Green space is increasingly recognized as an important component of the urban environment. Adequate management and planning of urban green space is crucial to maximize its benefits for urban inhabitants and for the urban ecosystem in general. Inventorying urban vegetation is a costly and time-consuming process. The development of new remote sensing techniques to map and monitor vegetation has therefore become an important topic of interest to many scholars. Based on a comprehensive survey of the literature, this review article provides an overview of the main approaches proposed to map urban vegetation from high-resolution remotely sensed data. Studies are reviewed from three perspectives: (a) the vegetation typology, (b) the remote sensing data used and (c) the mapping approach applied. With regard to vegetation typology, a distinction is made between studies focusing on the mapping of functional vegetation types and studies performing mapping of lower-level taxonomic ranks, with the latter mainly focusing on urban trees. A wide variety of high-resolution imagery has been used by researchers for both types of mapping. The fusion of various types of remote sensing data, as well as the inclusion of phenological information through the use of multi-temporal imagery, prove to be the most promising avenues to improve mapping accuracy. With regard to mapping approaches, the use of deep learning is becoming more established, mostly for the mapping of tree species. Through this survey, several research gaps could be identified. Interest in the mapping of non-tree species in urban environments is still limited. The same holds for the mapping of understory species. Most studies focus on the mapping of public green spaces, while interest in the mapping of private green space is less common. The use of imagery with a high spatial and temporal resolution, enabling the retrieval of phenological information for mapping and monitoring vegetation at the species level, still proves to be limited in urban contexts. Hence, mapping approaches specifically tailored towards time-series analysis and the use of new data sources seem to hold great promise for advancing the field. Finally, unsupervised learning techniques and active learning, so far rarely applied in urban vegetation mapping, are also areas where significant progress can be expected.

Glyphosate-based herbicides are widely used in agriculture, particularly in Europe, where they are commonly applied to terminate winter cover crops prior to spring planting. However, their use remains controversial due to concerns over potential environmental and health impacts. Accurate spatial information on glyphosate application is critical for exposure assessment, yet mapping efforts using remote sensing remain limited, especially with open-access data under real-world field conditions. This study presents a remote sensing and machine learning-based methodology to identify agricultural parcels where glyphosate was applied for cover crop termination. Using Flanders, Belgium, as a case study during the 2023 and 2024 season, we compiled a training dataset consisting of parcels with crops chemically terminated with glyphosate, parcels with ongoing vegetative cover or bare soil, and parcels with crops mechanically terminated. Sentinel-2 imagery was used to derive spectral indices sensitive to chlorophyll and carotenoid content. We applied a threshold-based approach to analyze time-series spectral variations following termination. Two machine learning classifiers, Decision Tree (DT) and Random Forest (RF), were trained and evaluated using field-validated parcel data. Glyphosate-treated parcels exhibited a distinct spectral response: a rapid decline in chlorophyll-sensitive indices followed by a delayed increase in carotenoid-sensitive indices. These trends contrasted clearly with those observed in mechanically terminated or untreated parcels. The DT and RF classifiers achieved overall accuracies of 0.92 and 0.95, respectively. This study demonstrates the potential of combining open-access satellite data and machine learning to detect glyphosate applications at scale, offering a cost-effective framework for herbicide monitoring in support of environmental and public health assessments.

Context Ground-nesting bees are essential pollinators in natural and urban ecosystems, yet their nesting habitats are increasingly threatened by anthropogenic activities such as urbanization and agricultural practices. Although local ecological drivers of nesting are well understood, applying this knowledge at the landscape scale remains challenging due to limited field data, the spatial scale at which nesting aggregations occur and spatial heterogeneity in urban environments. Objectives This study uses a Bayesian framework to combine landscape, soil, vegetation, and disturbance related predictors with prior ecological knowledge to model habitat suitability for ground-nesting bees across an urban landscape. We illustrate this approach for the grey-backed mining bee (Andrena vaga) in Braunschweig, Germany. Methods We derived spatial indicators representing known nesting habitat factors (floral resource proximity, sun exposure, soil texture, vegetation density, probability of tillage, and flood risk) from remote sensing and GIS datasets. A Bayesian logistic regression model integrated these indicators with 55 observed nest locations and literature-based priors to generate habitat suitability and uncertainty maps. Results The resulting model achieved an average fivefold spatial cross-validation ROC-AUC score of 0.76 (95% posterior credible interval: 0.70–0.81) and provided spatially explicit estimates of prediction uncertainty. Sparsely vegetated areas with high insolation proved most suitable, yet all environmental variables were important. The model outputs form a decision-support layer for identifying and prioritizing suitable nesting areas in urban planning contexts, incorporating both predicted nesting probability and associated uncertainty. Conclusions By integrating ecological knowledge through informative priors, the Bayesian framework enables robust habitat suitability assessment even with limited field observations, reducing overfitting risks common in data-sparse urban ecological studies. Importantly, our approach induces spatially explicit uncertainty estimates that highlight areas where predictions are less supported by observed data. This offers critical guidance for targeted conservation planning in areas with the highest potential. The framework’s dependence on widely available remote sensing and GIS data suggests potential transferability to other ground-nesting bee species and cities, though validation across contrasting urban landscapes will be needed to confirm this.