Skip to main content
The search returned 9 results
Now showing 1 - 9 of 9
  • Some of the metrics are blocked by your 
    Item type:Publication, Access status: Open Access ,

    Machine learning models to complete rainfall time series databases affected by missing or anomalous data

    In recent years, artificial intelligence in geosciences is spreading more and more, thanks to the availability of a large amount of data. In particular, the development of automatic raingauges networks allows to get rainfall data and makes these techniques effective, even if the performance of artificial intelligence models is a consequence of the coherency and quality of the input data. In this work, we intended to provide machine learning models capable of predicting rainfall data starting from the values of the nearest raingauges at one historic time point. Moreover, we investigated the influence of the anomalous input data on the prediction of rainfall data. We pursued these goals by applying machine learning models based on Linear Regression, LSTM and CNN architectures to several raingauges in Tuscany (central Italy). More than 75% of the cases show an R2 higher than 0.65 and a MAE lower than 4 mm. As expected, we emphasized a strong influence of the input data on the prediction capacity of the models. We quantified the model inaccuracy using the Pearson's correlation. Measurement anomalies in time series cause major errors in deep learning models. These anomalous data may be due to several factors such as temporary malfunctions of raingauges or weather conditions. We showed that, in both cases, the data-driven model features could highlight these situations, allowing a better management of the raingauges network and rainfall databases.
      33  35
  • Some of the metrics are blocked by your 
    Item type:Publication, Access status: Open Access ,

    Deep learning models to predict flood events in fast-flowing watersheds

    This study aims to explore the reliability of flood warning forecasts based on deep learning models, in particular Long-Short Term Memory (LSTM) architecture. We also wish to verify the applicability of flood event predictions for a river with flood events lasting only a few hours, with the aid of hydrometric control stations. This methodology allows for the creation of a system able to identify flood events with acceptable errors within several hours' notice. In terms of errors, the results obtained in this study can be compared to those obtained by using physics-based models for the same study area. These kinds of models use few types of data, unlike physical models that require the estimation of several parameters. However, the deep learning models are data-driven and for this reason they can influence the results obtained. Therefore, we tested the stability of the models by simulating the missing or wrong input data of the model, and this allowed us to achieve excellent results. Indeed, the models were stable even if several data were missing. This method makes it possible to lay the foundations for the future application of these techniques when there is an absence of geological-hydrogeological information preventing physical modeling of the run-off process or in cases of relatively small basins, where the complex system and the unsatisfactory modeling of the phenomenon do not allow a correct application of physical-based models. The forecast of flood events is fundamental for correct and adequate territory management, in particular when significant climatic changes occur. The study area is that of the Arno River (in Tuscany, Italy), which crosses some of the most important cities of central Italy, in terms of population, cultural heritage, and socio-economic activities.
      38  355
  • Some of the metrics are blocked by your 
    Item type:Publication, Access status: Open Access ,

    Seasonal rainfall trends of a key Mediterranean area in relation to large-scale atmospheric circulation: How does current global change affect the rainfall regime?

    Current global warming causes a change in atmospheric dynamics, with consequent variations in the rainfall regimes. Understanding the relationship between global climate patterns, global warming, and rainfall regimes is crucial for the creation of future scenarios and for the relative modification of water management. The aim of this study is to improve knowledge of the relationship between North Atlantic Oscillation (NAO), East Atlantic (EA), and Western Mediterranean Oscillation (WeMO) with the seasonal rainfalls in Tuscany, Italy. The study area occupies a strategic position since it lies in a transition zone between the wet area of northern Europe and the dry area of the northern coast of Africa. This research, based on a statistical correlation method and on linear models, is designed to understand the relationship between seasonal rainfalls and climate patterns. The results of this study demonstrate that the use of linear models can yield more information than traditional statistical corre- lations. The results show a decrease in rainfall in the warm period of the year, namely in the summer, when its expression is most visible. This phenomenon is ascribable to current global warming, which causes an increase in sea-surface temperatures. An increase in the Northern Atlantic Sea Surface Temperature and in the Mediterra- nean Sea Surface Temperature causes a reduction of the Iceland Low, with an extension of the Azores High. Moreover, an increase in the Genoa Gulf SST
      27  105
  • Some of the metrics are blocked by your 
    Item type:Publication, Access status: Restricted ,

    A finite element Galerkin/least-squares method for computation of multicomponent compressible–incompressible flows

    (2012-08-30) ; ; ; ;
    Longo, A.; Istituto Nazionale di Geofisica e Vulcanologia, Sezione Pisa, Pisa, Italia
    The space–time Galerkin/least-squares finite element method with discontinuity capturing (ST-GLSDC), developed by Hughes and collaborators [Shakib et al. A new finite element formulation for computational fluid dynamics: X. The compressible Euler and Navier–Stokes equations. Comput Methods Appl Mech Eng 1991;89:141–219], allows to study both compressible and incompressible single-fluid one-component flows. It is effective in the stabilization of the numerical solution without introducing excessive overdiffusion. In this work the development by Hauke and Hughes [A comparative study of different sets of variables for solving compressible and incompressible flows. Comput Methods Appl Mech Eng 1998;153:1–44] to pressure primitive variables is extended to single-fluid multicomponent compressible and incompressible flows of gas–liquid mixtures at local mechanical and chemical equilibrium. The stabilized algorithm is implemented in a parallel C++ library, which is tested on several benchmarks. The solution of the system of equations for the conservation of mass of each component, and of momentum and energy of the global mixture, requires the introduction of mass fractions as primitive variables to describe mixture composition. The weak formulation, the stabilization parameters, and the time-marching algorithm are rewritten in terms of the expanded set of variables, keeping similarity with the formulation in pressure variables.
      388  45
  • Some of the metrics are blocked by your 
    Item type:Publication, Access status: Open Access ,

    The stochastic quantization method and its application to the numerical simulation of volcanic conduit dynamics under random conditions

    (2010) ; ; ; ;
    Peruzzo, E.; Scuola Normale Superiore, Pisa, Italy
    Stochastic Quantization (SQ) is a method for the approximation of a continuous probability distribution with a discrete one. The proposal made in this paper is to apply this technique to reduce the number of numerical simulations for systems with uncertain inputs, when estimates of the output distribution are needed. This question is relevant in volcanology, where realistic simulations are very expensive and uncertainty is always present. We show the results of a benchmark test based on a one-dimensional steady model of magma flow in a volcanic conduit.
      197  264
  • Some of the metrics are blocked by your 
    Item type:Publication, Access status: Restricted ,

    Heterogeneous large total CO2 abundance in the shallow magmatic system of Kilauea volcano, Hawaii

    Due to its very low solubility in silicate melts, CO2 concentrations in melt inclusions (MIs) within crystals are commonly orders of magnitude less than the total concentration in the multiphase magma, strongly limiting the possibility to constrain CO2 abundance based on the dissolved quantities. Here we develop a statistical method to process MI data, which allows analytical uncertainties to be taken into account together with the peculiar features of the local saturation surface. The method developed leads to retrieve total H2O and CO2 concentrations in magma as well as the gas phase abundance at the time of magma crystallization. Application to a set of 29 high-resolution secondary ion mass spectrometry (SIMS) MI data from a single specimen of the 1842–1844 eruption of Kilauea, Hawaii, reveals the existence of heterogeneous total CO2 abundance, and of at least 2–6 wt % total CO2 in some magma batches, two orders of magnitude higher than the dissolved amounts and 30–50 times more abundant than the corresponding total H2O content. Heterogeneous total volatile concentrations are interpreted as due to a combination of degassing and gas flushing in magma subject to convective motion at shallow depth where P < 100 MPa. In such a view, the magma rising to shallow depth in the volcanic system carries initially a total volatile content ≤1 wt %, corresponding to the determined low total CO2 population, and consistent with previous global estimates. The high CO2 populations correspond to progressive CO2 enrichment due to degassing at low P and flushing from a deep CO2-rich gas. A total CO2 content >1 wt % is likely to characterize the >30 km deep magma, not represented in the analyzed inclusions, from which a CO2-rich gas phase exsolves and decouples from the liquid.
      222  18
  • Some of the metrics are blocked by your 
    Item type:Publication, Access status: Open Access ,

    A numerical code for the simulation of magma-rocks dynamics

    We present a numerical code for the simulation of the dynamics of compressible to incompressible, multicomponent ows, based on the _nite element algorithm by Hauke & Hughes (1998). Balance equations for mass, momentum, energy and composition are solved with space-time Galerkin least-squares and discontinuity-capturing stabilizing techniques. The code is used to study the dynamics of convection and mixing in magmatic systems such as replenishment of magma chambers and volcanic conduits, and it reveals the occurrence of previously not described processes. The uid-structure interaction of fully coupled magma-rock dynamics is being implemented by using the deforming-spatial domain method by Tezduyar (2006), that intrinsecally includes moving meshes.
      486  207
  • Some of the metrics are blocked by your 
    Item type:Publication, Access status: Restricted ,

    Numerical simulation of the dynamics of fluid oscillations in a gravitationally unstable, compositionally stratified fissure

    We have simulated the dynamics of convection, mixing and ascent of two basaltic magmas differing in their volatile and crystal content, giving rise to a gravitationally unstable configuration along a dyke or fissure. Numerical simulations are performed by a recently developed code which describes the transient 2D dynamics of multicomponent fluids from the incompressible to the compressible regime, and the initial and boundary conditions are inspired to the paroxysmal eruption which occurred at Stromboli in 2003 (D'Auria et al. 2006). Multicomponent (H2O+CO2) saturation is accounted for by modelling the non-ideal equilibrium between the gas phase and the melt. The numerical results show the formation of a rising bulge of light magma, and the sink of discrete batches of dense magma towards deep fissure regions. Such dynamics are associated with a complex evolution of the pressure field, which shows variations occurring over a wide spectrum of frequencies. A first order analysis of the propagation of such pressure disturbances through the country rocks shows that the pre-eruptive fissure dynamics are able to produce mm-size, mainly radial deformation of the volcano, and a detectable seismic signal with spectral peaks at periods of about 50 s.
      211  24
  • Some of the metrics are blocked by your 
    Item type:Publication, Access status: Restricted ,

    Numerical simulation of convection and mixing in magma chambers replenished with CO2-rich magma

    (2006) ; ; ; ;
    Longo, A.; Istituto Nazionale di Geofisica e Vulcanologia, Sezione Pisa, Pisa, Italia
    Magma convection and mixing, and periodic refilling, commonly occur in magma chambers.We show here that the presence of CO2 in the refilling magma is a very efficient mean of inducing buoyant-driven plume rise and large scale convection. Numerical simulations performed with an appositely developed code for the transient 2D dynamics of multicomponent compressible to incompressible fluids reveal several features of the processes of plume rise, convection and mixing in magma chambers associated with chamber refilling. A parametric study on CO2 abundance in the refilling magma shows that progressively larger amounts of this volatile produce a shift from simple plume rise and spreading near the chamber top, to complex patterns of flow circulation and large scale vorticity and mixing. Lower chamber depth and lower magma viscosity largely enhance the efficiency of mixing and convection, favoring the formation of multiple vortexes migrating with time.
      241  28