Cities increasingly implement car-reducing measures to improve sustainability, public health, and urban liveability. However, their impacts are often evaluated using aggregate indicators such as traffic volumes or modal shares, providing limited insight into the behavioural mechanisms through which travellers adapt. In particular, little is known about how responses to reduced car accessibility differ across socio-demographic groups.

This thesis investigates the effects of car-reducing measures on travel behaviour using the redesign of Hofplein in Rotterdam as a case study. The intervention reduces road capacity by transforming a car-oriented traffic square into a multimodal public space, thereby reducing car accessibility at an important node in the transport network. The study examines both the overall behavioural impacts of this accessibility reduction and the extent to which these impacts vary between traveller groups.

An activity-based transport model (ActivitySim) was used to simulate traveller behaviour before and after the intervention. Accessibility changes resulting from the Hofplein redesign were first estimated using the V-MRDH traffic assignment model and subsequently incorporated into ActivitySim. To improve behavioural realism, the model was calibrated against both observed modal shares and literature-based travel-time elasticities using Bayesian optimisation. Simulated travellers were subsequently linked to ten mobility personas to analyse behavioural responses across different socio-demographic groups.

The calibrated model reproduced both observed modal shares and benchmark travel-time elasticities reasonably well. Simulation results indicate that the intervention reduces car use while increasing walking, cycling, e-bike use, and public transport demand. However, modal shift explains only part of the response. Many travellers continue travelling by car by selecting alternative routes, while others adapt through destination changes or trip suppression. The analysis further shows that impacts are distributed unevenly across the population, with car-oriented personas experiencing the largest accessibility reductions and behavioural responses.

These findings demonstrate the value of activity-based transport models for evaluating transport interventions. By capturing individual behavioural adaptation and distributional effects, such models provide a more detailed understanding of the impacts of car-reducing measures and can support more equitable transport policy development. ... Read more

Aviation accounts for approximately 3.5% of net anthropogenic climate forcing, a share expected to grow. Hydrogen and sustainable aviation fuels remain years from deployment, so near-term mitigation is needed. Contrails, ice clouds formed when aircraft fly through cold, humid air, are the largest contributor to aviation’s non-CO₂ climate impact and warm within hours rather than decades. Because contrail-forming regions are vertically thin, small altitude adjustments offer a promising mitigation route. This thesis investigates their feasibility within existing air traffic management constraints, using open remote sensing and flight surveillance data.
Chapter 2 explains that contrails form when engine exhaust mixes with sufficiently cold, humid air (the Schmidt-Appleman Criterion, SAC) and persist only in ice-supersaturated regions (ISSRs).
Chapter 3 tests whether open data can identify contrail formation. Contrails identified in GOES-16 imagery were matched to OpenSky trajectories and ERA5 data. SAC and ISSR together predicted persistent formation with 87% accuracy, validating ERA5 for this purpose.
Chapter 4 addresses formation altitude. A novel method estimates contrail altitude from the shadows contrails cast in Landsat 8 imagery, using solar geometry and the Hough transform to pair them. Validated against ADS-B GPS altitude, it is accurate to within a few hundred meters. A ResUNet with a novel SR loss, which penalizes deviations in Hough space to enforce contrail linearity, automates this. Contrail detection performed well, shadow detection less so.
Chapter 5 asks whether routine step-climb altitude changes suffice to avoid ISSRs. Of 5.7 million flights analyzed, about 3.5% produced persistent contrails. Among flights that could leave the ISSR within 2000 ft, about 50% could do so with a step of ±1000 or ±2000 ft.
Chapter 6 develops an open-source climate trajectory optimizer on OpenAP.top and applies it to nearly 1.4 million Euro-Atlantic flights. Climate-optimized trajectories reduced persistent contrail time by roughly 15% relative to fuel-optimal routing, at 20–25% more fuel. Since fuel-optimal routes are about 23% more efficient than flown trajectories, climate-optimal fuel burn stays comparable to current operations.
Chapter 7 examines operations and economics. For 1.6 million flights, optimization on ARPEGE forecasts differed from an ERA5 benchmark by only 0.8% in flight time and 2.2% in fuel burn, but retained about 25% more contrail time. In 91.3% of Flight Information Regions, traffic changed by at most five aircraft per three-hour window. The average cost was €1551 per flight.
Together, the chapters show that contrail mitigation is technically feasible at network scale without unacceptable safety or capacity impact. The main barrier is the reliability of humidity forecasts, not the optimization. Because a small fraction of flights drives most of the contrail impact, targeting them yields large benefit at moderate cost. This requires better numerical weather prediction, regulatory clarity on forecast-error accountability, and open data sharing. ... Read more

Arithmetic proficiency is an important educational concern, as the arithmetic competence of secondary school students aged 13-15 is declining. Online arithmetic puzzle games could play a role in reversing this trend. However, it remains insufficiently understood how the features of these puzzle games relate to the development of students’ arithmetic proficiency. The objective of this study was to determine how various features of online arithmetic puzzle games contribute to the development of procedural fluency, strategic competence and adaptive reasoning of secondary school students aged 13–15. A literature review was conducted to identify the game mechanics of serious games that have a positive influence on student engagement, as engagement is key in achieving the learning outcomes, and to establish the objectives of the three strands in the context of arithmetic education. Based on these findings, the scope of this study was focused on four aspects in the context of online arithmetic puzzle games: student engagement, the recognition of procedural errors, the selection and adaptation of problem-solving strategies and the evaluation of solution processes. Nine participants aged 13-15 have played a selection of six online arithmetic puzzle games across multiple sessions. The participants were observed while playing the games and interviewed after each session. The findings show that participants generally enjoyed playing the games and demonstrated a strong desire to be challenged by the puzzles. A multiplayer mode and adaptive difficulty were identified as the mechanics with the greatest positive influence on student engagement. Online arithmetic puzzle games provide limited opportunities for students to develop their ability to recognise procedural errors, as relatively few procedural errors were made while playing. In contrast, the puzzle games provide opportunities for students to develop strategic competence, as the games require students to select and adapt a variety of solution strategies in response to the challenges presented. Students generally do not evaluate their solution processes independently when solving a puzzle, suggesting that the games provide limited support for the development of adaptive reasoning. The findings of this study may help educators to make more informed decisions when selecting and implementing online arithmetic puzzle games for educational purposes. To determine whether and how online arithmetic puzzle games can contribute to the long-term development of students’ arithmetic proficiency, further research is needed. ... Read more

Planners and policy makers face complex and pressing decisions under climate change. To complement traditional planning processes, new tools and methods are needed to make rapid and integrated assessments of climate change impacts and adaptation options. In this perspective, we reflect on the potential of policy hackathons to support climate adaptation planning and policy development. Drawing on five hackathons conducted in Brazil and the Netherlands, we illustrate how these short, intensive, co-creation events enabled first-order assessments of climate risks and adaptation options, fostered interdisciplinary collaboration, and catalyzed action. We argue that policy hackathons are a promising approach for exploring urgent and emergent questions, identifying knowledge gaps, and generating momentum within planning and policy-making processes. Hackathons also support the continuous learning and innovation required by adaptation. Most importantly, within planning and policy processes, hackathons can contribute to accelerating climate action. ... Read more