Abstracts (2023)
Ariel Guersenzvaig (Elisava)
Title: Can machine learning make naturalism about Health truly naturalistic? A reflection on a data-driven concept of Health
In this talk, using the notion of health as a leitmotif, we critically examine the hypothetical application of big data and machine learning to generate a naturalist (i.e., neutral, objective, and value-free) definition of health. Our objective is to analyze, clarify, and explore various philosophical questions raised by the use of machine learning for epistemic purposes. The paper argues that a sweeping "dataist" approach cannot be used to generate naturalist definitions, as prior theories and values persist. The talk further argues against delegating the final authority for defining complex concepts like health to AI systems, as this task necessitates ethical judgment and capacities for deliberation that AI currently lacks. We also caution against granting creators and deployers of AI systems discretionary authority to determine these definitions outside of wider social discussions, and advocate for ongoing public engagement on normative notions. To conclude, the talk will briefly outline some applied projects that are currently being carried out at Elisava Research, which are related to data and well-being and involve participatory techniques.
Isabel Ruiz-Mallén (Universitat Oberta de Catalunya)
Title: Examining the multiple co-benefits of nature-based solutions for climate adaptation in school environments
The European research project COOLSCHOOLS (https://coolschools.eu) explores and analyzes the impacts of implementing of Nature-based climate school shelters, which we understand as co-created nature-based, enabling environments for responding to climate change in schoolyards that ensure wellbeing, social equity, learning and inclusivity regards schoolchildren, other school community members and the wider neighborhood community of actors around schools. The research builds and expands on the experience of Barcelona as a pioneer European city in the implementation of nature-based solutions for climate change adaptation in school environments through the Refugis Climàtics and Transformem els patis municipal programs. We combine natural, social, and education sciences and rely on qualitative and qualitative data to shed light on the transformation capacities of nature-based climate school shelters regarding accessibility and equity, biodiversity and green connectivity, inclusive governance and climate education.
Mark Nieuwenhuijsen (ISGlobal)
Title: Urban Burden of Disease Estimation for Policy Making (UBD Policy)
The Urban Burden of Disease Estimation for Policy Making (UBD Policy) project aims to estimate the health impacts resulting from urban and transport planning related exposures in nearly 1000 European cities. So far, quantitative burden of disease assessments have been conducted for the year 2015 to estimate the premature mortality due to current levels of major environmental stressors such as air pollution, noise, urban heat islands and lack of green space. Future work of UBD Policy will extend the health impact estimation for the nearly 1000 European cities to 2018, 2021 and 2024, and explore new methods and health and well-being indicators (e.g., quality of life, subjective general health, well-being, happiness, stress, annoyance, sleep disturbance, productivity, etc.). Further aims of the project are to rank cities based on exposure levels and health impacts, develop a Healthy City Index (that takes into account exposure levels, health outcomes and urban design features) and establish a knowledge translation hub; to ultimately strengthen evidence-based decision making across Europe.
The objectives of the presentation are: (1) introduction to UBD Policy, (2) previous findings from quantitative burden of disease assessments, specifically for the city of Barcelona, and (3) future work of UBD Policy.
Mercè Crosas - Barcelona Supercomputing Center (BSC)
Title: Data Science for the Social Sciences
The social sciences play a pivotal role in the well-being of cities, in addition to engineering, urban and environmental sciences, and should be considered as a component of science-based solutions to most city´s challenges. This talk will first introduce the new Computational Social Sciences Program at the Barcelona Supercomputing Center, and then will propose what is needed to incorporate social science research into data science approaches towards the improvement the well-being of cities.
Sílvia Casorran - Area Metropolitana de Barcelona (AMB)
Title: Transforming mobility and public space to reach liveable cities
Mobility is one of the most challenging issues nowadays in urban areas. Places need to be daily accessible to people and goods, but the great invention of motorized transport has created several unwanted impacts on health and the environment: air pollution, noise, CO2 emissions (heat island effect), traffic crashes, lack of public space for people and greenery, sedentary lifestyle, and etc.
Cities are trying to cope with all these negative impacts by transforming public space into a more livable place. This means removing part of the space dedicated to cars to give it back to people, bikes, and public transit. However, at the same time, there is significant social resistance, as the car is still considered a symbol of progress. What strategies are being implemented nowadays in the Barcelona metropolitan area to address this situation?
Jan Mateu - Barcelona Supercomputing Center (BSC,UPC)
Title: Air quality data in Barcelona: data fusion and probabilistic estimations
Air pollution is considered one of the main leading environmental risk factors, especially at the city level. In Barcelona, the only two traffic stations of air quality (Eixample and Gràcia) repeatedly reported NO2 exceedances of the WHO thresholds. Building a NO2 database at high resolution is not easy; the difficulties in characterizing spatial gradients in urban areas, the low number of monitoring stations, and the uncertainty linked to urban air quality models challenge accurate NO2 assessments. We present a comprehensive database of NO2 in Barcelona using state-of-the-art data-fusion methodologies. This database includes the estimation of uncertainty in NO2 predictions at a high spatio-temporal resolution. We have found that 13 % of the Barcelona area has a probability of 70% or higher of exceeding the NO2 annual legal limit of 40 µg m−3. We will discuss how these results interact with other disciplines, such as public health, policy-making, and mobility.
Giovanni Maccani (Ideas for change)
Title: New Data Institutions: from citizen science to data altruism
At Ideas for Change we have specialized in the design and development of projects in which technology and data are put at the service of humans' well-being, and not the other way around. For some years now we have been developing innovation projects and accompanying companies, public institutions, social entities and citizens in the development of research projects that contribute to a positive impact on a large scale.
Through the development of projects in the field of (extreme) citizen science and after leading several pilots in the Barcelona area (EU-funded projects such as Making Sense, WeCount or DNoses), we have detected some recurring pain points, such as the complexity of maintaining relevant citizen engagement over time, the subject of data reliability or the difficulties in developing models that are sustained beyond pilot stages. For this reason, and for some years now, we have been working on the development of new data institutions (under the protection of new European regulations such as the Data Governance Act), starting with a citizen data cooperative (SalusCoop) founded in 2017, and the recently created data altruism association DataLog (2023). DATALOG is a citizen data Association that offers the opportunity to aggregate and manage consumption data in the Barcelona metropolitan area. Using advanced analysis and prediction algorithms, we want to help citizens to make informed decisions to fight energy poverty and address climate emergency issues.
Pau Garcia (Domestic Data Streamers)
Title: Artificial Ignorance: An exploration of non-functional uses of AI
This talk will delve into the collaborative research done by the Domestic Data Streamers team, which comprises artists, programmers, journalists, and designers. The primary objective of this research is to explore the expansive capabilities of artificial intelligence beyond its conventional applications in automation, bias mitigation, workflow control, and optimization. Specifically, the focus will be on investigating how AI can significantly influence and transform social practices and the arts, highlighting its potential for substantial impact in these fields.
Gerard Gaya (CREAF) / Montserrat Busto (Servei Meteorologic de Catalunya)
Title: Citizen Science and Climate Change in Barcelona
Barcelona Reconnecta is a citizen science project carried out in the geographical area of Barcelona by the citizen observatory RitmeNatura from CREAF, the xarxa FenoCat from the Servei Meteorològic de Catalunya and the Autonomous University of Barcelona since November 2022, thanks to the collaboration of the Climate Emergency Action Plan 2030 of the Barcelona City Council. The logo of Barcelona Reconnecta is inspired by the Pla Clima de l'Ajuntament de Barcelona, which collaborates with the project.
Its main objective is to facilitate the reconnection of the population of the city of Barcelona with nature through phenological monitoring and to raise awareness of the impact of climate change in their immediate environment. It aims to deepen this reconnection, making it resilient and lasting as an essential value to perceive the natural world around us and to learn to measure it, becoming aware of the changes that are occurring and the direct impact and influence of climate change in this environment.
Marc Figuls (Factual Consulting)
Title: Safety assessment for cycling infrastructures
Lane Patrol, a robust road safety assessment tool designed for cycling infrastructure, adheres to the international and evidence-based standard CycleRAP. This comprehensive solution integrates both hardware and software components. It involves a specially equipped bicycle, fitted with a camera capable of capturing geolocated images at 20-meter intervals along the cycling infrastructure, encompassing cycle lanes and other roads designated for bicycle use. The coding software meticulously analyzes over 50 attributes derived from the CycleRAP methodology, including delineation, intersection type, lane width, and traffic interactions.
The CycleRAP algorithm, a pivotal component of Lane Patrol, conducts a risk analysis of the entire assessed network. This algorithm provides a nuanced understanding of risks, differentiating between potential crashes involving bicycles and vehicles, bicycle-to-bicycle scenarios, bicycle-pedestrian interactions, and solo bicycle-related risks.
Lane Patrol goes beyond assessment, offering actionable insights for road safety improvements. The solution proposes various safety treatments, presenting a quantitative simulation to demonstrate the potential impact of these enhancements. This includes a comprehensive before-and-after analysis, aiding in informed decision-making for the implementation of safety measures.
Lane Patrol has been successfully deployed in diverse cities, including Barcelona, Madrid, Oxford, Skopje, and Helsinki, underscoring its adaptability and effectiveness in addressing road safety challenges in different urban environments.
John Palmer (Universitat Pompeu Fabra)
Title: Citizen Science to Nowcast Disease-Vector Mosquitoes In Barcelona
This presentation will explain how citizen science can be combined with novel technologies and traditional mosquito surveillance to gain valuable insights about urban disease-vector mosquitoes for neighborhood-level control and public health interventions. It will describe an ongoing study and operational platform that brings these three data streams together in near real-time to estimate the spatio-temporal distribution of tiger mosquitoes (Aedes albopictus) at resolutions sufficiently fine-grained to facilitate day-to-day decision-making by public health authorities. The system combines data from the Mosquito Alert citizen science system, AI-driven “smart traps”, and traditional mosquito traps.