María Castrillo Melguizo

arXiv2026

Wattnet: matching electricity consumption with low-carbon, low-water footprint energy supply

The environmental impact of electricity consumption is commonly assessed through its carbon footprint (CF), while water-related impacts are often overlooked despite the strong interdependence between energy and water systems. This is particularly relevant for electricity-intensive activities such as data center (DC) operations, where both carbon emissions and water use occur largely off-site through electricity consumption. In this work, we present Wattnet, an open-source tool that jointly assesses the CF and water footprint (WF) of electricity consumption across Europe with high temporal resolution. Wattnet implements an electricity flow-tracing methodology that accounts for local generation mixes, as well as for cross-border electricity imports and exports at a 15-minute resolution. Operational and life-cycle impact factors are used to quantify and compare local (generation-based) and global (consumption-based) footprints for multiple European regions during 2024. Wattnet includes a 72-h forecasting module to facilitate informed energy-aware decision-making. The results demonstrate that neglecting electricity flows and temporal variability can lead to significant misestimations of both CF and WF, particularly in countries with high levels of electricity trade or hydropower dependence. Furthermore, the joint analysis reveals trade-offs between decarbonisation and water use, highlighting the prominent role of reservoir-based hydropower in increasing WF even in low-carbon systems. Wattnet facilitates informed energy-aware decision-making, while also enhancing transparency regarding the environmental impacts of electricity consumption for end users and policymakers.

physics.soc-ph↗

arXiv2022

Forecasting COVID-19 spreading trough an ensemble of classical and machine learning models: Spain's case study

In this work we evaluate the applicability of an ensemble of population models and machine learning models to predict the near future evolution of the COVID-19 pandemic, with a particular use case in Spain. We rely solely in open and public datasets, fusing incidence, vaccination, human mobility and weather data to feed our machine learning models (Random Forest, Gradient Boosting, k-Nearest Neighbours and Kernel Ridge Regression). We use the incidence data to adjust classic population models (Gompertz, Logistic, Richards, Bertalanffy) in order to be able to better capture the trend of the data. We then ensemble these two families of models in order to obtain a more robust and accurate prediction. Furthermore, we have observed an improvement in the predictions obtained with machine learning models as we add new features (vaccines, mobility, climatic conditions), analyzing the importance of each of them using Shapley Additive Explanation values. As in any other modelling work, data and predictions quality have several limitations and therefore they must be seen from a critical standpoint, as we discuss in the text. Our work concludes that the ensemble use of these models improves the individual predictions (using only machine learning models or only population models) and can be applied, with caution, in cases when compartmental models cannot be utilized due to the lack of relevant data.

cs.LG↗