Cities are getting hotter, and not only because of climate change. The way cities are built plays a major role in how much heat they store and release, especially during summer and at night. This effect is known as the Urban Heat Island: urban areas are often several degrees warmer than the surrounding countryside.
A recent study led by researchers from Universitat Rovira i Virgili explores how different urban features influence this excess heat across cities in the Iberian Peninsula. Using high-resolution climate data and machine learning techniques, the team analyzed how temperature varies at the scale of city blocks, rather than treating cities as single, uniform units.
The researchers combined detailed meteorological information with data describing urban form, such as building height, building density, population density, and the amount of vegetation. This allowed them to estimate how each factor contributes to daily maximum heat intensity in cities like Barcelona, Madrid, or Valencia.
Their results are clear and intuitive. Taller buildings, denser construction, and higher population density all increase urban temperatures. For example, increasing average building height by just one meter can raise peak urban temperatures by almost 0.1 °C. Likewise, areas with more built-up surfaces tend to trap more heat. In contrast, vegetation has a cooling effect. Increasing green cover can reduce urban heat by more than one degree in some cases, confirming the importance of trees, parks, and green spaces.
One key advantage of this work is its local focus. Instead of producing city-wide averages, the model estimates heat intensity for each small area of the city, making it especially useful for urban planners. This level of detail can help identify heat-risk hotspots and design targeted solutions, such as increasing tree cover or adjusting building layouts.
As heatwaves become more frequent and intense, especially in southern Europe, studies like this provide practical tools to make cities healthier, more resilient, and more comfortable places to live.
Associate Professor at Universitat Rovira i Virgili since 2024 and member of the ECoMMFiT research group, with a well-established track record in fluid dynamics, heat and mass transfer, and advanced numerical modeling. He holds a PhD in Engineering from URV and has developed his research career at internationally renowned institutions such as the University of Ottawa and the City University of New York, as well as through the Beatriz Galindo program as a Distinguished Researcher.
His research focuses on the role of turbulence in transport and dispersion processes, with applications ranging from multiphase plumes in the marine environment and aerosol dispersion to urban air quality. In recent years, he has extensively integrated Machine Learning and Artificial Intelligence techniques to address complex problems in environmental science, health, and engineering, including clinical prediction in intensive care units, the impact of transport and maritime traffic on atmospheric pollution, and water network management.
He is the author or co-author of 38 scientific articles with significant impact, has supervised and co-supervised several doctoral theses, has led competitive national and international research projects, and has received recognition for the social impact of his research.
Associate Professor of Fluid Mechanics in the Department of Mechanical Engineering at Universitat Rovira i Virgili (URV) and Director of the Interuniversity Master’s Degree in Computational Fluid Mechanics (URV–UNIR). He began his academic career at URV in 1993 and, after defending his doctoral thesis in 1997 and completing a postdoctoral stay at the Boundary Wind Tunnel Laboratory of the University of Western Ontario (Canada), he obtained a permanent Associate Professor position in 2002. As of June 2025, he holds the ANECA accreditation as Full Professor.
He is a member of the ECoMMFiT and IU-RESCAT research groups, and his research activity focuses on data analysis, pattern recognition, image processing, and machine learning applied to fluid mechanics, heat transfer, and environmental and biomedical flows. He has carried out significant work in particle image velocimetry (PIV), including a patent on novel tracer particles, and more recently has oriented his research toward machine learning applications in health and the environment, contributing to the development of tools with recognized social impact.
He has served as principal investigator on several competitive national research projects and has led numerous technology transfer projects. He is co-author of 57 JCR-indexed articles, with an h-index of 19 according to Scopus, has co-supervised multiple doctoral theses, holds four research six-year periods (sexennis) and six teaching five-year periods (quinquennis), and has held several relevant academic management positions at URV.
Full Professor of Fluid Mechanics in the Department of Mechanical Engineering at the Universitat Rovira i Virgili since 2000, where he has taught undergraduate, postgraduate, and doctoral courses in various degree programs at ETSEQ and ETSE. He holds a degree in Chemical Sciences (1980) and a PhD in Chemical Sciences (1986) from the University of Barcelona, where he began his academic career, which he complemented with training and research stays at the Universities of Cambridge and Toronto.
He carried out a fifteen-month postdoctoral stay at the University of Cambridge, focused on the application of artificial intelligence algorithms and laser visualization systems, followed by a subsequent stay at the University of Toronto to study the effect of distortion on the turbulent wake of a cylinder. Before fully joining the field of Fluid Mechanics, he served as a lecturer in the areas of Statistics and Operations Research and Chemical Engineering.
He has participated in more than 60 competitive research projects, 35 of them as principal investigator, in the fields of fluid mechanics, heat transfer, and turbulence. He is a member of the consolidated research group ECoMMFiT, has supervised five doctoral theses, and is the author of more than 50 scientific publications and monographs, as well as 75 contributions to international scientific conferences.