Abstract Urban landscape modifications are proposed as a potential solution to mitigate urban heat islands (UHIs). Previous research has explored the effects of planar and vertical urban landscape characteristics on land surface temperature (LST). However, these studies predominantly assume a yet-to-be-validated linearity in these effects, and overlook the multifaceted nature of these characteristics. As a result, the nuanced impact of urban landscapes and the accurate relative importance of planar versus vertical characteristics in shaping LST remain unclear. This study, focusing on Wuhan, China, investigates the nonlinear associations between planar and vertical urban landscapes and LST using random forest models. Our findings indicate that planar landscape characteristics contribute more significantly to LST variation than vertical characteristics (73% vs. 11%). Nevertheless, despite the seemingly small contribution of vertical characteristics, their impact is nearly equivalent to that of location characteristics (12%) and substantially surpasses that of socioeconomic factors (4%). Water area proportion, building density, greenspace proportion, and floor area ratio emerge as the most influential urban landscape characteristics. Moreover, our results reveal prevalent but distinct nonlinear relationships between urban landscapes and LST, identifying four patterns: L−/inverted-L- shape, stepwise-decline−/increase, inverted-U-shaped, and N-shaped patterns. Our research provides valuable insights into multidimensional urban planning strategies for effective UHI mitigation. Taking into account the revealed nonlinearity, we also propose planning strategies to address the complex challenges of regulating urban thermal environments.

In low fertility and mortality countries, migrations are increasingly becoming the main demographic component inducing fast population turnover at different geographic scales. Spain is a privileged case of study since it has become a new international migration destination in the XXI century. In 2023, foreign born population surpassed 17% after two migration waves following economic cycles, interrupted by two shocks, the Great Recession and the Covid-19 pandemic. At the regional level, some areas concentrate the majority of immigrations and population growth. Indeed, the latest data show that while some regions reach 26% of foreign-born residents, others barely exceed 4%. These accelerated changes are key to understand how demography can shift from slow to fast under certain conditions. The objectives are twofold: first, determine the role of different demographic components in population change by NUTS-3 regions and their characteristics measured by a socioeconomic index; second, detect if a regional demographic convergence is taking place. The results show evidence of accelerated population change following the rise of international migration and the effects of the crisis and Covid-19. Overall migration is increasingly becoming the main demographic component in detriment of natural increase in most regions. Regional convergence models perform well indicating the increasing role of migrations in population change throughout all regions. Nevertheless, some shortcomings have to be pointed out regarding the lack of convergence of emigrations in the most vulnerable regions. This approach is useful tool for policymakers to identify population changes affecting the design and coverage of public services.