Characterization of microplastics ingested by marine benthos - a methodological and field-experimental study

Microplastics have become ubiquitous pollutants in the marine environment. Ingestion of microplastics by a wide range of marine organisms has been recorded both in laboratory and field studies. Despite growing concern for microplastics, few studies have evaluated their concentrations and distribution in wild populations. Further, there is a need to identify cost-effective standardized methodologies for microplastics extraction and analysis in organisms. In this thesis I present: (i) the results of a multi-scale field sampling to quantify and characterize microplastics occurrence and distribution in 4 benthic marine invertebrates from saltmarshes along the North Adriatic Italian coastal lagoons; (ii) a comparison of the effects and cost-effectiveness of two extraction protocols for microplastics isolation on microfibers and on wild collected organisms; (iii) the development of a novel field- based technique to quantify and characterize the microplastic uptake rates of wild and farmed populations of mussels (Mytilus galloprovincialis) through the analysis of their biodeposits. I found very low and patchy amounts of microplastics in the gastrointestinal tracts of sampled organisms. The omnivorous crab Carcinus aestuarii was the species with the highest amounts of microplastics, but there was a notable variation among individuals. There were no substantial differences between enzymatic and alkaline extraction methods. However, the alkaline extraction was quicker and cheaper. Biodeposit traps proved to be an effective method to estimate mussel ingestion rates. However their performance differed significantly among sites, suggesting that the method, as currently designed, is sensible to local environmental conditions. There were no differences in the ingestion rates of microplastics between farmed and wild mussels. The estimates of microplastic ingestion and the validated procedures for their extraction provide a strong basis for future work on microplastic pollution.

Design for uncertainty of aerospace structures

The aim of this work is to present a general overview of state-of-the-art related to design for uncertainty with a focus on aerospace structures. In particular, a simulation on a FCCZ lattice cell and on the profile shape of a nozzle will be performed. Optimization under uncertainty is characterized by the need to make decisions without complete knowledge of the problem data. When dealing with a complex problem, non-linearity, or optimization, two main issues are raised: the uncertainty of the feasibility of the solution and the uncertainty of the objective value of the function. In the first part, the Design Of Experiments (DOE) methodologies, Uncertainty Quantification (UQ), and then Uncertainty optimization will be deepened. The second part will show an application of the previous theories on through a commercial software. Nowadays multiobjective optimization on high non-linear problem can be a powerful tool to approach new concept solutions or to develop cutting-edge design. In this thesis an effective improvement have been reached on a rocket nozzle. Future work could include the introduction of multi scale modelling, multiphysics approach and every strategy useful to simulate as much possible real operative condition of the studied design.

Deep learning for computer vision: a comparison between convolutional neural networks and hierarchical temporal memories on object recognition tasks

In recent years, Deep Learning techniques have shown to perform well on a large variety of problems both in Computer Vision and Natural Language Processing, reaching and often surpassing the state of the art on many tasks. The rise of deep learning is also revolutionizing the entire field of Machine Learning and Pattern Recognition pushing forward the concepts of automatic feature extraction and unsupervised learning in general. However, despite the strong success both in science and business, deep learning has its own limitations. It is often questioned if such techniques are only some kind of brute-force statistical approaches and if they can only work in the context of High Performance Computing with tons of data. Another important question is whether they are really biologically inspired, as claimed in certain cases, and if they can scale well in terms of "intelligence". The dissertation is focused on trying to answer these key questions in the context of Computer Vision and, in particular, Object Recognition, a task that has been heavily revolutionized by recent advances in the field. Practically speaking, these answers are based on an exhaustive comparison between two, very different, deep learning techniques on the aforementioned task: Convolutional Neural Network (CNN) and Hierarchical Temporal memory (HTM). They stand for two different approaches and points of view within the big hat of deep learning and are the best choices to understand and point out strengths and weaknesses of each of them. CNN is considered one of the most classic and powerful supervised methods used today in machine learning and pattern recognition, especially in object recognition. CNNs are well received and accepted by the scientific community and are already deployed in large corporation like Google and Facebook for solving face recognition and image auto-tagging problems. HTM, on the other hand, is known as a new emerging paradigm and a new meanly-unsupervised method, that is more biologically inspired. It tries to gain more insights from the computational neuroscience community in order to incorporate concepts like time, context and attention during the learning process which are typical of the human brain. In the end, the thesis is supposed to prove that in certain cases, with a lower quantity of data, HTM can outperform CNN.

Studio multi-banda dal sub-parsec al kilo-parsec dell'emissione del getto nella radio galassia IC 1531

In questo lavoro di tesi si presenta il primo studio multi-scala e multi-frequenza focalizzato sul getto della radiogalassia IC1531 (z=0.026) con i satelliti Chandra, XMM-Newton e Fermi con l’obiettivo di tracciarne l’emissione alle alte energie; definire i processi radiativi responsabili dell’emissione osservata e stimare i principali parametri fisici del getto; stimare l’energetica del getto alle diverse scale. La sorgente è stata selezionata per la presenza di un getto esteso (≈5’’) osservato in radio e ai raggi X, inoltre, era riportata come possibile controparte della sorgente gamma 3FGLJ0009.6-3211 presente nel terzo catalogo Fermi (3FGL). La presenza di emissione ai raggi γ, confermata dal nostro studio, è importante per la modellizzazione della SED della regione nucleare. L’emissione X del nucleo è dominata da una componente ben riprodotta da una legge di potenza con indice spettrale Γ=2.2. L’analisi dell’emissione in banda gamma ha evidenziato una variabilità su scale di 5 giorni, dalla quale è stato possibile stimare le dimensioni delle regione emittente. Inoltre viene presentato lo studio della distribuzione spettrale dell’energia della regione nucleare di IC 1531 dalla banda radio ai raggi γ. I modelli ci permettono di determinare la natura dell’emissione gamma e stimare la potenza cinetica del getto a scale del su-pc. Gli osservabili sono stati utilizzati per ottenere le stime sui parametri del modello. La modellizzazione così ottenuta ha permesso di stimare i parametri fisici del getto e la potenza trasportata del getto a scale del sub-pc. Le stime a 151MHz suggerisco che il getto abbia basse velocita' (Γ≤7) e angolo di inclinazione rispetto alla linea di vista 10°<ϑ<20°; nel complesso, il trasporto di energia da parte del getto risulta efficiente. L’origine dell’emissione X del getto a grandi scale è consistente con un’emissione di sincrotrone, che conferma la classificazione di IC1531 come sorgente di bassa potenza MAGN.

Spectral distortions of the cosmic microwave background and small-scale gravitational waves as new windows on the Early Universe

In this thesis, we explore constraints which can be put on the primordial power spectrum of curvature perturbations beyond the scales probed by anisotropies of the cosmic microwave background and galaxy surveys. We exploit present and future measurements of CMB spectral distortions, and their synergy with CMB anisotropies, as well existing and future upper limits on the stochastic background of gravitational waves. We derive for the first time phenomenological templates that fit small-scale bumps in the primordial power spectrum generated in multi-field models of inflation. By using such templates, we study for the first time imprints of primordial peaks on anisotropies and spectral distortions of the cosmic microwave background and we investigate their contribution to the stochastic background of gravitational waves. Through a Monte Carlo Markov Chain analysis we infer for the first time the constraints on the amplitude, the width and the location of such bumps using Planck and FIRAS data. We also forecast how a future spectrometer like PIXIE could improve FIRAS boundaries. The results derived in this thesis have implications for the possibility of primordial black holes from inflation.