Expanding Continuous Carbon Isotope Measurements of CO₂ and CH₄ in the Italian ICOS Atmospheric Consortium: First Results from the Continental POT Station in Potenza (Basilicata)

Abstract

Carbon isotope fractionation is an efficient tool used for the discrimination and differentiation of sinks and emission sources. Carbon dioxide (CO₂) and methane (CH₄) are among the key drivers of climate change, and a detailed evaluation of variations in the ¹³C/¹²C ratio in either compound provides vital information for the field of atmospheric sciences. The Italian atmospheric ICOS (Integrated Carbon Observation System) consortium is currently implementing δ¹³C-CO₂ and δ¹³C-CH₄ measurements, with four observation sites now equipped with Picarro G2201-i CRDS (Cavity Ring-Down Spectrometry) analyzers. In this work, results from the first two months of measurements performed at the Potenza station in southern Italy between 20 February and 20 April 2025 are presented and constitute the first evaluation of continuous atmospheric δ¹³C-CO₂ and δ¹³C-CH₄ measurements from an Italian station. These results provide a first insight on how these measurements can improve the current understanding of CO₂ and CH₄ variability in the Italian peninsula and the central Mediterranean sector. Although preliminary in nature, the findings of these measurements indicate that fossil fuel burning is responsible for the observed peaks in CO₂ concentrations. CH₄ has a generally stable pattern; however, abrupt peaks in its isotopic delta, observed during March, may constitute the first direct evidence in Italy of Saharan dust intrusion affecting carbon isotope fractionation in the atmosphere. This study also introduces an analysis of the weekly behavior in isotopic deltas.

1. Introduction

The stable isotopes of carbon differ in their number of neutrons while maintaining the same electronic configuration and chemical behavior; ¹²C (carbon-12), which constitutes approximately 98.9% of the carbon found in nature, is predominant compared to ¹³C (carbon-13), whose abundance is estimated at around 1.1% [1,2]. Unlike the radioactive isotope ¹⁴C, also referred to as radiocarbon and used as an effective tracer to discriminate natural and anthropogenic sources [3,4,5,6,7], these isotopes are not subject to radioactive decay, thus ensuring their nuclear stability. The isotopic ratio δ¹³C is referenced against an international standard named VPDB (Vienna Pee Dee Belemnite), based on the ¹³C/¹²C ratio observed in a marine fossil belonging to the species Belemnitella americana [8] of the genus Belemnitella [9] discovered in the Pee Dee geological formation in South Carolina, USA [10]. Deviations from the VPDB standard are expressed in per mille or per mil (‰) and are widely used in source apportionment [11]. Specifically, δ¹³C is valuable in identifying carbon cycle dynamics [12], tracing seasonal biochemical processes [13,14,15,16], and interpreting isotopic signals in paleoclimatic and geological studies [17,18,19,20,21,22]. Stable carbon isotopic ratios have also been used to study the first possible evidence of life on Earth, pinpointed from the biogenic isotopic fingerprint observed in carbon inclusions within zircon minerals dating to the Hadean Eon (4.1 billion years ago) [23].

Stable carbon isotopes are closely tied to the understanding of climate change, serving as tracers for the processes involved in the carbon cycle and greenhouse gas dynamics [24]. Isotopic signatures, or the relative distribution of isotopes in a given sample, provide a sort of “fingerprint” unique to various natural and anthropogenic processes, enabling the identification and distinction of sources and emission dynamics of greenhouse gases such as CO₂ and CH₄ [25,26].

For CO₂, δ¹³C allows, for instance, differentiation between carbon emissions from fossil fuels like coal and oil [27,28] and those of natural origin, such as emissions resulting from volcanic activity [29] or peculiar soil respiration processes [30,31]. Furthermore, stable isotopes are highly valuable tools in studying the dynamics of the carbon cycle: by analyzing isotopic variations, it becomes possible to understand how oceans and forests act as sinks, absorbing CO₂ from the atmosphere, or as sources, releasing carbon under specific conditions [14,15].

The isotopic trends of atmospheric CO₂ and CH₄ have shown significant variations over time, influenced by a combination of anthropogenic and natural factors [32,33]. For CO₂, δ¹³C has exhibited a steady decline, primarily reflecting the increasing contribution of emissions from fossil fuels that are depleted in ¹³C [25,34,35]. Seasonal effects play a key role, as photosynthesis and plant respiration drive annual variations linked to vegetative growth cycles [36,37]. Additionally, phenomena such as El Niño can alter δ¹³C values [38]. Atmospheric background values of δ¹³C-CO₂ are approximately −8‰, with seasonal fluctuations caused by the CO₂ uptake of photosynthesis during the summer and increased anthropogenic emissions during the winter [39,40]. In fact, as fossil fuels are characterized by a fingerprint in the ≈−25 to ≈−30‰ range [41], anthropogenic emissions cause CO₂ concentrations to increase and δ¹³C-CO₂ to decrease [42]; further analyses can discriminate coal emissions from those attributable to gasoline [43,44].

The differentiation of CO₂ sources based on the carbon isotopic ratio relies on the distinct isotopic signature of fossil-derived carbon, the result of photosynthetic isotopic discrimination that occurred during the formation of ancient biomass, wherein autotrophic organisms preferentially assimilated the lighter isotope δ¹²C instead of the heavier δ¹³C [13]. The organic matter thus produced was relatively depleted in δ¹³C and, following long-term geological processes of burial and diagenesis, was converted into fossil fuels [45]. Upon combustion, these fuels release CO₂ retaining the original isotopic depletion, with δ¹³C values typically averaging around −28‰. This contrasts with natural emissions such as those from volcanic activity or contemporary biospheric respiration, which exhibit less negative δ¹³C values.

For CH₄, observed trends indicate a shift in the global budget, with natural sources such as wetlands driving changes in the total isotopic fingerprint of CH₄ [46,47]. Anaerobic methanogenesis processes in wetlands are characterized by isotopic fractionation, during which lighter carbon isotopes (¹²C) are preferentially utilized in biochemical reactions, releasing CH₄ with highly negative isotopic values. The isotopic fractionation reflects the underlying mechanisms of CH₄ production—primarily hydrogenotrophic methanogenesis (via CO₂ reduction), typically yielding δ¹³C signatures as low as −80‰ or lower, and acetoclastic methanogenesis (from acetate fermentation), which results in relatively less negative values, often between −65‰ and −50‰. This underscores the variability and complexity of natural contributions [48]. The processes that drive changes in the global CH₄ budget are still not fully understood, and the study of isotopic fractionation in CH₄—also accounting for ¹⁴C—has become a key tool to constrain current uncertainties in models [49]. δ¹³C-CH₄ values are generally more depleted than their δ¹³C-CO₂ counterparts: pyrogenic emissions are in the −13 to −25‰ range, while thermogenic emissions are in the −25 to −55‰ range and biogenic emissions are particularly depleted in δ¹³C, reaching the −55 to −70‰ range [50,51,52]. Fossil fuel-related CH₄ can be isotopically distinguished between two principal sources: fugitive emissions and those resulting from incomplete combustion. The former arise from unintentional releases occurring during the extraction, processing, and transportation of natural gas and petroleum: this thermogenic CH₄ is typically characterized by δ¹³C values ranging from −25‰ to −55‰, reflecting its deep geological origin [49]. In contrast, CH₄ produced via the incomplete combustion of fossil materials (coal, petroleum, wood, gasoline, etc.)—commonly referred to as pyrogenic methane—exhibits less negative δ¹³C signatures, generally within the −13‰ to −29‰ range, and presents a more heterogeneous isotopic composition, strongly influenced by the nature of the combusted material and by combustion efficiency [53].

The current δ¹³C-CH₄ value for the atmospheric background is ≈−47‰ [54], with a reported decline to values closer to the −47.5‰ mark [55]. Minor differences have been reported between the northern and southern hemisphere [56].

Analytical techniques for measuring stable isotopologues are primarily distinguished based on their measurement principles, grouping into isotope ratio mass spectrometry (IRMS) [57] and laser absorption spectroscopy (LAS) techniques [58]. The IRMS technique is based on the separation of ions according to their mass-to-charge ratio (m/z) using a magnetic field, enabling the direct measurement of isotopic ratios relative to reference standards with a known isotopic composition. This technique requires the conversion of the sample into a gaseous phase and is traditionally used in laboratory settings, although developments such as continuous-flow IRMS (CF-IRMS) [59] have improved analysis speed. Conversely, LAS relies on the absorption of laser light at specific wavelengths by CO₂ molecules, where each isotopologue absorbs light radiation at slightly different frequencies. LAS techniques measure the mixing ratio of individual isotopologues, from which isotopic ratios are calculated, and include various methodologies such as tunable diode laser absorption spectroscopy (TDLAS) [60], quantum cascade laser spectroscopy [61,62] (QCLAS), cavity ring-down spectroscopy (CRDS) [63,64,65,66], off-axis integrated cavity output spectroscopy (OA-ICOS) [67], and Fourier transform infrared spectroscopy (FTIR) [68]. LAS techniques are generally better suited for in situ measurements and offer a higher measurement frequency compared to IRMS.

In Italy, measurements of δ¹³C were limited to specific circumstances, such as an evaluation of the isotopic fingerprint of Mount Etna’s CO₂ outputs [69]. Until recently, the country lacked a well-established and organized consortium of atmospheric stations performing continuous δ¹³C measurements of CO₂ and CH₄.

In this work, we present the first results of the measurements performed near Potenza, southern Italy, at the Institute of Methodologies for Environmental Analysis (Istituto di Metodologie per l’Analisi Ambientale, IMAA) of the National Research Council of Italy (Consiglio Nazionale delle Ricerche, CNR) as the new Class 1 atmospheric station in the Integrated Carbon Observation System (ICOS) research infrastructure [70]. These results constitute the first step towards the full implementation of these measurements in the developing cross-country network. A recent study [71] highlighted the addition of the Potenza site as an extra step towards filling notable gaps in the southern European network of ICOS stations, which are not equally distributed across the continent. In addition to presenting the preliminary results, this study relies on the methodology first introduced by C.D. Keeling to pinpoint the isotopic fingerprint of sources [72], thus allowing us to provide an insight into source apportionment at the site. Therefore, this study constitutes a preliminary report on the expansion of advanced greenhouse gas measurements in the central Mediterranean, integrating existing research with new insights into source apportionment and the attribution of observed peaks to natural or anthropogenic sources.

This work is divided as follows: Section 2 describes the Italian network of continuous carbon isotope measurements and the main characteristic of the POT site; Section 3 describes the instrument and methodologies used for these measurements; Section 4 shows the results of the measurements; and Section 5 and Section 6 discuss the results and conclude the paper, respectively.

2. The National Consortium, with a Focus on POT

The ICOS Italian consortium (Figure 1A) is expanding its instrumentation capacity by integrating continuous carbon isotope measurements into its atmospheric research consortium. Currently, four monitoring stations are equipped with Picarro G2201-i analyzers (Santa Clara, CA, USA) to measure δ¹³C-CO₂ and δ¹³C-CH₄, strengthening the consortium’s alignment with European research efforts in atmospheric composition studies. The first station, Lampedusa (LMP), operated by ENEA (Italian National Agency for New Technologies, Energy and Sustainable Economic Development) and located in the Strait of Sicily, 135 km east from Tunisia, initiated these measurements, setting the foundation for long-term isotopic monitoring. Subsequently, under the ITINERIS project (Italian Integrated Environmental Research Infrastructures System), three additional stations belonging to the national consortium were equipped with the isotope analyzer: In July 2024, the Lamezia Terme (LMT), a regional coastal World Meteorological Organization—Global Atmosphere Watch (WMO/GAW) station in the region of Calabria was equipped with the G2201-i analyzer, generating a dataset of continuous measurements (July–December) [73]. The northernmost station, Monte Cimone (CMN), is also performing continuous measurements of stable carbon isotopes in CO₂ and CH₄, further expanding the consortium’s geographical coverage. The latest addition, the ICOS POT station (February 2025) (Figure 1B,C), completes the current national consortium, ensuring broader observational capabilities.

POT is a new ICOS Class 1 atmospheric station and is part of the CIAO (CNR IMAA Atmospheric Observatory) [74,75], which hosts two ESFRI (European Strategy Forum on Research Infrastructures) research infrastructures (RIs) involved in atmospheric characterization: ICOS and Aerosols, Clouds, and Trace Gases Research Infrastructure (ACTRIS). These two pan-European RIs are dedicated to studying atmospheric phenomena from the near surface to the lower stratosphere, producing high-quality, standardized, and open-access data [76,77].

As of today, the site has successfully completed Step 1 of the ICOS labeling process. This indicates that both the site and its infrastructure have been assessed and approved by ICOS. To achieve official recognition as an operational ICOS station, the final phase, Step 2, must still be completed. This process is currently underway and requires all instruments to be in regular operational mode.

The station is situated on a plain enclosed by low mountain ranges and is located within 150 km of the western, southern, and eastern coastlines of the Mediterranean Sea. This positioning exposes the station to a typical mountainous climate significantly influenced by Mediterranean atmospheric dynamics: the resulting climate pattern is characterized by predominantly arid, hot summers, and cold winters. The surrounding area mainly consists of non-irrigated arable land, forests, and grasslands and, accordingly with ICOS recommendations, the site is located far from large cities or heavy anthropogenic sources that could affect GHG observations: the closest urban center is the regional capital of Potenza (~9 km NE, 819 m a.s.l.; 64.100 inhabitants, 365 inhabitants/km²), and some small villages (Tito, the most populous one, has less than 7500 inhabitants) are at about 10 km from the site. A high-speed road is located 1 km north of the site while the nearest highway (A3) is about 30 km west of the site.

This positioning exposes the station to a typical mountainous climate significantly influenced by Mediterranean atmospheric dynamics. The measurement station consists of a 104 m tall tower with sampling points at 100 m, 50 m, and 10 m heights, in accordance with ICOS requirements for continental stations. The top sampling level (≥100 m) is used to avoid local influences and collect data on atmospheric processes across broader areas, enabling insights into regional greenhouse gas sources and sinks [78]. Air sampling at lower levels (10 m and 50 m) helps to analyze the vertical distribution of greenhouse gas concentrations to understand local phenomena. Situated in the southern region of the Italian peninsula, the station holds a strategic position within the Mediterranean basin, linking the hints from atmospheric circulation from different European region domains and allowing the observation of interactions and transitions between them, providing insights into atmospheric dynamics, gas transport, and other environmental phenomena that may not be evident when measuring in only one of the regions. Measurements of stable carbon isotopes from tall towers have provided numerous details on the extent and nature of anthropogenic and natural emission sources at various scales [79].

In POT’s case, whose measurement footprint—shaped by its configuration as a tall-tower atmospheric station—extends over a broad and regionally representative area, similar to what has been observed at sites like LMP and CMN [80,81], stable isotope measurements would provide the first estimate of δ¹³CO₂ and δ¹³CH₄ background levels in the central Mediterranean Basin.

3. Data Gathering and Evaluation Methodology

The measurements evaluated in this study were performed for 60 continuous days, from 20 February to 20 April 2025. A 60-day window was selected as it represents a sufficiently long period to capture meaningful isotope variability while remaining well within the manufacturer’s recommended calibration interval. Measurements were taken using a Picarro G2201-i (Santa Clara, CA, USA) CRDS analyzer; the dataset used in this work is available on the ITINERIS (Italian Integrated Environmental Research Infrastructures System) HUB [82]. Air was sampled from the top level of the tower (100 m) by Synflex 1300 aluminum/polyethylene composite tubing of 12 mm OD (EATON, Dublin, Ireland). Two vacuum pumps were used: one pump, KNF model KTE N815 (KNF, Schenkon, Switzerland), drew air from the top level of the tower, while the other one (instrument pump) delivered the sampled air to the analyzer. Sampled air passes through two filters: One of 2 mm porosity (M&C F2, Ratingen, Germany) provides protection from particulate and allows us to also detect possible water intrusion due to tubing damages, thanks to the external glass bulb. The second filter of 0.5 mm porosity (Swagelok, Solon, OH, USA) is placed at the inlet of the analyzer and serves to protect the instrument’s internal filter from any intrusion of finer particulate matter. The analyzer also measures the water vapor content of the sample, but internal water correction is insufficient for this type of analyzer [83]. To avoid artifacts in the measurements, a Perma Pure MD 070-144 S-4 Nafion dryer (Perma Pure, Lakewood, NJ, USA) was connected in line. The flow rate of the sampled air into the instrument was approximately 25 mL/min.

The Picarro G2201-i used at POT station operates in CO₂-CH₄ simultaneous mode, which allows us to measure CO₂ and CH₄ mole fractions, as well as the carbon isotope fractionation of both compounds. The reported precision in this mode for δ¹³C-CO₂ is <0.16‰, while that of δ¹³C-CH₄ in High Precision (HP) is <1.15‰. The HP mode of CH₄ has been selected due to measured concentrations being in the HP range (1.8–12 ppm); High Dynamic Range mode (HR) is recommended for higher concentrations (10–500 ppm), which were not observed during the entire measurement period. The instrument performs one measurement per second. All measurements were aggregated on both an hourly and a 10 min basis to generate a dataset yielding a 98.47% (hours) and 98.26% (10 min blocks) coverage rate during the observation period (20 February–20 April). During the entire measurement period, H₂O was constantly below the Guaranteed Spec Range limit of 2.4% recommended by the manufacturer: the highest reported value of water vapor was in fact 0.055%, with an average of 0.013% ± 0.016% (one standard deviation). All data were processed using an algorithm developed by the CNR-ISAC (National Research Council—Institute of Atmospheric Sciences and Climate) Lamezia Terme, Italy, to generate a G2201-i dataset in the Calabria neighboring region [73]. The algorithm parses through raw data and excludes from further processing all measurements that do not have optimal status flags.

The instrument started data gathering operations shortly before the beginning of the measurements evaluated in this study, and its initial conditions reflected the manufacturer’s setup and configuration. This guarantees the quality of the data and the absence of drifts for the whole period considered in our study (60 days): the manufacturer recommends long-term calibration via the primary standard every 3–4 months [84].

In order to verify the stability of measurements, G2201-i mole fractions of CO₂ and CH₄ corrected for H₂O (“dry”) were compared with the Picarro G2401 CRDS counterparts operating at POT at the same time. A total of 1015 h of measurements of both instruments were compared using the Bland & Altman methodology [85,86] using the blandr package and library in R 4.5.1 [87]. The graphical results yielded by the method (Figure 2) and the statistics of the comparison (

Table 1. Statistical results of the Bland & Altman method applied to G2201-i and G2401 measurements of CO₂ and CH₄ at POT station.

Statistics	CO2	CH4
Max. Means	446.36	2.08
Min. Means	420.45	1.96
Max. Difference	35.63	0.13
Min. Difference	−16.21	−0.08
Bias	1.71	−0.00 (5)
Bias std. dev.	3.20	0.00 (9)
Bias up 95% CI	1.93	−0.00 (4)
Bias low 95% CI	1.48	−0.00 (5)

) indicate good agreement between the two instruments. These results do not constitute an alternate procedure for G2201-i calibrations; however, they provide information on the stability of measurements during the study period at POT.

This analysis aims to provide an initial insight on carbon isotope variability over the country and to gather information about the ranges in isotopic deltas that should be considered in future calibration procedures accordingly to the recommendations of both the manufacturer [84] and—specifically for CH₄—the IAEA (International Atomic Energy Agency) [88].

Daily [89] and weekly [90] cycles were investigated based on the methodologies seen in previous studies from the other Italian sites equipped with the same kind of instrument. The daily cycle evaluated in this study considers, for the first time, the 10 min block aggregation in addition to hourly averages and standard deviations. It is important to note that this analysis is representative of the specific observation period between February and April and does not reflect annual patterns.

4. Results

4.1. Observed Values

Figure 3 and Figure 4 report CO₂ and CH₄ and their isotopic deltas, respectively. Along with the original data collected, a 36 h moving average is reported.

Mole fractions and isotopic deltas show distinct behaviors, which are attributable to changes in the balance between different sources of emissions. Specifically, δ¹³C-CH₄ tendencies are affected by abrupt increases, followed by a gradual decrease.

Table 2. Average concentrations (ppm) and isotopic deltas (‰) of CO₂ and CH₄ during the study period, aggregated on a monthly basis with their respective standard deviation intervals (±1σ).

Month	CO2		CH4
	Conc. (ppm)	δ13C (‰)	Conc. (ppm)	δ13C (‰)
February	435.65 ± 5.58	−7.11 ± 0.39	2.03 ± 0.02	−48.11 ± 0.47
March	431.53 ± 3.91	−6.56 ± 0.44	2.02 ± 0.02	−47.36 ± 1.06
April	429.82 ± 3.47	−5.54 ± 0.22	2.02 ± 0.02	−48.55 ± 1.00

shows the results of these measurements and their respective standard deviation intervals.

4.2. Daily Cycle Variability

Other studies from the national consortium have highlighted peculiar patterns in daily cycle variability, largely driven by local circulation patterns and changes in anthropic activities [89]. In this work, the daily cycle is evaluated with enhanced time resolution via the implementation of 10 min blocks. The results are shown in Figure 5 for CO₂ and CH₄ variability and Figure 6 for their respective isotopic deltas. The data are aggregated on a monthly basis to reflect longer timescale changes during the period.

Overall, the aggregation by month allows us to pinpoint, for CO₂ in particular, that with increasing temperatures and photosynthesis, δ¹³C-CO₂ leans towards higher values. Regarding CH₄, there is no monotonic behavior with time: a decrease is observed from February to March but then values in April are higher than in March. This is probably linked to anomalies observed in March. This reflects also on δ¹³C-CH₄: the highest values are observed in March for all hours of the day.

4.3. Weekly Cycle Variability

The evaluation of weekly cycles has been used in other studies to highlight anthropic influences, as natural processes—unlike daily, seasonal, and annual cycles—are not affected by changes during a standard week [90,91,92]. Weekly cycles (MON-SUN) of CO₂ and CH₄ during the period March–April are shown in Figure 7, while the isotopic deltas are shown in Figure 8.

The weekly data for February were not considered because the sampling period was only eight days.

The plots do not allow us to determine substantial differences between weekdays. Therefore, the data were further evaluated to assess the statistical significance of possible differences in the averages between weekday (WD, MON-FRI) and weekend (WE, SAT-SUN) concentrations/isotopic deltas. In order to apply the non-parametric Kruskal–Wallis test [93], the normality of the data distribution was assessed via the Shapiro–Wilk test [94] in R. CO₂ and CH₄ concentrations, as well as their respective isotopic deltas, which yielded p-values < 0.05, thus indicating that the data did not follow a normal distribution. Consequently, the Kruskal–Wallis tests allowed us to determine the presence of a statistically significant difference between WD and WE concentrations of CO₂, while δ¹³C-CO₂, CH₄, and δ¹³C-CO₂, did not show a consistent weekly cycle.

4.4. Keeling Plot and Source Apportionment

Keeling plots [72] is an effective methodology by which it is possible to pinpoint the isotopic fingerprint of emission sources. In this study, the first Keeling plot computed by the Italian atmospheric monitoring consortium is shown (Figure 9); however, this is limited to CO₂ only, as the analysis of CH₄ did not yield statistically relevant results. The evaluated data are categorized on a monthly basis, an approach employed in research works such as Hoheisel et al. [95].

All regression lines are statistically very significant (p-value < 0.001), while the R² values range between 0.10 (March) and 0.86 (February). The intercepts, which according to the Keeling plot methodology are representative of the characteristic isotopic fingerprint of the observed emission source(s), indicate that increased CO₂ concentrations are compatible with fossil fuel emissions. March’s poor correlation, and its intercept leaning towards a higher δ¹³C-CO₂ value, further corroborate the hypothesis by which the month was affected by anomalous events for the period. Notably, Saharan dust intrusions recorded in mid to late March (10–14, 16–18 and 22–24) contributed to several disturbances in the data and such events were not observed during April. The effects of Saharan intrusions on CO₂ isotopic fractionation, in addition to the effects on CH₄, need to be further evaluated in future research.

5. Discussion

At the new ICOS POT station (Figure 1), the preliminary data gathered by a Picarro G2201-i analyzer of δ¹³C-CO₂ and δ¹³C-CH₄ were evaluated to provide the first insight on isotopic trends in the area. The study of GHGs at scales ranging from global to local is extremely important for the introduction of emission mitigation measures [96]. The importance of carbon isotope fractionation as a tool to discriminate between anthropogenic and natural sources has been known for decades [97]; however, the Italian ICOS atmospheric consortium lacked the infrastructure and instruments to perform these analyses. With a national consortium now in development [73], carbon isotope measurements are expected to increase the level of accuracy in source apportionment efforts performed in Italy. This study presents the first results related to the measurements of stable carbon isotopes in CO₂ and CH₄, representing the first published findings within the national consortium.

The accuracy of the measurements evaluated in the measurements is affected by the current lack of a national standard for isotopic measurements of δ¹³C-CO₂ and δ¹³C-CH₄. The measurements’ duration (60 days) was specifically selected to remain in an acceptable range of uncertainty, as the manufacturer Picarro recommends calibration with primary standards at least once every 3–4 months, in addition to regular calibration procedures [84]. The measurement period is characterized by an optimal coverage rate, over 98% for both the hourly and 10 min aggregations.

Although calibration standards of isotopes are presently not available in the national consortium, the stability of G2201-i measurements at POT was assessed via a direct comparison with a Picarro G2401 CRDS analyzer measuring mole fractions of CO₂ and CH₄ at the same site. The Bland & Altman methodology [85,86], which is commonly used to compare two instruments measuring the same parameter(s), allowed us to verify that G2201-i measurements were in good agreement with their G2401 counterparts (Figure 2, Table 1). This procedure does not replace full calibration using isotopic standards; however, it still allowed us to ensure the stability levels required to perform the assessments presented in this work.

The general tendencies observed during the study period allow us to highlight differences in the behavior of both compounds (Figure 3) compared to their isotopic deltas (Figure 4). CO₂’s mole fractions show a decline typical of transitions from low to high temperatures (Figure 3A), which results in an increase in δ¹³C-CO₂ (Figure 4A). This behavior is consistent with increased photosynthetic activity, which normally peaks during the summer and results in the lowest CO₂ concentrations and the highest δ¹³C-CO₂ values of the year [36]. CH₄ does not show a clear transition between February and April (Figure 3B); however, abrupt changes in δ¹³C-CH₄ are observed during March (Figure 3B), matching the occurrence of three Saharan dust intrusion events that affected Southern Italy in mid to late March [98,99]. These values diverge from those observed in nearby urban areas [100].

In the atmosphere, CH₄ is characterized by more uncertainties in the attribution of sources and sinks, with natural emissions being a key driver of global scale shifts [101]. Changes in δ¹³C-CH₄ may be caused by sinks such as chlorine (Cl) or hydroxyl radical (•HO), which are known to have a characteristic isotopic fingerprint [102,103]. A previous study on CH₄ variability in the neighboring region of Calabria highlighted the potential of δ¹³C shifts in CH₄ as a tool to assess the effects of this sink [89]. The observed changes in CH₄’s isotopic delta could be attributable to the interaction between Saharan dust and sodium chloride (NaCl), which results in mineral dust–sea spray aerosol (MDSA) [104]. The consequent chemical reactions can release reactive Cl radicals [54], which can alter the atmosphere’s oxidizing capacity [105]. This phenomenon may then cause an increase in isotopic deltas [106,107], which requires further investigation.

Potential shifts that occurred during the study period can be seen in Table 2. CO₂ concentrations show a gradual decline, which is compatible with reduced anthropogenic emissions and increased photosynthetic activity in the transition from cold (435.65 ± 5.58 ppm) to warm seasons (429.82 ± 3.47 ppm); this also results in an increase in δ¹³CO₂ due to the isotopic fractionation of plants that favor the lighter isotopologue ¹²CO₂ (from −7.11 ± 0.38‰ to −5.54 ± 0.22‰). CH₄ shows overall stable concentrations during the study period (2.02 ± 0.02 ppm); however, its isotopic fingerprint is less depleted in March (−47.36 ± 1.06‰) compared to February (−48.11 ± 0.47‰) and April (−48.55 ± 1.00‰), further corroborating the hypothesis by which Saharan dust intrusions in March may have triggered isotopic fractionation processes in CH₄.

The daily cycle of CO₂ (Figure 5A) follows a characteristic pattern: the lowest concentrations are recorded during the central hours of the day, coinciding with peak photosynthetic activity and a consequent increase in δ¹³C-CO₂ (Figure 6A), while they decrease at night. The δ¹³C-CO₂ values highlight the daily variations in atmospheric CO₂: the increase in δ¹³C-CO₂ during daytime results from isotopic fractionation associated with biochemical photosynthesis processes [37]. This phenomenon is due to the preferential uptake of the lighter isotopologue, ¹²CO₂, by plants, leaving higher values of ¹³CO₂ in the atmosphere. Moreover, an increase in the average δ¹³C-CO₂ value is observed from February to April, coinciding with the rise in photosynthetic activity typical of the warm boreal period [39,40]. Additionally, this variation is accompanied by a reduction in CO₂ emissions from thermogenic activities, which are characterized by an enrichment in ¹²CO₂. The average concentration of CO₂ tends to decrease from February to April, likely due to the reduced use of domestic heating sources as the transition from winter to spring occurs. CH₄, on the other hand, exhibits less pronounced daily variations (Figure 5B), with a slight decrease in concentration during the day, though less significant compared to CO₂, especially when considering the relative abundance of both compounds in the atmosphere. However, its average mole fraction does not appear to follow typical seasonal variations: a decrease in concentration is observed from February to March, as expected, followed by an increase from March to April, contrary to the usual pattern of CH₄. This phenomenon, currently under investigation in an ongoing study [98], could be another piece of evidence of Saharan dust influence on the chemistry of CH₄ in the central Mediterranean [108], in particular on its stable isotope chemistry [106,107]. With the average δ¹³C-CH₄ values measured in March being consistently different from their February and April counterparts (Figure 6B), the effects of rapid anomalies such as Saharan dust events deviating from seasonal trends further supports the hypothesis of a Cl effect on δ¹³C-CH₄.

Based on the methodologies introduced by other studies from the national consortium [90,109], an evaluation was aimed at the weekly pattern of CO₂ and CH₄ (Figure 7), and their respective isotopic variations (Figure 8). These evaluations allow us to determine the extent of anthropogenic emissions on continuous measurements, as anthropic activities are affected by weekly patterns that are not present in nature and therefore cannot be reported in daily, seasonal, and annual cycles. Possible influences linked to a weekly cycle include, but are not limited to, reduced photosynthesis caused by increased atmospheric pollution, which would result in a lower δ¹³C-CO₂ value [110,111]. The analysis showed that no clear weekly patterns are present as the differences are primarily of a seasonal nature and reflect the variability of CO₂ and CH₄ in a Mediterranean climate. One exception is CO₂ during the month of April, which yields higher δ¹³C-CO₂ and lower mole fractions during weekends, possibly reflecting reduced anthropogenic emissions and a higher influence of photosynthesis as a sink.

Finally, the Keeling plot [72] method was used to assess the extent and influence of specific sources over POT’s preliminary results. The Keeling plot assessment of CH₄ did not yield statistically significant results; however, that of CO₂ (Figure 9) provided the first insight from the national consortium on the influence of anthropogenic sources of CO₂. In fact, the intercepts of regression lines indicate emission sources with a δ¹³C of ≈−25/−30‰, consistent with fossil fuel burning. A lower R² is reported for March as the values are distributed over a wider range and likely reflect the influence of Saharan dust events on photosynthesis. Overall, the results of the measurements allowed us to highlight both general trends, such as environmental responses to seasonal changes in the central Mediterranean, and specific circumstances such as Saharan dust events.

6. Conclusions

The results of δ¹³CO₂ and δ¹³CH₄ measurements performed over 60 continuous days (20 February–20 April 2025) at the new ICOS POT station in Potenza, Italy, using a Picarro G2201-i CRDS analyzer are presented. POT is part of the CIAO (CNR IMAA Atmospheric Observatory) and integrates other measurements performed in the area. The data gathered at POT during the measurement period were evaluated to assess general tendencies during a period characterized by a winter-to-spring transition and provided the first insight into stable carbon isotope variability in the atmosphere from the developing cross-country consortium, which includes three more stations. These preliminary results constitute the first step towards a more coordinated measurement coordination effort within the consortium itself.

The preliminary results indicate that trends driven by changes in anthropogenic emissions and photosynthetic activity in the transition from low to high temperatures in a Mediterranean climate can be affected by rapid events, i.e., Saharan dust intrusions and their consequent effects on δ¹³C-CO₂ and δ¹³C-CH₄. δ¹³C-CH₄ in particular showed abrupt changes that may be attributable to chemical reactions between sodium chloride and mineral dust and changes in atmospheric oxidation; these phenomena, which are currently under investigation, represent the first direct evidence of similar carbon fractionation processes in the area. Although a decrease in CH₄ concentrations is observed from the beginning to the end of the measurements, the trend observed does not align with its usual seasonality. The data on CO₂ mole fractions and δ¹³C-CO₂ reveal a degree of variability consistent with the typical tendencies observed in the northern hemisphere.

Analyses of the daily and weekly cycles also provided insights on the balance between natural and anthropic influences. The daily cycles of CO₂ and δ¹³C-CO₂ show tendencies compatible with the effects of photosynthesis, which are expected to peak during the boreal summer season: specifically, lower CO₂ concentrations are directly linked to higher δ¹³C-CO₂ values during diurnal hours, a pattern consistent with photosynthesis. Weekly cycles do not show specific patterns with the exception of CO₂ during April, thus indicating that changes in local anthropic activities across the course of a standard week do not have a tangible impact on local measurements at POT. Finally, a Keeling plot of CO₂ allowed us to determine, during the transition from cold (February) to warm (April) months, that higher CO₂ concentrations are almost entirely attributable to fossil fuel burning and related phenomena with a similar isotopic fingerprint. These results, although preliminary in nature, provide an initial foundation for the isotopic characterization of the site and the observational footprint of the station. A more comprehensive characterization of the Mediterranean’s isotopic footprint will be achieved by analyzing the relationship between isotopic compositions and meteorological parameters, such as wind speed and direction, to determine the origin and nature of the intercepted air masses and integrating the results obtained at this station with those from other stations within the national consortium, contributing to a deeper understanding of regional biogeochemical processes.