Cuoco, Emilio

The δ18O and δ2H of rivers and springs were investigated in order to characterize the groundwater recharge sources around Nyiragongo and Nyamulagira volcanoes, in the Democratic Republic of the Congo and Rwanda. Water samples were collected monthly between November 2013 and October 2014 from 5 major rivers, 3 major cold springs, 3 tepid springs and 1 hot spring. The temperatures of each spring were nearly constant over the sampling period attesting for their groundwater character, while the temperatures of the rivers were much more variable. The rivers monthly δ2H and δ18O range from 􀀀 6.8‰ to 1.9‰ and 􀀀 3.1‰ to 1.6‰, respectively, while springs showed depleted values that span from 􀀀 10.2 to 􀀀 1.1‰ for δ2H and 􀀀 3.6 to 􀀀 1.9‰ for δ18O. Catchment morphology (formed of depression, upper footslope and medium to high gradient-mountains) and the local tectonic discontinuity (fissures and faults) regulate the surface runoff and subsurface flow, control the precipitation infiltration zones and hence the aquifers recharge areas. Chemical and isotopic (δ18O and δ2H) compositions of springs and rivers reveal the presence of shallow and deep aquifers, with some waters having intermediate isotope composition. Three different recharge zones characterized by different altitudes were identified: the first is found at low altitude ranging from ~1800 m to ~2150 m, the second and intermediate recharge zone in the altitude range from ~2180 m to ~2500 m at the upper footslope area, while the third and highest recharge area is located in the altitudes range from ~2620 to ~3220 m. The two upper recharge areas are the most fractured and fissured zones allowing rapid infiltration of depleted precipitations which recharge deep aquifers found in the tepid and hot springs. Based on their chemical and isotopic composition, waters from the shallow and deep aquifers have been considered representative of mixing end members. During their ascent to the surface, water from the deep aquifer mixes with that of shallow aquifer yielding the tepid springs of intermediate chemical and isotopic composition, while the other keep their original fingerprint corresponding to the isotopically depleted hot spring.

A theoretical pattern for Fe and As co-precipitation was tested directly in a groundwater natural system. Several monitoring wells were sampled to identify the different endmembers that govern the hydrodynamics of the Ferrarelle Groundwater System in the Riardo Plain (Southern Italy). In agreement with recent investigations, we found a mix of a deep and a shallow component in different proportions, resulting in a specific chemical composition of groundwater in each well depending on the percentages of each component. The shallow component was characterized by EC ~ 430 µS/cm, Eh ~ 300 mV, Fe ~ 0.06 µmol/L and As ~ 0.01-0.12 µmol/L, while the deep component was characterized by EC ~ 3400 µS/cm, Eh ~ 170 mV, Fe ~ 140 µmol/L and As ~ 0.59 µmol/L. A general attenuation of As and Fe concentration that was not due to a simple dilution effect was observed in the mixing process. The oxidation of Fe(II) to Fe(III) produces solid precipitates which adsorb As from solution and then co-precipitate. The reactions pattern of Fe(II) oxidation and As adsorption gave a linear function between [As] and [Fe], where the angular coefficient depends on the [O2]/[H+] ratio. Chemical data obtained from our samples showed a very good agreement with this theoretical relationship. The investigated geochemical dynamics represented a natural process of attenuation of Fe and As, two undesirable elements that usually affect groundwater quality in volcanic aquifers in central-southern Italy, which are exploited to supply drinking water.

We present the main features of the permanent monitoring system managed by the Istituto Nazionale di Geofisica e Vulcanologia-Osservatorio Vesuviano in the Campi Flegrei caldera. Eruptive history of this active volcano shows that the majority of the eruptive events has been characterised by high explosivity and was accompanied by pyroclastic density currents. Its last eruption occurred in AD 1538 and in the next centuries the Campi Flegrei caldera has experienced several episodes of bradyseism and also the progressive increasing of the urbanisation in the area (west of Naples). Monitoring the dynamics of a mainly explosive volcano completely embedded in a very populated area is a challenging task. In order to detect any variation in the physical and chemical parameters of the Campi Flegrei caldera, the Istituto Nazionale di Geofisica e Vulcanologia-Osservatorio Vesuviano manages a permanent multi-parametric monitoring system. All the recorded h24 continuous data are transmitted to the Monitoring Room of the Osservatorio Vesuviano in Naples, where they are acquired, processed and evaluated to define changes in the dynamical state of the volcano. The caldera, since the end of 2004, is experiencing a bradyseismic episode characterised by a low velocity rate uplift, low energy earthquakes and increasing in the magmatic components of fumarolic fluids. The monitoring and surveillance activity of the Campi Flegrei caldera plays a crucial role in the volcanic emergency plan that includes evacuation of approximately 500,000 people before the beginning of the eruption.

An integrated approach using chemical and microbial indicators has been tested in two different sites of the Campania Plain (Southern Italy) with different land use covering and different hydrogeological features in order: (1) to define the water-rock interaction processes, (2) to differentiate sources of pollution in a detailed way (3) to evaluate the degree of water quality in the studied alluvial aquifer and (4) to identify the most worrying elements for human's health. Groundwater have showed a HCO3-Ca signature for both investigated sites, and a progressive enrichment in alkali ions has been highlighted moving from the boundary of the plain toward the coastal areas, due to groundwater interaction with volcanic rocks along the flow path. The application of the Factor Analysis allowed to identify different sources of pollution, which were attributed to (a) leaks in the sewer system for the Agro-Aversano Area and also the spreading of manure as fertilizers in agricultural activities for the Caiazzo Plain. Furthermore, it has been highlighted that the use of major elements, trace elements and microbiological indicators, allows to accurately differentiate contamination processes in progress. In fact, from the results of the Factor Analysis applied in the Agro-Aversano area, no significant statistically relationships between major elements and microbiological indicators of fecal contamination were highlighted, unlike the Caiazzo plain where statistically significant correlations have been found between major and trace elements and microbiological indicators. The use of a Groundwater Quality Index has shown general poor water quality for the majority of analyzed samples due to the high amount of Nitrate and Fecal indicators. The use of a Health Risk Assessment highlighted that Nitrate coupled with Fluoride represent the most important concern for human health compared to the all investigated parameters in both sites.

The present investigation provides measurements of radon (222Rn) concentration levels in 20 thermal waters at the Campi Flegrei volcanic caldera, an important geothermal system with hydrothermal manifestations in the Neapolitan area (Southern Italy). We used a RAD7® Radon-in-air detector equipped with Big Bottle RAD H2O and DRYSTIK accessories (Durridge Co. Inc.). Water samples with different chemical and/or physical conditions, not used as drink waters, are taken from continental thermal groundwaters, springs, lakes, pools and one sub merged thermal spring. The waters are mostly chlorine to bicarbonate, except of a few sulphate types sampled at the hydrothermal discharge areas of Solfatara and Pisciarelli, central in the caldera. Water temperature and pH values range from 18.1 to 91.3 ◦C and from 2 to 8, respectively. Sampling and measurement of radon in groundwater are complicated by the high volatility of the gas; a method is here proposed. In some of the 20 sites double or triple samples were collected by using different volume polyethylene terephthalate (PET) bottles, diluting sample with blank water, and modifying flow of pumped wells. We suggest that dilution can be considered when water is i) not enough to fill in the PET, resulting in large head space in the sampler, ii) too hot determining damage of the PET or iii) too saline to clog the Big Bottle System. Dissolved radon concentrations vary from 0.1 ± 0.1 to 1146 ± 57 Bq/L with an average value of 152 Bq/L, using the CAPTURE program, the default RAD7 data acquisition program. Similar values in radon concentration are obtained using the method proposed in De Simone et al. (2015) ranging between 0.1 ± 5.8 and 1286 ± 98 Bq/L with an average value of 167 Bq/L. The hottest and most acidic sulphate waters refer to a small boiling pool at Pisciarelli hydrothermal discharge area and have nearly zero 222Rn content. 222Rn concentrations from this study are mostly below the reference level of 1000 Bq/L recommended for human health protection by the European Commission and the most adopted in the scientific community (Catao ˜et al., 2022). No correlation has been observed between temperature, pH, major anions and radon content values, nor between rock composition since it is almost homogeneous trachyte at the study sites. 222Rn levels therefore appear to reflect the local sedimentological, structural or hydrogeological conditions. The levels of 222Rn here presented are an important background for the scientific community that will intend to define the natural fluctuations of dissolved 222Rn in relation with seasons, environment, hydrogeology or volcanic dynamics at the geohazardous Campi Flegrei area.

The Riardo basin hosts groundwater exploited for the production of high quality, naturally sparkling, bottled water (e.g., Ferrarelle water), and circulating in a system constituted by highly fractured Mesozoic carbonates, overlain by more impervious volcanic rocks of the Roccamonfina complex. The two formations are locally in hydraulic connection and dislocated by deep-rooted faults. The study aimed at elucidating groundwater origin and circulation, using isotopic tracers (δ18O, δ2H, δ11B and 87Sr/86Sr) coupled to groundwater dating (Tritium, CFCs and SF6). Besides recharge by local precipitation over the Riardo hydrogeological basin, stable isotope ratios in water indicated an extra-basin recharge, likely from the elevated surrounding carbonate reliefs (e.g., Maggiore and Matese Mts.). The mineralization process, promoted by the deep CO2 flux, controls the B and Sr contents. However, their isotopic ratios did not allow discriminating between circulation in the volcanic and in the carbonate aquifers, as in the latter the isotopic composition differed from the original marine signature. Groundwater model ages ranged from ~ 30 years for the volcanic endmember to > 70 years for the deep, mineralized end-member, with longer circuits recharged at higher elevations. Overall, the results of this study were particularly relevant for mineral water exploitation. A recharge from outside the hydrogeological basin could be evidenced, especially for the more mineralized and valuable groundwater, and an active recent recharge was detected for the whole Riardo system. Both findings will contribute to the refinement of the hydrogeological model and water budget, and to a sustainable development of the resource.

Single rainwater samples were collected in the city of Goma (~1,1 million inhabitants), eastern Democratic Republic of the Congo, from January to June 2013 to draw a baseline for rainwater chemical composition and quality as influenced by the permanent plume of Nyiragongo volcano. This was a better period for a baseline as the neighboring Nyamulagira volcano, only 15 km apart, had no important degassing from its central crater, and hence the recorded volcanic influence on the rainwater chemistry was solely from Nyiragongo's lava lake which has been active since May 2002. The baseline for the rainwater chemistry and quality is important for this highly populated region where rainwater is the unique potable water source for the inhabitants of many villages surrounding the volcanoes, and for some of the inhabitants of the city of Goma. Our results show that samples collected at the crater rim of Nyiragongo were more acidic with pH ranging from 3.70 to 3.82, while the majority of rainwater samples collected in downtown Goma city and to the northeastern zone of the volcano had pH close to 5.7; which represents the value for rainwater from unpolluted continental areas (Berner and Berner, 2012). However, the pH was as low as 3.93 to the west of Nyiragongo volcano because the volcanic plume is directed westward by the dominant local wind direction. The western part of the city of Goma as well as the small town of Sake and many villages (e.g. Rusayo, Mubambiro, Kingi, …) are located in this zone, and experience endemic fluorosis caused by high fluoride in the available water. The mean F- in this zone was 0.38 mg/L, while the southern and northeastern zones had mean F- concentrations on 0.44 and 0.01 mg/L respectively; even though concentrations higher than the WHO guidelines were found in few samples from the western zone (1.69 mg/L) and from the southern zone (3.44 mg/L). Compared to data from Cuoco et al. (2012) obtained during the Nyamulagira 2010 eruption, and from Balagizi et al.2017 and Liotta et al., 2017 obtained during the intense degassing of both Nyiragongo and Nyamulagira lava lakes; we have noted similarity in the spatial variation of the pH, but samples from the present study showed notable lower concentrations of major elements. This is the case for fluoride which is strictly of volcanic origin. For the other major elements, anthropogenic sources, mainly the traffic and wind-blown dust; or other non-volcanic natural sources influenced their concentrations. Thus, the anions (Cl- and SO42-) and cations (Na+, K+, Mg2+, and Ca2+) from the present study are either lower compared to that previously reported in the literature for the Virunga, or are both comparable for the zones impacted by anthropogenic activities.

The Italian Apennines are among the most important sources of freshwater for several Italian regions. With evidences of deep CO2-rich fluids intruding into aquifers in the nearby central-southern Apennines, a thorough investigation into the geochemistry of groundwater became critical to ensure the water quality in the area. Here, we show the main hydrogeochemical processes occurring in the Matese Massif (MM) aquifer through the investigation of 98 water samples collected from springs and water wells. All waters were classified as HCO3 type with Ca dominance (from 50% up to 97%) and variable amount of Mg (from 1% up to 49%). A multivariate statistical approach through the application of the factor analysis (FA) highlighted three main hydrogeochemical processes: (i) water-carbonate rock interactions mostly enhanced in peripheral areas of the MM by CO2 deep degassing; (ii) addition of NaCl-rich components linked to recharging process and to water mixing processes of the groundwater with a thermal component relatively rich in Cl, Na, and CO2; (iii) anthropogenic activities influencing groundwater composition at the foothills of MM. Furthermore, the first detailed TDIC, pCO2, and δ13C-TDIC distribution maps of the MM area have been created, which track chemical and isotopic anomalies in several peripheral areas (Pratella, Ailano, and Telese) throughout the region. These maps systematically highlight that the greater the amount of dissolved carbon occurs the heavier the C isotope enrichment, especially in the peripheral areas. Conversely, spring waters emerging at higher altitudes within MM are only slightly mineralized and associated with δ13C-TDIC values mainly characterized by recharging processes with the addition of biogenic carbon during the infiltration process through the soil.

We present a new geometrical method capable of quantifying and illustrating the outcomes of a three-component mixing dynamics. In a three-component mixing sce nario, classical algebraic equations and endmember mixing analysis (EMMA) can be used to quantify the contributions from each fraction. Three-component mixing of natural waters, either in an element–element plot or by using the EMMA mixing sub space is described by a triangular shaped distribution of sample points where each endmember is placed on an apex, while each side corresponds to the mixing function of the two endmembers placed at the apex, considering the third endmembers' con tribution equal to zero. Along each side, the theoretical mixing fractions can be com puted using mass balance equations. Samples with contributions from three endmembers will plot inside the triangle, while the homogeneous barycentric coordi nate projections can be projected onto the three sides. The geochemistry observed in the mineralized Ferrarelle aquifer system (southern Italy) results from three component mixing of groundwater, each with diagnostic geochemical compositions. The defined boundary conditions allow us to parameterize and validate the proce dures for modelling mixing, including selection of suitable geochemical tracer

Groundwater resources constitute the main source of clean fresh water for domestic use and it is essential for food production in the agricultural sector. Groundwater has a vital role for water supply in the Campanian Plain in Italy and hence a future sustainability of the resource is essential for the region. In the current paper novel data mining algorithms including Gaussian Process (GP) were used in a large groundwater quality database to predict nitrate (contaminant) and strontium (potential future increasing) concentrations in groundwater. The results were compared with M5P, random forest (RF) and random tree (RT) algorithms as a benchmark to test the robustness of the modeling process. The dataset includes 246 groundwater quality samples originating from different wells, municipals and agricultural. It was divided for the modeling process into two subgroups by using the 10-fold cross validation technique including 173 samples for model building (training dataset) and 73 samples for model validation (testing dataset). Different water quality variables including T, pH, EC, HCO3-, F-, Cl-, SO42-, Na+, K+, Mg2+, and Ca2+ have been used as an input to the models. At first stage, different input combinations have been constructed based on correlation coefficient and thus the optimal combination was chosen for the modeling phase. Different quantitative criteria alongside with visual comparison approach have been used for evaluating the modeling capability. Results revealed that to obtain reliable results also variables with low correlation should be considered as an input to the models together with those variables showing high correlation coefficients. According to the model evaluation criteria, GP algorithm outperforms all the other models in predicting both nitrate and strontium concentrations followed by RF, M5P and RT, respectively. Result also revealed that model's structure together with the accuracy and structure of the data can have a relevant impact on the model's results.