figure 2. Greenhouse gas and aerosol data coverage, in addition to meteorological parameters, show both daily (Figure 3 and Figure 4) and hourly (Figure 5 and Figure 6) variability during the observation period.

Following the analyses seen in other research papers focused on LMT data [111,113,114,117], daily cycles of greenhouse gases, aerosols, and meteorological parameters have been assessed. LMT has been proven in multiple studies to have a peculiar daily cycle and this research effectively integrates a new variable, peplosphere height variability, into these cycles.

In the case of GHGs and aerosols, the analysis has considered four distinct wind regimes, each defined by specific direction (WD) and speed (WS) thresholds: breeze, not complete (NC) breeze, eastern synoptic, and western synoptic (Table 3). In the case of CO₂ (Figure 7B) and CH₄ (Figure 7C), although the breeze and NC breeze yield higher values during night-time hours, diurnal concentrations are lower and do not show differences in terms of wind regime. However, CO (Figure 7A), eBC (Figure 7D) and PM (Figure 7E, 7F, 7G) show a greater variability in morning and evening hours that is compatible with inversions in wind patterns from the northeastern-continental to the western-seaside corridors, and vice versa. PM (Figure 7F, 7G) is particularly susceptible to night-time fluctuations. The susceptibility of pollutant concentrations to inversions in wind patterns was first reported in Cristofanelli et al. (2017) [111] and further supported in D’Amico et al. (2024c) [117], which analyzed the first 2020 COVID-19 lockdown in Italy to highlight the nature of pollutant peaks linked to wind circulation in lockdown versus ante/post lockdown periods.

Key meteorological and environmental parameters show well-defined daily cycles (Figure 8). In particular, SCA values (Figure 8D) show a peak linked to increased pollutant inversions that occur during inversions.

The percentile roses in Figure 9 differentiate between gases and aerosols in terms of spatial distribution. Two types of behavior are reported: minor spatial variations, as seen in the case of CO₂ (Figure 9B) and CH₄ (Figure 9C); intermediate variations, as in the case of CO (Figure 9A), eBC (Figure 9D), and PM_2.5 (Figure 9F); substantial variations, seen specifically in PM₁₀ (Figure 9G). These differences reflect not only the susceptibility of each parameter to wind regimes, but also indicate distinct emission sources. CO and eBC, which are both effective tracers of combustion processes, were linked to local wildfires in a previous research study by Malacaria et al. (2024) [122]. CH₄ is characterized by both natural and anthropogenic sources, and local peaks are linked to northeastern-continental winds [113]. In Cristofanelli et al. (2017) [111], methane peaks at LMT were specifically attributed to sources located nearby.

Particulate matter and reported differences based on particle size are indicators of distinct emission sources. In fact, as described in section 1, PM is heterogeneous in nature. Anthropogenic PM_2.5 emissions are generally linked, for example, to fossil fuel burning and domestic heating, while PM₁₀ is linked to natural sources such as sea spray, biomass burning attributable to agriculture, Saharan dust events [121], and wildfires [122]. Among the anthropogenic sources of PM₁₀, incomplete fossil fuel burning processes are common. The findings of this research study allow to narrow down the nature of observed PM depending on the four wind regimes that affect LMT observations: the breeze is linked to diurnal sea spray from the west and nocturnal smog from the northeast; the not complete (NC) breeze results into a southern corridor and PM peaks that are hereby interpreted as biomass burning linked to agricultural activities; the western synoptic combines sea spray outputs, combined with continental-anthropogenic PM; the eastern synoptic is linked to all PM categories.

In Figure 10, PBLH data have been plotted to show the overall variability during the observation period, as well as the daily cycle (Figure 10C). This cycle in particular shows daily PBLH variations dependent on the four distinct categories of wind regimes, which in turn have an impact on gas and aerosol concentrations observed at LMT. Examples of the four categories are shown in Figure 11.

Qplots testing the correlation of PBLH, gases and aerosols have been plotted in Figure 12 through 17 under all four wind regimes. An apparent inverse relationship or anticorrelation between PBLH and CO₂ and CH₄ concentrations can be seen in Figure 13 and Figure 14. This suggests that as the PBLH increases, CO₂ and CH₄ concentrations tend to decrease, indicating that lower peplosphere heights correspond to limited vertical mixing and the accumulation of these gases near the surface.

When synoptic conditions are favorable to sea breeze development, colder air masses from the sea with low marine aerosol content are advected over land in the early morning and interact with the nighttime boundary layer. After the onset of the sea breeze, an internal boundary layer develops from the coastal discontinuity and the height of the maximum backscatter threshold from the Nimbus decreases, likely due to the advection of the marine aerosols above the PBL, thus creating a discontinuity in aerosol concentration and size distribution. Later in the morning, when the breeze is well developed, the convection takes over and mixes marine and continental aerosols, creating a homogeneous content of aerosols filling the convective layer.

During stationary synoptic flow with wind speed typically >5 m/s, marine aerosols are mixed with their continental counterparts and the height of the boundary layer detected by LMT’s ceilometer remains constant. During sea breeze days, at the onset of the breeze, the findings of this research and past studies indicate that the sea breeze advection of marine aerosols causes a nonhomogeneous columnar distribution, inducing a low LiDAR signal-to-noise ratio above the internal boundary layer [102].

Figure 12, Figure 13, Figure 14, Figure 15, Figure 16 and Figure 17 show that the advection of cleaner air with marine aerosols from the sea with respect to the land aerosols, flowing above the internal boundary layer, after the breeze onset, causes a vertical discontinuity of aerosol concentration and thus a reduction of the LiDAR vertical range. In case of breeze, the resulting values of PBL vary as expected following the daily cycle over land, i.e., low during the stable night and high during the unstable days: during synoptic flow, the PBL is constantly higher compared to unstable conditions. It is therefore reported that during the night, synoptic PBLH is larger than during the day. A reason can be that during summer the breeze always develops, adding and modulating the synoptic flow (Figure 11). A well-developed breeze, adding speed, would produce a large quantity of marine aerosols advected on land flowing over the land; aerosol layer would contribute to the detection of a lower PBLH.

Therefore, the findings of this research provide evidence of direct peplospheric influences over the surface concentrations of greenhouse gases and aerosols.

5. Conclusions

This research work relied on an enhanced characterization of peplospheric variability at the Lamezia Terme (code: LMT) regional coastal WMO/GAW observation site in Calabria, Southern Italy. The study is based on five continuous months (May-September 2024) of PBL data obtained by a Nimbus ceilometer, plus greenhouse gas, aerosol, meteorological, and environmental data gathered by other instruments. The integration of multiple datasets has ensured a more detailed understanding of PBL variability at the site, as the previous campaign on PBL data was focused entirely on one month of observations performed during the 2009 summer season, and no correlations with the surface concentrations of key pollutants were possible at the time.

This study has introduced four distinct categories of wind regimes (breeze, not complete breeze, eastern synoptic, western synoptic) and tested the correlation of PBLH variability with the concentrations of carbon monoxide (CO), carbon dioxide (CO₂), methane (CH₄), equivalent black carbon (eBC), and particulate matter (PM) under each regime.

Overall, these findings demonstrate the need to effectively integrate PBLH data and wind regimes in urban air quality assessments due to the influence of PBLH patterns on the surface concentration of key pollutants. Such a study demonstrates the importance of correlating circulation, complex orography and boundary conditions at the interface between land and sea when dealing with the assessment of natural and anthropogenic pollutants. The availability of physical/chemical and aerosol data allows to test the correlation between PBLH variability and a number of factors.

The LMT observation site is therefore a challenging natural laboratory for capturing different circulation patterns in which natural and anthropic-induced pollutant concentrations show different behaviors. Future research integrating the findings shown in this study with additional parameters will further demonstrate the importance of considering all these factors in sustainable policies/regulations and air quality monitoring.