OA Earth-prints Repository
Not a member yet
16457 research outputs found
Sort by
A Deep Convolutional Neural Network for Detecting Volcanic Thermal Anomalies from Satellite Images
No full textThe latest generation of high-spatial-resolution satellites produces measurements of high-temperature volcanic features at global scale, which are valuable to monitor volcanic activity. Recent advances in technology and increased computational resources have resulted in an extraordinary amount of monitoring data, which can no longer be so readily examined. Here, we present an automatic detection algorithm based on a deep convolutional neural network (CNN) that uses infrared satellite data to automatically determine the presence of volcanic thermal activity. We exploit the potentiality of the transfer learning technique to retrain a pre-trained SqueezeNet CNN to a new domain. We fine-tune the weights of the network over a new dataset opportunely created with images related to thermal anomalies of different active volcanoes around the world. Furthermore, an ensemble approach is employed to enhance accuracy and robustness when compared to using individual models. We chose a balanced training dataset with two classes, one containing volcanic thermal anomalies (erupting volcanoes) and the other containing no thermal anomalies (non-erupting volcanoes), to differentiate between volcanic scenes with eruptive and non-eruptive activity. We used satellite images acquired in the infrared bands by ESA Sentinel-2 Multispectral Instrument (MSI) and NASA & USGS Landsat 8 Operational Land Imager and Thermal InfraRed Sensor (OLI/TIRS). This deep learning approach makes the model capable of identifying the appearance of a volcanic thermal anomaly in the images belonging to the volcanic domain with an overall accuracy of 98.3%, recognizing the scene with active flows and erupting vents (i.e., eruptive activity) and the volcanoes at rest. This model is generalizable, and has the capability to analyze every image captured by these satellites over volcanoes around the world.Published3718OSV1: Verso la previsione dei fenomeni vulcanici pericolosiJCR Journa
Sources and migration pathways of methane and light hydrocarbons in the subsurface of the Southern Po River Basin (Northern Italy)
No full textThis paper presents new chemical and isotopic data on gases from deep oil and gas fields, bubbling gases, dissolved gases in groundwaters and dry seeps of the Southern Po River Basin (Emilia-Romagna, Italy), aiming to (i) characterize and differentiate the various types of deep natural gases; (ii) identify the source(s) of methane and light hydrocarbons in shallow aquifers and surface gas-rich emissions; (iii) propose a conceptual model of natural fluid migration pathways in the sedimentary prism of the Southern Po River Basin. Based on the isotopic composition of CH4 and C2–C4 n-alkanes, CH4/(C2H6+C3H8) ratio, relative proportion of the C7 hydrocarbons and relative concentration of cyclic compounds with respect to the total cyclic abundance, three main deep reservoirs of hydrocarbons are identified in the subsurface of the Southern Po River Basin: (1) microbial gas hosted in Pliocene-Pleistocene marine sediments, (2) thermogenic gas hosted in Miocene deposits and (3) thermogenic gas produced in Triassic carbonates. Helium isotopes of these deep fluids indicate an almost pure crustal origin (Rc/Ra values = 0.014–0.04), with negligible contributions from mantle-derived helium. A variable contribution of atmosphere-derived fluids is highlighted by low 4He/20Ne (down to 5.42) and 40Ar/36Ar (≤319.5) values. Comparison of chemical and isotopic signatures of deep and surficial hydrocarbon occurrences suggests that methane in shallow groundwaters or gas seeps is sourced by microbial gas migrating upward from deep Plio-Pleistocene reservoirs, with no detectable contributions of Triassic or Miocene thermogenic hydrocarbons. At shallow depths (roughly around 20–50 m.b.g.l.), Plio-Pleistocene microbial methane appears to be mainly stored in anoxic aquifers. However, where CH4 further migrates upwards and reaches aerobic environments (e.g., aquifers or soils), it readily undergoes a process of exothermic microbial oxidation mediated by methanotrophic bacteria. Where the structural architecture of the sedimentary sequence favors the migration of fluids, the methanotrophic biofilter is bypassed and CH4 is discharged through soil diffuse degassing or gas bubbling at water wells. We argue that microbial consumption might be able to bio-sequester significant amounts of Plio-Pleistocene deep-sourced methane in the form of CO2 and biomass. Such process might be widespread in the subsurface of the Southern Po River Basin and, possibly, in other foreland basins worldwide.Published1059819T. Geochimica dei fluidi applicata allo studio e al monitoraggio di aree sismicheJCR Journa
Hydrological deformation of karst aquifers detected by GPS measurements, Matese massif, Italy
No full textWe used hydrological and geodetic observations to characterize the effects of hydrological forcing on the displacements observed by GPS (Global Positioning System) measurements collected by stations located in the karst area of the Matese massif (Apennines, central-southern Italy). The latter is one of the main karst massifs of the central-southern Apennines, characterized by steep slopes, high mountain peaks (up to 2050 m a.s.l.), and wide endorheic areas, playing a fundamental role in recharging the large groundwater resource of the massif. Integrated statistical analysis of rainfall and spring discharge time series provided insight into timescales characterizing the hydrological dynamic of the Matese massif, highlighting how hydraulic conditions of the water table at a specific time mainly depend on long antecedent periods of rainfall. Nevertheless, the intense daily rainfall occurring during the wet season is responsible for the abrupt increase of the discharge of the Matese’s springs, which show a typical karst behavior. Statistical analyses of time series show robust correlations between hydrological conditions of the karst aquifer and GPS displacements observed at stations placed over and around the massif. We find that the observed outward and inward deformations of the massif (horizontal dilatation and contraction) are controlled by the water table variations, which are in turn controlled by the temporal variations of the groundwater recharge due to rainfall. The detected deformation patterns are intimately related to the seasonal and multi-year characteristics of the recharge/discharge processes and allowed us to track with geodetic measurements the different phases of the water cycle in the karst aquifer of the Matese massif.Published240OST1 Alla ricerca dei Motori GeodinamiciJCR Journa
An application of the NonErgodic ground ShaKing (NESK) approach to an historical earthquake scenario: The case-study of the 1915 Fucino earthquake (central Italy)
No full textThe 1915, Mw6.7, Fucino earthquake is one of the most destructive events occurred in the central-southern Apennines (Central Italy) in pre-instrumental era, involving normal faulting in a deep alluvial basin. This study shows the application of the empirical non-ergodic approach (NESK method) for mapping ground shaking related to this historical event, taking into account the regional features of source, propagation and site contributions. Corrections of the source-region and spatially correlated maps of site and path residuals are combined with median prediction at the reference rock (i.e. without site amplification) to generate spatially variable ground shaking and associated variability in terms of peak ground acceleration and spectral ordinates at vibration periods from 0.01s to 2s. The method captures the main spatial non-stationarities and anisotropies of the shaking fields produced by this earthquake in and around the Fucino basin. In particular, we obtain patterns of seismic motion quite in accordance with the results of other methods and the macroseismic intensity field. Marked amplifications of the shaking in the long-periods are also captured, due to the coupling of 3D site effects, especially in the deeper portion of the basin, with propagation effects mainly focused towards the eastern part of the fault. These results confirm that the non-ergodic shaking scenarios from NESK can provide useful indications even in the case of very complex seismological and geological contexts, such as in the case of strong events in deep sedimentary basins.Published1076225T. Sismologia, geofisica e geologia per l'ingegneria sismicaJCR Journa
Along-strike variations of strain partitioning within the Apennines determined from large-scale multi-temporal InSAR analysis
No full textIn the Apennine Mountains of the Italian Peninsula, GPS data display 3–4 mm/yr of divergent motion oriented N50°E between the Adriatic and Tyrrhenian coastlines. However, the mechanisms driving this extension remain debated and along-strike variations of extension within the actively deforming belt remain poorly constrained. Here, we derived the first large-scale extensional and vertical velocity field for the Apennines by multi-temporal InSAR analysis of 7 years of Sentinel-1 data at the scale of the entire range, improving the spatial resolution and vertical accuracy of existing GPS measurements. The results reveal along-strike variations of extensional rates and gradients, with extension concentrated on single fault systems in the north, consistent with the loci of seismicity and recent moderate earthquakes, and distributed throughout the central Apennines, where the range is widening. Vertical surface displacements do not resolve any active long-wavelength uplift of the orogenic belt and, on average, show more subsidence than uplift relative to the Tyrrhenian and Adriatic coasts. This work provides the first InSAR-based geodetic map of differential extension and uplift within the Italian Peninsula. Our results are compatible with a pure shear extensional model of the crust, driven by both boundary and gravitational forces.Published230076OST1 Alla ricerca dei Motori GeodinamiciJCR Journa
Characterization of trace elements in thermal and mineral waters of Greece
No full textNatural thermal and mineral waters are widely distributed along the Hellenic region and are related to the geodynamic regime of the country. The diverse lithological and tectonic settings they are found in reflect the great variability in their chemical and isotopic composition. The current study presents 276 (published and unpublished) trace element water data and discusses the sources and processes affecting the water by taking into consideration the framework of their geographic distribution. The dataset is divided in groups using temperature- and pH-related criteria. Results yield a wide range of concentrations, often related to the solubility properties of the individual elements and the factors impacting them (i.e. temperature, acidity, redox conditions and salinity). Many elements (e.g. alkalis, Ti, Sr, As and Tl) present a good correlation with temperature, which is in cases impacted by water rock interactions, while others (e.g. Be, Al, Cu, Se, Cd) exhibit either no relation or an inverse correlation with T possibly because they become oversaturated at higher temperatures in solid phases. A moderately constant inverse correlation is noticed for the vast majority of trace elements and pH, whereas no relationship between trace element concentrations and Eh was found. Seawater contamination and water-rock interaction seem to be the main natural processes that influence both salinity and elemental content. All in all, Greek thermomineral waters exceed occasionally the accepted limits representing in such cases serious harm to the environment and probably indirectly (through the water cycle) to human health.Published78376–783936A. Geochimica per l'ambiente e geologia medicaJCR Journa
La storia sismica di Siena: stato delle conoscenze (aggiornato a gennaio 2022)
No full textTo this day it remains impossible to determine exactly ‘when’ the next large earthquakes will take place. However, it is possible to say ‘where’ they are most likely to happen and ‘how’ they might interact with a given stretch of country and the buildings, infrastructures and people therein. In order to achieve such a probabilistic forecasting several kinds of studies are needed. Geological, geophysical and seismotectonic studies help to identify the structures (if any) capable of generating earthquakes and to understand how they function. The historical study of past earthquakes (historical seismology) is also very important, particularly in countries such as Italy, where the most destructive earthquakes can have very long return periods (centuries or even millennia) and one needs to go as far back in time as possible in order to draw a reliable picture of the trends of seismicity. Historical seismologists search for historical evidence of the effects of past earthquakes on the studied areas, translate these raw data into macroseismic intensity grades and derive from them the parameters identifying each earthquake. The Italian seismic catalogue is one of the richest and longest-reaching in the world. The current Parametric Catalogue of Italian Earthquakes CPTI15 version 4.0 (released in January 2022) spans the period from 1000 AD to 2020, in which period it lists 4860 earthquakes with epicentral location within the current boundaries of Italy or just outside them, with a set threshold of intensity (I ≤ V MCS) or instrumental magnitude (I ≤ Mw 4.0 or higher). The Italian Macroseismic Database DBMI15, version 4.0 (January 2022), contains 123981 macroseismic intensity data based on observations of effects related to 15343 Italian locations and 4894 earthquakes occurred in Italy and its immediate surroundings between 1000 AD and 2020. From these data the “seismic history” of each locality can be derived, i.e. a list of all the macroseismic effects observed in that locality within the time span covered by the reference catalogue. The maximum intensity attested at a place can be considered (with due caution) as an empirical evaluation of the highest intensity threshold that – as far as current knowledge goes – it should be reasonably expected in the future. The level of relative completeness of seismic histories is very variable, for many reasons. For instance, understandably enough, historically relevant towns are most likely to have a longer and more detailed seismic history than small, marginal villages. Siena, whose seismic history is discussed here, is the second Tuscan locality for the quantity of available macroseismic observations (147), after Florence (234) and before Lucca (108). The current seismic history of Siena is therefore comparatively rich and well documented. Not all of the 147 intensity data constituting the seismic history of Siena reflect the effects of ‘local’ earthquakes, i.e., those with epicentral location in the Sienese or Tuscan Provinces but all the earthquakes responsible for the major seismic effects observed in Siena are of local origin. On the basis of currently available knowledge, the territory of the Province of Siena is historically characterised (or rather in the last millennium or so) by fairly frequent seismicity, but much less relevant than that localised along the central-northern Apennine belt (from Garfagnana-Lunigiana to Mugello and the Upper Valtiberina), where Mw 6.0 threshold has been reached several times in this period, and sometimes exceeded. In Siena there are 25 seismic resonances evaluated as having an intensity ≤VI MCS. This should correspond to a scenario of effects in which approximately half of the buildings undergoes slight damage (slight cracks in plasterwork, possible fall of a few tiles or stones from chimneys); 25% of the buildings is affected by moderate damage (slight cracks in walls, considerable fall of plasterwork, tiles and many chimneys); no more than 5% of buildings is so damaged as to become temporarily uninhabitable. This intensity was assigned to Siena in relation to several earthquakes whose epicentral location occurs within a radius of a few kilometres from the town. No matter how rich the historical macroseismic data base available for Siena, there are still good margins for improvement. This is demonstrated by ongoing research initiatives, including the systematic perusal of serial historical sources (diaries, newspapers and historical seismic questionnaires). It can be hoped that forthcoming updates of the seismic history of Siena may lead to a conspicuous increase in the quantity of available data, though probably not to significant changes in the information framework briefly outlined in this paper.Published27-36OST1 Alla ricerca dei Motori Geodinamic
High frequency attenuation of S waves in alluvial deposits of the central Po Plain (northern Italy)
No full textThis article has been accepted for publication in Geophysical Journal International ©:The Author(s) 2023. Published by Oxford University Press on behalf of the Royal Astronomical Society. All rights reserved.Uploaded in accordance with the publisher's self-archiving policy.
All rights reserved.Estimation of local seismic response plays a key role in assessing local seismic hazard and particularly in the design of shaking scenarios. Modelling local seismic response involves knowing of the shear wave velocity (Vs) and quality factor (Qs) profiles for the site in question. The many techniques that have been developed to assess Vs in surface deposits produce reliable measurements of Vs , but these rarely correspond to direct measurements of Qs . The latter is often considered through damping measures from laboratory tests on small-scale soil samples, which can provide information primarily on intrinsic attenuation, neglecting the contribution of scattering effects. In this paper, using seismic recordings obtained at the surface and in boreholes at 100 m depth, we estimate an average value of Qs of some characteristic alluvial deposits of the Po Plain (northern Italy). Data come from a microseismic network which sampled an almost uniform lithology in the central Po Plain and consisted of three surface and four borehole stations with an interstation distance of about 2 km. The average value of Qs of the shallowest 100 m of the sedimentary strata, Qs100, is estimated by considering: (1) the high-frequency attenuation of seismic waves due to propagation through the corresponding stratigraphy and (2) the interference between incident and surface-reflected waves observed at borehole stations. We parametrize the first through k0_100, the difference between the values of the spectral decay parameter kappa (k) estimated at the surface and at the boreholes depth, respectively. We use the second in order to compute Vs100, the time-averaged Vs referred to the uppermost 100 m stratigraphy. We obtain: k0_100 = (11 ± 3) ms, Vs100 = (309 ± 11) m s −1 and Qs100 = 31 ± 10. At the surface, the estimated values of the site-specific kappa, k0, are found to range from 75 to 79 ms. As expected, these results are in good agreement with studies performed in other sites characterized by sandy or clayey lithologies, and can be usefully used in site response analysis at sites where the rigidity is mainly controlled by lithostatic pressure.Comune di Minerbio (grant: “Sperimentazione ILG Minerbio”; grant number: 0913.010).Published2075–2094OST2 Deformazione e Hazard sismico e da maremotoJCR Journa
Rapid provision of maps and volcanological parameters: quantification of the 2021 Etna volcano lava flows through the integration of multiple remote sensing techniques
No full textAt active volcanoes recurring eruptive events, erosive processes and collapses modify the edifice morphology and impact monitoring and hazard mitigation. At Etna volcano (Italy) between February and October 2021, 57 paroxysmal events occurred from the South-East Crater (SEC), which is currently its most active summit crater. Strombolian activity and high lava fountains (up to 4 km) fed lava flows towards the east, south and south-west, and caused fallout of ballistics (greater than 1 m in diameter) within 1–2 km from the SEC. The impacted area does not include permanent infrastructure, but it is visited by thousands of tourists. Hence, we rapidly mapped each lava flow before deposits became covered by the next event, for hazard mitigation. The high frequency of the SEC paroxysms necessitated integration of data from three remote sensing platforms with different spatial resolutions. Satellite (Sentinel-2 MultiSpectral Instrument, PlanetScope, Skysat and Landsat-8 Operational Land Imager) and drone images (visible and thermal) were processed and integrated to extract digital surface models and orthomosaics. Thermal images acquired by a permanent network of cameras of the Istituto Nazionale di Geofisica e Vulcanologia were orthorectified using the latest available digital surface model. This multi-sensor analysis allowed compilation of a geodatabase reporting the main geometrical parameters for each lava flow. A posteriori analysis allowed quantification of bulk volumes for the lava flows and the SEC changes and of the dense rock equivalent volume of erupted magma. The analysis of drone-derived digital surface models enabled assessment of the ballistics’ distribution. The developed methodology enabled rapidly and accurate characterisation of frequently occurring effusive events for near real-time risk assessment and hazard communication.Published58OSV3: Sviluppo di nuovi sistemi osservazionali e di analisi ad alta sensibilitàJCR Journa
Joint Inversion of Geodetic and Strong Motion Data for the 2012, Mw 6.1–6.0, May 20th and May 29th, Northern Italy Earthquakes: Source Models and Seismotectonic Interpretation
No full textAbstract We present the first rupture models of the two mainshocks of the 2012 northern Italy
sequence, determined by jointly inverting seismic and geodetic data. We aim at providing new insights into
the mainshocks for which contrasting seismotectonic interpretations are proposed in literature. Sources'
geometric parameters were constrained by seismic reflection profiles, 3-D relocations and focal mechanisms
of mainshocks/aftershocks. Site-specific velocity profiles were used to model accelerograms affected by strong
propagation effects related to the Po basin. Our source models differ significantly from previous ones relying
on either seismic or geodetic data. Their comparison against geological sections and aftershock distribution
provides new insights about the ruptured thrust faults. The May 20th Mw6.1 mainshock activated the Middle
Ferrara thrust-ramp dipping ∼45° SSW-wards, breaking a main eastern slip patch 4–15 km deep in Mesozoic
carbonates (maximum slip 0.7–0.8 m) and Paleozoic-Triassic basement rocks, and a small western patch in
the basement. The May 29th Mw6.0 mainshock featured two separated asperities along the Mirandola thrustramp dipping ∼42° S-wards: an eastern asperity 4–15 km deep in Mesozoic carbonates and basement rocks
(maximum slip 0.7 m) and a deeper western one (7–16 km depth) mainly in the basement (slip peak 0.8 m).
On-fault aftershocks were concentrated within the basement and Mesozoic carbonates, devoiding highslip zones. Slip and aftershock distribution was controlled by the rheological transition between Mesozoic
carbonates and Cenozoic sediments. Unlike previous thin-skinned tectonic interpretations, our results point to
a complex rupture process along moderately dipping (40°–45°) thrust-ramps deeply rooted into the Paleozoic
crystalline basement.
Plain Language Summary The two M6 mainshocks of the 2012 Italy sequence are the strongest
earthquakes ever observed in the Po Plain, a strategic region for the Italian economy. The mainshocks
ruptured blind thrust-faults, however their source models and seismotectonic interpretation are still debated
because the thrust-system architecture is controversial. Contrasting thick-skinned and thin-skinned tectonic
models are proposed. In thick-skinned interpretations, shortening is accommodated by thrust-ramps
rooted into the crystalline basement that represent main seismogenic structures, whereas in thin-skinned
interpretations, shortening and seismicity are controlled by listric faults splaying out from dècollement levels
in the sedimentary crust. A comprehensive analysis of the mainshocks' source represents an opportunity to
provide new insights into the seismogenesis in northern Italy and on a broader scale into seismotectonics of
thrust-and-fold belts. We get a complete picture of the mainshocks kinematics by jointly inverting, for the
first time, seismic and geodetic data, and unravel rupture heterogeneities not resolved by previous studies.
By integrating source models with aftershock locations and geological models, we propose a comprehensive
seismotectonic interpretation of the sequence. We conclusively identify the ruptured faults that correspond to
thrust-ramps rooted into the crystalline basement and evidence the key role played by lithological changes in
the rupture process.Publishede2022JB026278OST3 Vicino alla fagliaJCR Journa