Editorial: Rising stars in remote sensing 2025: advancements in time series analysis

Remote sensing time series analysis is transforming how we observe, interpret, and respond to environmental change. Rather than providing only snapshots of land, water, and vegetation conditions, dense satellite image archives now allow researchers to examine trajectories: when change occurs, how quickly systems respond, whether recovery follows disturbance, and where persistent degradation or instability may be emerging. This Research Topic, “Rising Stars in Remote Sensing 2025: Advancements in Time Series Analysis,” brings together contributions from emerging scholars whose work demonstrates the growing applied breadth of time series remote sensing across coastal, forest, reservoir, terrestrial, and arid ecosystems (Figure 1).

A central theme across the Research Topic is the movement from mapping change to diagnosing response trajectories. In (“Recording seagrass growth in Mosquito Lagoon, Florida after Hurricanes Ian and Nicole”), Insalaco et al. use semi-monthly Harmonized Landsat–Sentinel imagery and Random Forest classification to examine rapid seagrass recovery following two major hurricanes in 2022. Their work shows the value of dense temporal monitoring for capturing ecological responses that would be difficult to interpret from isolated pre- and post-event imagery. The study documents a striking post-disturbance trajectory: seagrass remained limited into early 2023 but recovered to pre-collapse levels by summer 2023, raising important questions about disturbance, recovery, water quality, and coastal ecosystem resilience.

Disturbance detection is also central to (“Mapping small-scale logging disturbances in tropical forests using Sentinel-1 time series and an extensive ground truth dataset”), where Mercier et al. address one of the persistent challenges in tropical forest monitoring: identifying small-scale forest degradation that may remain invisible in global forest-cover products. Using dense Sentinel-1 SAR time series, extensive tree-felling records, and UAV-borne lidar data, the authors develop and evaluate approaches for detecting very small logging disturbances in the Congo Basin. Their Fused-Lasso Change Detection approach represents the clearest methodological innovation in the Research Topic, showing how radar time series can help overcome cloud-cover limitations and improve detection of subtle, short-lived forest disturbances, especially where selective and artisanal logging create canopy gaps much smaller than conventional monitoring thresholds.

The Research Topic also demonstrates the importance of long-term time series for monitoring hydrological change. In (“Monitoring the dual-season hydrological dynamics of the Pong reservoir in Himachal Pradesh, India”), Sarda and Kumar use 3 decades of Landsat imagery to evaluate pre- and post-monsoon reservoir dynamics. By comparing water indices, hydrological consistency, and relative water depth, the study identifies seasonal and multi-decadal changes in the Pong Reservoir, an important Ramsar site. MNDWI proved effective for mapping reservoir water cover, and areas of high hydrological consistency declined substantially in the pre-monsoon season. This work illustrates how established spectral-index approaches, when applied consistently through time, can provide useful information for conservation planning, water-resource management, and monitoring of human-modified aquatic systems.

Vegetation dynamics in dryland environments are addressed in (“Multi-index Assessment and Spatiotemporal Dynamics of Vegetation Cover in Desert Regions of Xinjiang Based on Pixel Binary Models”). Xin et al. examine fractional vegetation cover in Xinjiang’s desert regions from 2016 to 2025, comparing EVI, NDVI, and MSAVI within a pixel binary model framework. Their results show that EVI outperformed NDVI and MSAVI for vegetation–soil discrimination and fractional vegetation cover estimation in sparsely vegetated arid landscapes. By mapping contrasting vegetation trajectories in the Junggar and Tarim Basins, this study highlights the need for regionally adapted index selection and careful validation when monitoring low-cover ecosystems where soil background effects, mixed pixels, and spatial heterogeneity complicate vegetation retrieval.

The broader conceptual significance of these applied studies is developed in (“Terrestrial ecosystems are in transition”). Wang et al. synthesize evidence that climate change, land-use change, altered disturbance regimes, and biogeochemical imbalances are eroding ecosystem resilience and increasing the risk of critical transitions. Their review emphasizes the importance of remote sensing time series for detecting early-warning signals, monitoring vegetation dynamics, and identifying ecosystems approaching thresholds. In doing so, the article provides a wider framework for interpreting the other contributions in this Research Topic: time series remote sensing is not only a mapping tool, but also a means of diagnosing resilience, vulnerability, and possible transition pathways.

Taken together, the articles in this Research Topic show the diversity of questions now being addressed through remote sensing time series analysis (Figure 1). They span marine and coastal vegetation recovery, tropical forest degradation, reservoir hydrology, dryland vegetation monitoring, and ecosystem transition theory. Most contributions apply well-established tools, including machine learning classification, spectral water and vegetation indices, pixel-based trend analysis, and multi-decadal image archives. Their value lies in applying these approaches to environmental systems where timing, persistence, and spatial heterogeneity matter. At the same time, Mercier et al.’s radar time-series analysis points toward the need for further algorithmic development for difficult monitoring problems, including cloud-prone environments, small disturbances, mixed pixels, and rapidly changing landscape features.

Several cross-cutting lessons emerge from the Research Topic. First, disturbance detection depends strongly on sensor choice and temporal density: optical time series are powerful for tracking vegetation and water dynamics, while SAR offers key advantages where clouds or canopy structure limit optical observation. Second, validation remains a central challenge. High-resolution UAV, lidar, field, and image-interpretation datasets can build confidence at local scales, but applying methods to coarser global products or to regions without detailed ground data remains difficult. Third, many studies can diagnose what changed, when, and where, but the attribution of drivers—such as rainfall variability, water quality, land management, grazing, sedimentation, or human activity—remains a major Frontier.

The work highlighted here points toward three priorities for the field: integrating AI and time series methods with ecological theory, bridging scale gaps between local validation and regional or global products, and expanding robust monitoring approaches in data-limited regions, including much of the Global South. This Research Topic therefore celebrates a set of strong individual contributions while also identifying broader challenges for the emerging generation of remote sensing scientists. As satellite archives lengthen, revisit times improve, and analytical methods continue to advance, remote sensing time series will play an increasingly central role in diagnosing environmental response trajectories and supporting timely, spatially explicit decisions for conservation, restoration, water-resource management, and climate adaptation.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. The authors used generative AI tools, including ChatGPT, to assist with language refinement and figure design. All content was reviewed and verified by the authors, who take full responsibility for the final published article.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Bijeesh Kozhikkodan Veettil, Văn Lang University, Vietnam

Reviewed by: Siham Acharki, Mohammed VI Polytechnic University, Morocco