This work deals with the problem of face parameterization from the video of a speaking person and estimating Parkinson’s disease and the progress of its symptoms based on face parameters. It describes the syntax and function of the program that was created within this work and solves the problem of face parameterization. The program formats the processed data into a time series of parameters in JSON format. From these data, a dataset was created, based on which artificial intelligence models were trained to predict Parkinson’s disease and the progress of its symptoms. The process of model training and their results are documented within this work.

As the most produced grain crop world-wide, 5% of corn is lost due to stalk lodging (above-ground structural failure of the stalk near the roots). Current modeling methods lack the ability to manipulate the stalk architecture. In contrast, parameterized models enable advanced analyses such as sensitivity and optimization studies. This thesis advances previous work on a parameterized cross-sectional model of maize stalk morphology and investigates the validity of a parameterized three-dimensional model. The parameterized cross-sectional model is based upon previous work that approximated the cross-section of maize stalks using an ellipse plus principal components. Validation of the cross-sectional model was done by evaluating the structural response in four loading cases: axial tension/compression, bending, transverse compression, and torsion. 2D prismatic extrusions of specimen specific cross-sections were tested under the load conditions and compared against 2D prismatic models of the parametrized cross-sections. Analysis of the 2D prismatic model consisted of a parameter sensitivity analysis to determine influential morphological features, and a load bearing analysis to quantify the proportion of load borne by each material tissue. Validation of the parameterized 3D model was completed by comparing the structural response of the parameterized 3D model against empirical test data. A comparison against CT-based finite element models was also done to quantify the level of predictive discrepancy caused by geometric parameterization. The elliptical 2D prismatic model responded with less than 5% error for axial tension/compression, bending, and transverse compression, suggesting that the ellipse model is sufficient for analyses and 3D parameterization. The 2D prismatic model maintained an error less than 10% under a torsion load. The parameter sensitivity analysis revealed that ellipse parameters are significantly more influential to stalk strength than material or finer geometric details. In the load bearing analysis, the rind bore a median of over 90% of the load in axial tension/compression, bending, and torsion, but less than 10% of the load in transverse compression. The parameterized 3D model validates yielding correlations with empirical test data resulting in R2 values of 0.82 in flexural stiffness, and in critical buckling a value of 0.71. Comparison with CT-based models resulted in very strong correlations with R2 values of 0.81 in flexural stiffness, and for critical buckling a value of 0.87. The parameterized 3D model validates and can be used in future studies.

No description available.

No description available.

Bakgrund: Sveriges riksdag har lagstiftat ett långgående miljömål om att Sverige som nation ska vara klimatneutralt till år 2045. I Sverige står bygg- och fastighetsbranschen för ca 18 % av det totala nationella utsläppet av växthusgaser. Det växande behovet av byggande gör att det blir allt viktigare att ta hänsyn till ökande klimatpåverkan från byggnadsmaterialtillverkningen, byggandet och förvaltningen. Dokumentation av bygg- och anläggningsprojekts klimatpåverkan i form av Livscykelanalyser (LCA) börjar att efterfrågas av flera offentliga och privata beställare. Det är dock väldigt tids- och resurskrävande att utföra en LCA för ett projekt. Beräkning och planering av åtgärder för att minska klimatpåverkan, baserat på en LCA är inte ännu kopplat till tekniska projekteringen. Detta utgör ett hinder för att byggsektorn kan uppnå miljömålet om klimatneutralitet.Svenska byggföretag lagrar projektdata i olika IT-baserade produktions- och planeringssystem t.ex. kalkylsystem, digitala designprogram (CAD, Computer Aided Design), interna LCA-verktyg etc. Ett sätt för att effektivisera och göra mindre kostsamma LCA:er är att använda redan befintlig information i CAD-programvara. Den största offentliga beställaren på anläggningssidan Trafikverket ställer krav både på leverans av byggprojekthandlingar i ett digitalt format och rapportering av minskad klimatpåverkan i sina projekt. Idag saknas dock ett smidigt sätt att överföra projektinformation, t.ex. ingående byggnadsmaterial och dess mängder för konstruktionsdelar, från CAD-verktyg till LCA-programvara.Det nya byggklassifikationssystemet, CoClass erbjuder ett nytt sätt att strukturera digital projektinformation i CAD-verktygen. Det pågår en rad branschgemensamma initiativ för att utreda om användandet av CoClass kan förenkla överföring av projektinformationen från digitala modeller till LCA-programvaror.Syfte: Syftet med denna rapport var att ta fram en arbetsprocess för parametrisering av digitala modeller i enlighet med det nya CoClass-systemet och överföring av informationen på ett digitalt sätt till en LCA-programvara.Metod: Tillvägagångssättet baserades på flertalet semistrukturerade intervjuer. Intervjuerna inriktade sig mot konstruktörer och klimat/LCA-samordnare. I arbetet utfördes parametrisering av en digital modell för en betongbro i en CAD-programvara, Tekla och överföring av informationen till en LCA-programvara, Gabi.Slutsats: Resultatet från intervjuerna och testerna visar att det är tekniskt möjligt att använda CoClass för att parametrisera informationen i digitala modeller med avseende på LCA redan idag. Det har också konstaterats att det i nuläget saknas incitament för beställare, entreprenörer och konsulter att börja strukturera sina digitala modeller enligt CoClass- systemet. En ytterligare slutsats är att de ökade beställarkraven på minskandet av klimatpåverkan i bygg- och anläggningsprojekt skulle kunna bli en drivkraft för olika marknadsaktörer att gå över till CoClass-systemet. Implementering av CoClass- systemet i byggbranschen kommer skapa nya affärsmöjligheter med klimateffektiva konstruktioner och tekniska utföranden, vilket kommer stödja uppfyllandet av Sveriges mål om klimateffektivitet. / Background: The Swedish Parliament has legislated a long-term environmental objective that Sweden as a nation should be climate neutral by 2045. In Sweden, the construction and real estate industry accounts for about 18% of the total national greenhouse gas emissions. The growing need for expanding construction makes it important to take into account the increasing climate impact of production of building materials, construction and maintenance. Documentation of the construction project's carbon footprint in terms of life cycle analysis (LCA) begins to be requested by several public and private clients. Creating an LCA-report today is a very time and resource consuming process. There is no connection between calculation and planning of climate impact, based on an LCA and the technical project planning which constitutes an obstacle to reach the climate targets.Swedish construction companies store project data in various IT-based production- and planning systems, eg. calculation systems, digital design programs (CAD, Computer Aided Design), internal LCA tools, etc. The Swedish Transport Administration (Trafikverket) sets demands on both delivery of construction documentation in a digital format and reports of reduced climate impact in their projects. But currently the procedure lacks an effective way to transfer project information, such as building materials and their amount of construction parts, from CAD tools to LCA software.The new building classification system, CoClass, offers a new way of organizing digital project information in CAD tools. Initiatives from the construction industry are underway to investigate whether the use of CoClass can simplify the transfer of project information from digital models to LCA software.Purpose: The purpose of this report is to develop a process of parameterization in digital models in accordance with the new CoClass system and investigating ways of transferring information in a digital way to an LCA software.Method: In this report, the approach has been several semi-structured interviews. The interviews ware targeting building constructor and climate/LCA coordinators. In the work based on information from the interviews a parametrization of a concrete bridge was performed in a CAD-software, Tekla and further the information was transferred to LCA software, GaBi.Conclusion. The results of the interviews and the test shows that it is technically possible to use CoClass to parameterize the information in digital models regarding to LCA, today. There is currently no incentive for clients, entrepreneurs and consultants to start structuring their digital models according to the CoClass system. But the increased demand requirements for reducing climate impact in construction projects can become a spur for different market participants to switch to the CoClass system. Implementation of the CoClass system in the construction industry will create new business opportunities with climate-efficient constructions and technical designs, which will support the fulfillment of Sweden's climate efficiency targets.

Το αντικείμενο της διατριβής είναι η παρατηρησιακή και αριθμητική μελέτη των δυναμικών και φυσικών διεργασιών που συνδέονται με τη θερινή καταιγιδοφόρο δραστηριότητα στον Ελλαδικό χώρο. Αναλυτικότερα τα εξεταζόμενα αντικείμενα είναι τα εξής: 1. Η μελέτη των ηλεκτρικών εκκενώσεων στον Ελλαδικό χώρο κατά τη θερινή περίοδο για 4 τουλάχιστον έτη. Για την πραγματοποίηση των στόχων του παραπάνω αντικειμένου έχει ήδη γίνει η συλλογή και η επεξεργασία των ηλεκτρικών εκκενώσεων στη περιοχή του ευρύτερου Ελλαδικού χώρου για τις θερινές περιόδους των ετών 2003 έως και 2006. Τα δεδομένα που έχουν χρησιμοποιηθεί προέρχονται από το σύστημα καταγραφής των ηλεκτρικών εκκενώσεων της Βρετανικής Μετεωρολογικής Υπηρεσίας ATD (Arrival Time Difference). Τα γεγονότα τα οποία κατεγράφησαν στον Ελλαδικό χώρο για την περίοδο μελέτης, μετά τον απαραίτητο έλεγχο ποιότητας, κατανεμήθηκαν σε πλέγμα διαστάσεων 0,1° x 0,1° και υπολογίστηκε η πυκνότητα των ηλεκτρικών εκκενώσεων ανά τετραγωνικό χιλιόμετρο ανά θερινή περίοδο. Στις περιοχές με τις υψηλότερες τιμές πυκνότητας ηλεκτρικών εκκενώσεων συγκαταλέγονται η Μακεδονία, η Ήπειρος, ο ορεινός κεντρικός ηπειρωτικός κορμός καθώς και οι ορεινές περιοχές της Πελοποννήσου, γεγονός που αποτελεί μία πρώτη ένδειξη της σύνδεσης που υπάρχει μεταξύ της θέρμανσης του εδάφους, λόγω της ηλιακής ακτινοβολίας, και της ανάπτυξης θερινών καταιγίδων. Η άποψη αυτή ενισχύεται και από το γεγονός ότι το μέγιστο της δραστηριότητας παρατηρείται τις μεσημεριανές και πρώτες απογευματινές ώρες. 2. Η μελέτη των παραμετροποιήσεων ανοδικών κινήσεων (convection) και μικροφυσικής των αριθμητικών μοντέλων πρόγνωσης καιρού με στόχο την ορθότερη πρόβλεψη της θερινής καταιγιδοφόρου δραστηριότητας που συνοδεύεται από έντονη ηλεκτρική δραστηριότητα. Στο αντικείμενο αυτό κύριος στόχος είναι η σύγκριση τριών διαφορετικών σχημάτων παραμετροποίησης που χρησιμοποιούνται στο αριθμητικό μοντέλο πρόγνωσης καιρού MM5. Τα σχήματα αυτά είναι τα κάτωθι: Kain – Fritch, Betts – Miller – Janjic και Grell και χρησιμοποιούνται στην παραμετροποίηση του μοντέλου για τον υπολογισμό της βροχόπτωσης που προέρχεται από τις κατακόρυφες κινήσεις μεταφοράς. Κατά τη σύγκριση των τριών σχημάτων παραμετροποίησης τα καλύτερα στατιστικά σκορ εμφανίζει το σχήμα Kain – Fritch ενώ τα αποτελέσματα που δίνει το σχήμα του Grell είναι αρκετά κοντά με αυτά του προαναφερθέντος σχήματος. 3. Η τροποποίηση του σχήματος παραμετροποίησης κατακόρυφων κινήσεων στην ατμόσφαιρα των Kain – Fritsch με στόχο τη βελτιστοποίηση των αποτελεσμάτων της προβλεπόμενης βροχόπτωσης στην περιοχή του Ελλαδικού χώρου, κατά τη θερμή περίοδο του έτους Στο αντικείμενο αυτό έχει μελετηθεί στατιστικά η ικανότητα πρόβλεψης του πεδίου της βροχόπτωσης, εφαρμόζοντας ποικίλες τροποποιήσεις του σχήματος παραμετροποίησης Kain – Fritsch. Πιο συγκεκριμένα πέρα από την βασική έκδοση του σχήματος που χρησιμοποιείται στην επιχειρησιακή αλυσίδα του προγνωστικού μοντέλου ΜΜ5, αναπτύχθηκαν οκτώ συνολικά τροποποιήσεις. Εφαρμόζοντας τις παραπάνω τροποποιήσεις, πραγματοποιήθηκαν συνολικά 180 πειράματα για συνολικά 20 διαφορετικές ημέρες με έντονη βροχόπτωση κατά την θερμή περίοδο των ετών 2005 – 2007. Η σύγκριση των αποτελεσμάτων παρείχε ενθαρρυντικά στοιχεία, καθώς ωθώντας το σχήμα να παράξει περισσότερο υετό παρατηρήθηκε βελτίωση στην ικανότητα του μοντέλου να προβλέπει τον υετό. / The present Ph.D. Thesis is divided into three main parts. In the first part the lightning activity over Greece during the warm season (May-September) of the years 2003 through 2006 is investigated, in relation with the synoptic meteorological conditions that prevailed in the region. The study is based on the use of cloud-to-ground lightning activity data and of upper-air analyses. The analysis of the spatial variability of lightning shows that the highest “relative” flash densities are observed in Northern and Western Greece and in Central and Western Peloponnissos. The study of the synoptic patterns related to lightning is based on the analysis of 60 active and 60 inactive days in terms of lightning activity over Greece. The days with high lightning activity are characterised by a short wave trough at the 500 hPa level over the Ionian Sea. On the other hand, during the days with no lightning a northwest flow prevails over Greece. It was also found that high lightning activity is related with high values of absolute vorticity, equivalent potential temperature and convective available potential energy. In the second part the sensitivity of numerical model quantitative precipitation forecasts to the choice of the convective parameterization scheme (CPS) is examined for twenty selected cases characterized by intense convective activity over Greece, during the warm period of 2005 – 2007. Namely, the study is conducted using MM5 model and the following three different CPSs: Kain – Fritsch, Grell and Betts – Miller – Janjic. Sixty numerical simulations were carried out. The simulated precipitation was verified against raingauge measurements and lightning data. Verification results showed that for all three schemes the model presented a tendency to overestimate light to moderate rain while in general it underestimated the high precipitation amounts. The validation against both sources of data showed that among the three CPSs, the more consistent behaviour in quantitative precipitation forecasting was obtained by the Kain – Fritsch scheme that provided the best statistical scores. However, the differences of the results of statistical analysis between the Kain – Fritsch and Grell schemes were not large. In the third part the sensitivity of quantitative precipitation forecasts to various modifications of the Kain – Fritsch (KF) convective parameterization scheme (CPS) is examined for twenty selected cases characterized by intense convective activity and widespread precipitation over Greece, during the warm period of 2005-2007. The modifications include: (i) the maximization of the convective scheme precipitation efficiency, (ii) the change of the convective time step, (iii) the forcing of the convective scheme to produce more/less cloud material, (iv) changes to the trigger function and (v) the alteration of the vertical profile of updraft mass flux detrainment. In general, forcing the model to produce less cloud material improves the precipitation forecast for the moderate and high precipitation amounts.

Parameterization development and evaluation ideally takes a two-step approach (Lohmann et al., 2007). Insight into new processes, and initial parameterization formulation should be guided by theory, process-level observations (laboratory experiments or field studies) or, if these are unavailable, by high-resolution modelling. However, once implemented into large-scale atmospheric models, a thorough testing and evaluation is required in order to assure that the parameterization works satisfactorily for all weather situations and at the scales the model is applied to. Satellite observations are probably the most valuable source of information for this purpose, since they offer a large range of parameters over comparatively long time series and with a very large, to global, coverage. However, satellites usually retrieve parameters in a rather indirect way, and some quantities (e.g., vertical wind velocities) are unavailable. It is thus essential for model evaluation 1. to assure comparability; and, 2. to develop and apply metrics that circumvent the limitations of satellite observations and help to learn about parameterizations. In terms of comparability, the implementation of so-called \"satellite simulators\" has emerged as the approach of choice, in which satellite retrievals are emulated, making use of model information about the subgrid-scale variability of clouds, and creating summary statistics (Bodas-Salcedo et al., 2011; Nam and Quaas, 2012; Nam et al., 2014). In terms of process-oriented metrics, a large range of approaches has been developed, e.g. investigating the life cycle of cirrus from convective detrainment (Gehlot and Quaas, 2012), or focusing on the details of microphysical processes (Suzuki et al., 2011). Besides such techniques focusing on individual parameterizations, the data assimilation technique might be exploited, by objectively adjusting convection parameters and learning about parameter choices and parameterizations in this way (Schirber et al., 2013).In this chapter, we will first introduce the available satellite data, consider their limitations and the approaches to account for these, and then discuss observations-based process-oriented metrics that have been developed so far.

The North American monsoon (NAM) is the principal driver of summer severe weather in the Southwest U.S. With sufficient atmospheric instability and moisture, monsoon convection initiates during daytime in the mountains and later may organize, principally into mesoscale convective systems (MCSs). Most monsoon-related severe weather occurs in association with organized convection, including microbursts, dust storms, flash flooding and lightning. The overarching theme of this dissertation research is to investigate simulation of monsoon severe weather due to organized convection within the use of regional atmospheric modeling. A commonly used cumulus parameterization scheme has been modified to better account for dynamic pressure effects, resulting in an improved representation of a simulated MCS during the North American monsoon experiment and the climatology of warm season precipitation in a long-term regional climate model simulation. The effect of urbanization on organized convection occurring in Phoenix is evaluated in model sensitivity experiments using an urban canopy model (UCM) and urban land cover compared to pre-settlement natural desert land cover. The presence of vegetation and irrigation makes Phoenix a "heat sink" in comparison to its surrounding desert, and as a result the modeled precipitation in response to urbanization decreases within the Phoenix urban area and increase on its periphery. Finally, analysis of how monsoon severe weather is changing in association with observed global climate change is considered within the context of a series of retrospectively simulated severe weather events during the period 1948-2010 in a numerical weather prediction paradigm. The individual severe weather events are identified by favorable thermodynamic conditions of instability and atmospheric moisture (precipitable water). Changes in precipitation extremes are evaluated with extreme value statistics. During the last several decades, there has been intensification of organized convective precipitation, but these events occur with less frequency. A more favorable thermodynamic environment for monsoon thunderstorms is the driver of these changes, which is consistent with the broader notion that anthropogenic climate change is presently intensifying weather extremes worldwide.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.

Non disponible. / Aerosol particles are ubiquitous in the Earth’s atmosphere. Although a minor constituent of the atmosphere, the aerosol particles are linked to visibility reduction, adverse health effects and heat balance of the Earth. The secondary aerosols which are formed in the atmosphere from the gaseous phase : precursor gases become particles by nucleation and condensation (Seinfeld and Pandis, 1998) represents the largest source in a number concentration of atmospheric particles. The chemical reactions can play an important role by turning high volatility gases into species with low vapor pressure and thus high saturation ratio, i.e. creating favorable conditions for particulate matter formation. In this work the CHIMERE chemical transport model is used to ameliorate our understanding of the governing processes for aerosol formation and to investigate its capability to reproduce the mass and number concentrations and temporal evolution of the aerosols particles at high altitudes (as for example Puy de Dome research station), and in particular, evaluate its capacity to simulate the formation of new particles due to nucleation. For the studied cases it was investigated the impact of : a fine resolution topographical database on the accuracy of simulation of dynamical parameters at high altitude, of the use of different emissions databases in the accuracy of gas-phase and aerosol concentration predictions, what is the most adequate nucleation parameterization scheme for simulating new particle formation at high altitude and what is the influence of the choice of the primary particle size distribution on the prediction of new particle formation. Also the ability of the different theories to reproduce the occurrence or lack of a nucleation event is evaluated.