Ingmar Weber

Satellite monitoring registers reforestation only once planted vegetation develops enough canopy to be detected as forest, typically about five years after planting. We show that reforestation activity can instead be detected around the reported planting year by targeting the intervention's disturbance-and-recovery trajectory rather than canopy closure. The framework combines Sentinel-2 composites, Dynamic World land-cover probabilities, and AlphaEarth Foundations embeddings over a three-year window spanning the year before to the year after planting. At a field-referenced reforestation site in Kieni, Kenya, it flagged 98.8\% of the crown-referenced target area, a median of four years before Dynamic World recorded a transition. It also identified interventions with a planting-year disturbance that never reached the canopy-detection threshold within the observation window. The area under the receiver operating characteristic curve (AUROC) was 0.892 on a spatially disjoint test partition, with a median of 0.715 across eight geographically withheld regions and a mean of 0.699 across three prospective planting cohorts. In a stratified national audit in Kenya, 77\% of 300 flagged patches showed reforestation activity, as did 6 of 200 unflagged patches; across the large unflagged remainder of the country, this low rate implies substantial omission. The framework therefore supports screening rather than confirming success or estimating area. These results shift satellite monitoring from detecting forest outcomes to detecting restoration activity, opening a several-year window for monitoring interventions before canopy development becomes observable at 10\,m resolution.

Satellite monitoring registers reforestation only after planted vegetation develops sufficient canopy to be detected as forest, which typically takes about five years from the point of planting. We show that reforestation activity can instead be detected around the reported planting year using a three-year satellite observation window spanning the year before through the year after planting, by targeting the intervention's disturbance-and-recovery trajectory rather than canopy closure. At a field-referenced reforestation site in Kieni, Kenya, the framework flagged 98.8\% of the crown-referenced target area, doing so a median of four years before Dynamic World recorded a transition. Because detection targets restoration activity rather than its eventual canopy outcome, it also identified interventions that showed a planting-year disturbance but did not cross the canopy-detection threshold within the observation window.The framework achieved an AUROC of 0.892 on a spatially disjoint test partition, a median of 0.715 across eight geographically withheld regions, and a mean of 0.699 across three prospective planting cohorts. It integrates Sentinel-2 composites, Dynamic World land-cover probabilities, and AlphaEarth Foundations embeddings. In a stratified national audit in Kenya, 77\% of the 300 flagged patches showed reforestation activity, as did 6 of 200 unflagged patches, a lowrate that, across the unflagged remainder of the country, impliessubstantial omission. The framework therefore supports screening rather than confirming success or estimating area. These results shift satellite monitoring from detecting forest outcomes to detecting restoration activity, providing a several-year window for monitoring interventions before canopy development becomes observable at 10\,m resolution.

Political polarization poses major societal risks, yet a globally relevant, interdisciplinary understanding remains lacking, as nearly half of all studies focus on the United States. We argue for renewed effort to bring global equity to polarization research, highlighting insights robust across countries, those unique to specific contexts, and key literature gaps. Closing these gaps means overcoming cultural and systemic barriers, including data-access restrictions and misaligned incentives. It also demands interdisciplinarity bridging traditional approaches and those addressing the polarizing role of the internet, social media, and AI. Otherwise, efforts to counter polarization, and its democratic harms, risk resting on unsuitable evidence.

Accurate and timely estimates of migrant population stocks, disaggregated by age and sex, are critical for population projections and for understanding migration dynamics. This study proposes a hierarchical Bayesian model that extends previous work by incorporating age and sex disaggregation, using data from the Labour Force Survey (LFS) and digital traces from the Facebook Advertising Platform. A Bayesian multinomial–Dirichlet–Dirichlet model harmonizes age and sex profiles from the two sources, leveraging the Rogers–Castro framework to characterize migration age schedules and utilizing the conjugate nature of Dirichlet priors to ensure computational efficiency. We illustrate the framework using data on migrant populations in the United Kingdom for 2018 and 2019, based on the two sources: the Labour Force Survey and Facebook. The analysis identifies three distinct migrant groups with differing age and sex profiles: younger Western and Southern European migrants, slightly older Central and Eastern Europeans, and a predominantly older Irish migrant population. Facebook data enhances the coverage of younger migrants, who are often underrepresented in traditional surveys, while the LFS provides broader demographic context and helps benchmark the estimates to standard population definitions. The findings highlight the utility of integrating traditional and digital data sources to address gaps in migration statistics. This framework enables more accurate disaggregation of migrant population stock data and offers a scalable, computationally efficient methodology for improving migration estimates, particularly in contexts lacking ground-truth data. The approach also yields insights into migration patterns and demographic structures, with potential applications in policy planning and demographic research.

Immigration produces diverse outcomes, with some immigrants finding happiness in an improved quality of life, while others face emotional distress from unexpected challenges. Studying these emotional experiences is challenging due to limited longitudinal data. To address this, we curated high-quality data and analyzed Twitter activities of immigrants in the United States to explore how their expressed sentiments evolve post-migration, comparing them to non-immigrants and US locals in a quasi-causal sense. Our findings revealed that migration generally boosts positive sentiments among immigrants, even when discussing the same topics, and that this effect lasts up to one-year post-migration. However, there is no conclusive evidence that migration significantly impacts negative sentiments. These patterns indicate that the migration event is associated with an expansion of positive expressed sentiments among immigrants, rather than a reduction in negative ones, suggesting a reorientation of emotional expression that reflects adaptation or opportunity recognition following migration. Additionally, age and gender play a key role, with younger population and females more likely to express positive sentiments post-migration, while linguistic ties also help foster increased positivity. These insights advance our understanding of immigrants’ emotional adaptation and highlight key demographic and social factors that shape their post-migration experiences.