Wine spoilage poses a significant threat to the global wine industry, leading to substantial economic losses and jeopardising its sustainability. This sector is already under strain, contending with issues including climate change, global competition, increasing production costs, declining consumption and evolving consumer preferences. Volatile Sulphur Compounds (VSCs) influence wine quality with their abundance and low perception threshold. While some VSCs enhance wine aromatic complexity, most of them cause negative wine faults such as aromas of cabbage and rotten eggs leading to discarded wine and substantial economic losses. Despite their significance, current methods for detecting and quantifying VSCs (particularly light VSCs) are inadequate. Additionally, a comprehensive model of sulfur metabolism is lacking. This project aims to develop innovative strategies for the accurate quantification, assessment, and management of VSCs during wine fermentation. The specific objectives are: (1) to identify VSC origins and characterise their production dynamics using a novel real-time monitoring device; (2) to assess sulfur compound flux distribution within yeast metabolic networks; and (3) to create the first model-based strategies for predicting and controlling VSC formation. The project includes a secondment to experts at the CSIC in Spain, where I will refine the model. This experience will significantly enhance my skills in analytical techniques and metabolic modelling while expanding my professional network and career prospects. The results of the project will offer valuable tools for both researchers in yeast metabolism and the wine industry, leading to improvements in quality control and production efficiency to mitigate the challenges faced. This pioneering research will provide a solid foundation for future projects in yeast metabolism and winemaking where the developed tools can be utilised.

Eddy Current Testing (ECT) is an industrial procedure used to assess the reliability of the most critical facilities of nuclear power plants. Prior to practical usage, ECT requires a delicate phase of calibration and validation via numerical simulations. However, in the current state-of-the-art, they are limited by two critical issues: the poor precision of the results, and the limited geometrical modeling flexibility. This project will use the recently introduced Discrete de Rham (DDR) simulation method to overcome at once both these issues, focusing on three specific aims: laying the mathematical foundations of DDR methods for ECT simulation, building the practical computational tools to implement this method in a simulator, and using the simulator on real-life ECT scenarios. The project has a strong multidisciplinary nature, involving a combination of numerical analysis and engineering, and its originality and innovation lie on this interdisciplinary approach. The mathematical analysis of DDR methods for ECT is a complete novelty and represents a mine of mathematical problems in numerical analysis. The development of the EffECT simulation (open-source) will require both the transfer of the candidate’s engineering knowledge to the host institution and the training of the candidate on the very specific DDR mathematical methods. The project will thus make the candidate able to speak to different communities, improving his career prospects as an independent researcher. The planned communication activities of the project will target people both inside and outside academia, helping the latter to appreciate the effectiveness of the interaction between fundamental mathematical research and engineering in solving real-life problems. The arising results have the potential to increase the safety and efficiency of nuclear power plants and open new horizons in fundamental and applied level of numerical analysis of DDR methods, as well as other cutting-edge simulation methods.

Coastal and rural villages in Eastern Africa are more likely to have mitigated or even avoided poverty and outmigration during the last three decades of land desertification and natural disasters when developing a sustainable blue economy providing alternative resources and livelihoods. To test this core hypothesis, we will rely on a long-term causal inference framework combining satellite imagery, artificial intelligence algorithms, and spatial matching statistical methods, but also multidisciplinary field surveys. More precisely, BLUE-AFRICA has the ambition to quantitatively and causally assess on the long term whether (i) maintaining productive coral reefs through the establishment of fisheries management, (ii) investing in mariculture, or (iii) developing nature-based tourism can be potential transformation pathways on the eastern African coast to improve wealth assets and prevent outmigration when arable land and terrestrial resources are vanishing under climate change. Four main objectives will be reached: 1. Mapping and characterizing all rural socio-ecological systems (RSES) on the coasts of Madagascar, Mozambique, and Tanzania, but also blue economy activities. 2. Building, testing and validating artificial intelligence algorithms accurately predicting the level of human poverty and outmigration across space and time in Africa from satellite imagery and auxiliary covariates. 3. Testing the causal link between the establishment of a blue economy activity and the level of poverty and outmigration in the drying rural eastern Africa over a long time period (up to 30 years) using a spatial matching method between control and treated RSES and then fitting several response curve models. 4. Identifying, visiting and understanding the ‘bright’ vs. the ‘dark’ spots, i.e., the RSES with the highest vs. lowest level of economic development and net migration using a deviance approach from model expectations, in-depth field surveys and statistical comparisons.

Upon external mechanical loading, amorphous solids initially deform elastically, but for large enough applied stress they may fracture or flow as a plastic fluid. This transition, known as yielding, is one of the most fascinating open problems in condensed matter statistical physics. In fact, the exact nature of the transition is still unclear: some systems show abrupt (brittle) yielding (e.g., in oxide glasses) while others, such soft gels or foams, display ductility with a smooth and gradual increase of plastic deformation. A unique picture of the processes at play is yet missing and recent models and theories claim experimental support. Here I will use a use a novel colloidal glass to explore the yielding transition and its properties. Colloidal glasses are controlled through their volume fraction: it plays the same role of temperature in molecular ones, but it is much more difficult to precisely tune. I will overcome this difficulty exploiting a mixture of nanoparticles and non-colloidal polymer mesogels. The mesogels swell/shrink as a function of temperature, controlling the available ‘free’ volume for the (temperature-insensitive) nanoparticles. This approach allows for an unprecedented precise control of the volume fraction in a colloidal glass without changing particle-particle interaction or their size (e.g., in microgels). The versatility of the system will be exploited to investigate plastic activity for different level of glass equilibration and external applied stresses thanks to cutting-edge experiments involving rheology, time- and space- resolved visible scattering and synchrotron-based X-ray photon correlation. These approaches will be combined to obtain a comprehensive picture of the processes at play, allowing for a ground-breaking study from inter-particle distances up to macroscopic sizes. The yielding transition will then be tackled from a new point of view, potentially elucidating its nature and confronting emerging theories.

Our goal is to construct generalisations of the Hitchin and Wess--Zumino--Witten (WZW) and Knizhnik--Zamolodchikov (KZ) connections, both in geometric and deformation quantisation, and of their associated monodromy representations. The Hitchin connection achieved the quantisation of compact Chern--Simons theory and resulted in the construction of a topological quantum field theory. A different projectively flat connection provides a viable mathematical definition of correlation functions in the WZW model for conformal field theory. The resulting projectively flat vector bundles are isomorphic, and their monodromies have far-reaching applications in low-dimensional topology/geometry (quantum invariants of knots/3-manifolds) and representation theory (of mapping class/quantum/braid groups). Our guiding viewpoint is that the connections of Hitchin/WZW can be derived from the quantisation of moduli spaces of connections on Riemann surfaces. We will extend this further, focusing on meromorphic connections with high-order poles (i.e., wild singularities), generalising the above bundles and their applications. The motivation for this project is twofold. First, there is now a complete understanding of the Poisson/symplectic nature of isomonodromic deformations of wild singularitites, which are naturally amenable to quantisation. The quantum theory is much less developed than the classical one, and this naturally motivates us to close the gap using the latter as a guide. Second, recent work related the genus-zero WZW connection---that is, the KZ connection---to a new version of the Hitchin connection, and this was then used for the quantisation of moduli spaces of parabolic bundles. We want to pursue extensions of this identification; in particular, we will use the new flat connections constructed on the deformation quantisation side as candidates for `wild' Hitchin connections, in the geometric quantisation of wild character varieties: a complete novelty.