Publications

Urban landscape artificialization leads to increased flood hazards in the most vulnerable locations (Under review)

Rapid urbanization worsens cities’ environmental burden, amplifying flood risks due to climate change. Artificial landscapes lack ecological functions, impacting human lives. Modeling urban water flows aids in minimizing future flood risks, but it’s complex. We study on flood protection in densely populated cities in Europe, and highlights the need for land use policies integrating natural flood-regulating features in urban planning, accounting for climate change resilience.

Calibration for an Ensemble of Grapevine Phenology Models under Different Optimization Algorithms

Vine phenology modeling for winegrowers is crucial. Calibration factors affecting error variability were studied, indicating initial parameter boundary and grape variety as significant sources of uncertainty. Choice of optimization algorithm had an impact, while non-linear models outperformed linear models in predicting flowering timing. Understanding parameter boundaries and error variance is important for calibration. Optimization methods and continuous model improvement are important.

An expanded framing of ecosystem services is needed for a sustainable urban future

Urban activities impact ecosystem services. A model called ESTIMUM predicts changes in ecosystem services due to urbanization. Urban sustainability policies should consider impacts beyond the city boundaries. Dietary shifts can have a greater positive effect than urban greening. This study highlights the need for an expanded multi-scalar conceptualization of urban sustainability that accounts for impacts beyond city boundaries.

Projections of Climate Change Impacts on Flowering-Veraison Water Deficits for Riesling and Müller-Thurgau in Germany

Climate change is projected to affect grapevine cultivation in Germany, particularly impacting white wine cultivars Riesling and Müller-Thurgau. Using a soil-crop model and climate simulations, the study found that the flowering and veraison stages would be affected, with advancements in flowering and slight advancements in veraison. Müller-Thurgau would experience a shortened phenophase, while Riesling would see an extended phenophase in cool and high-elevation areas. Riesling would also face higher temperature increases, leading to increased water deficit stress. The study emphasizes the threat to white wine production and the need to reduce uncertainties in climate change impact assessment.

A Data-Driven Surrogate Modelling Approach for Acceleration of Short-Term Simulations of a Dynamic Urban Drainage Simulator

Surrogate modelling is an approach to develop a simpler and faster model emulating the outputs of a more complex simulator as a function of its inputs and parameters. Surrogate models are useful when the simulators are computationally expensive for applications such as model-based real time control, model calibration, design optimization, Monte Carlo based uncertainty propagation analysis, or sensitivity analysis. In the present research, we focus on the application of a Gaussian Processes Emulator to develop a data-driven surrogate model.

stUPscales, an R-Package for Spatio-Temporal Uncertainty Propagation across Multiple Scales with Examples in Urban Water Modelling

We present the R package stUPscales. This package constitutes a contribution in the state-of-art of open source tools that aim to support tasks of uncertainty propagation in the spatial and temporal domains. It opens the possibility of applying the software in studies involving Monte Carlo based uncertainty propagation analysis or long-term simulations across different scales when linking sub-models at different spatial and temporal scales, as it is often required in integrated environmental modelling.

EmiStatR, a Simplified and Scalable Urban Water Quality Model for Simulation of Combined Sewer Overflows

In this study, we present the EmiStatR model with parallel computing capabilities and validate it to independent observations as well as simulation runs of a complex and detailed model. We prove the hypothesis that for urban catchments it is possible to have a similar accuracy for simulating combined sewer overflow (CSO) systems, and CSO loads of chemical variables proxies of water pollution, by using the proposed simplified scalable model as compared to the complex detailed model which implements the dynamic wave routing for solving the complete one-dimensional Saint Venant flow equations.