Large‐scale dieback of ash trees (Fraxinus spp.) caused by the fungus Hymenoscyphus fraxineus is posing an immense threat to forest health in Europe, requiring effective monitoring at large scales. In this study, a pipeline was created to find ash trees and classify dieback severity using high‐resolution hyperspectral imagery of individual tree crowns (ITCs). Hyperspectral data were collected in four forest sites near Cambridge, UK, where 422 ITCs were manually delineated and labelled using field‐measurements of species and dieback severity (for ash trees). Four algorithms, namely linear discriminant analysis (LDA), principal components analysis coupled with LDA (PCA‐LDA), partial least squares discriminant analysis (PLS‐DA) and random forest (RF), were used to build classification models for species and dieback severity classification. The effect of dark‐pixel filtering on classification accuracy was evaluated. The best performing models were then coupled with automatic ITC segmentation to map species and ash dieback distribution over 16.8 hectares of woodland. We calculated and partitioned the coefficient of variation (CV) of the reflected ash spectra to find variable wavebands associated with dieback. PLS‐DA and LDA were most accurate for classifying ITC species identifies (overall accuracy >90%), whereas RF was most accurate for classifying ash dieback severity (overall accuracy 77%). Dark pixel filtering further increased the accuracy of species classification (+6%), but not disease classification. The reflectances of narrow blue (415 nm), red‐edge (680 nm) and NIR (760 nm) bands had high CV across disease classes and should be included if multispectral imagery were to be used to monitor ash dieback. The study demonstrates the possibility of using remote sensing to forward epidemiological research by monitoring forest pathogens in landscape scales, which would allow temperate forest managers to control pathogen outbreaks, assess associated impacts and restore affected forests much more effectively.

Rain‐triggered landslides cause significant socioeconomic loss and long‐term ecological impact, disrupting topsoil and seed banks and creating unfavourable conditions for vegetation establishment. Although hazard management effectively reduces landslide risk, most management plans primarily focus on the societal impacts (loss of property and human lives), while environmental impacts related to landslide occurrence and recurrence are often overlooked. The role of different vegetation types in mitigation strategies remains uncertain due to the time and spatial constraints of traditional field study methods. Therefore, the role of vegetation in landslide restoration should be reconsidered. In this study, we investigated post‐landslide vegetation recovery patterns, analysed the duration of vegetation recovery, identified factors influencing these patterns, examined the role of forests in occurrence and recurrence probability and explored the implications of these findings for each element within an integrated management framework for landslide restoration. We utilised long‐term landslide inventory data (1924–2018) covering the entire city of Hong Kong (~1100 km²) and combined it with structural data derived from airborne LiDAR scanning in 2020 to capture landslide recovery. Our findings revealed significant differences in recovery trajectories between the main bodies and foot areas of scars, with the former taking approximately 46 years to recover and the latter recovering in about 38 years. Scar age, average wind speed, elevation and proximity to the forest were identified as key factors influencing structural recovery on scars, while the proximity to forest also regulated recovery rate of scars. We observed decreased landslide occurrence in forested areas, with recurrent landslides primarily appearing on older, unrecovered scars on barren hillsides. Synthesis and applications: Our research highlights the dynamic patterns, recovery time and drivers of vegetation recovery on landslide scars; and the higher probability of landslide occurrence in non‐forest areas and increased recurrence on bare, low vegetated scars. This information provides valuable insights for informing restoration efforts and managing landslide‐prone areas. Such knowledge is crucial for developing effective mitigation strategies and enhancing the resilience of vulnerable ecosystems facing climate change and increased frequency of extreme weather events.

Burn-area products from remote sensing provide the backbone for research in fire ecology, management, and modelling. Landsat imagery could be used to create an accurate burn-area map time series at ecologically relevant spatial resolutions. However, the low temporal resolution of Landsat has limited its development in wet tropical and subtropical regions due to high cloud cover and rapid burn-area revegetation. Here, we describe a 34-year Landsat-based burn-area product for wet, subtropical Hong Kong. We overcame technical obstacles by adopting a new LTS fire burn-area detection pipeline that (1) Automatically uniformized Landsat scenes by weighted histogram matching; (2) Estimated pixel resemblance to burn areas based on a random forest model trained on the number of days between the fire event and the date of burn-area detection; (3) Iteratively merged features created by thresholding burn-area resemblance to generate burn-area polygons with detection dates; and (4) Estimated the burn severity of burn-area pixels using a time-series compatible approach. When validated with government fire records, we found that the LTS fire product carried a low area of omission (11%) compared with existing burn-area products, such as GABAM (49%), MCD64A1 (72%), and FireCCI51 (96%) while effectively controlling commission errors. Temporally, the LTS fire pipeline dated 76.9% of burn-area polygons within two months of the actual fire event. The product represents the first Landsat-based burn-area product in wet tropical and subtropical Asia that covers the entire time series. We believe that burn-area products generated from algorithms like LTS fire will effectively bridge the gap between remote sensing and field-based studies on wet tropical and subtropical fire ecology.

Summary: Tropical cyclones (TCs) sporadically cause extensive damage to forests. However, little is known about how TCs affect forest dynamics in mountainous terrain, due to difficulties in modelling wind flows and quantifying structural changes. Typhoon Mangkhut (2018) was the strongest TC to strike Hong Kong in over 40 yr, with gusts > 250 km h⁻¹. Remarkably, the event was captured by a dense anemometer network and repeated LiDAR surveys across natural forests and plantations. We mapped long‐term mean and extreme wind speeds using CFD models and analysed corresponding changes in canopy height, which uncovered TC‐forest dynamics at unprecedented scales (> 400 000 pixels, 1108 km²). Forest height was more strongly limited by wind exposure than by background topography, a limitation attributable to a dynamic equilibrium between growth and disproportionate TC damage to taller forests. Counterintuitively, wind‐sheltered forests also suffered heavy damage. As a result, canopies of wind‐sheltered forests were more rugged, which contrasted with flat‐topped forests at wind‐exposed sites. Plantations were more susceptible to TCs compared to natural rainforests of similar stature (canopy height change −0.86 m vs −0.39 m). Our findings highlight TCs as important, often overlooked factor that fundamentally shapes forest structure and dynamics.

Succession is the process and pattern of changes in communities following a disturbance over time. This process can be disrupted by recurrent disturbances or arrested at an early successional stage even if disturbances cease. Delayed succession can hinder the natural regeneration of vegetation, weakening ecosystem development, functions and services and in severe cases, increase the risk of fires. Despite its ecological significance, delayed succession is typically analysed in a site‐ or species‐specific context, lacking the broad spatial quantification needed for effective restoration planning. Using the tropical vegetated landscapes of Hong Kong as a model system, we identified the delayed succession phenomenon using a multi‐source remote sensing approach that integrates satellite‐based time‐series imagery and airborne‐based canopy height data. We then investigated the influence of biotic, abiotic and disturbance factors on the occurrence and spatial variability of delayed succession using XGBoost. Finally, we analysed the drivers of natural regeneration and predicted the recovery trajectories of disturbed grasslands following the cessation of disturbances. Our findings indicate that delayed succession affects 4.2% (30.78 km²) of Hong Kong's vegetated area, predominantly on ridges and mid‐slopes. Biotic factors were the primary drivers, with vegetation density (<0.2 m) being the most significant. Fire frequency emerged as a critical disturbance factor; the probability of delayed succession increased sharply after more than three fire instances during the study period. Notably, vegetation density (<0.2 m) shifted from mitigating to promoting delayed succession as fire frequency increased. Upon cessation of post‐fire disturbances, natural recovery was more dependent on the proximity of natural woodland remnants than on plantations. Synthesis and applications. Our study challenges the view that fire prevention alone is sufficient for restoring degraded landscapes. We show successful natural regeneration hinges on nearby remnant woodlands acting as crucial seed sources. This research provides a remote sensing framework to quantify the risk of delayed succession where ecosystems fail to recover within a short term. This enables managers to strategically prioritize resources, focusing fire prevention on high‐risk sites while directing active restoration like planting to areas with low natural recovery potential. This approach facilitates cost‐effective, climate‐resilient restoration.