Progress and Limitations in Forest Carbon Stock Estimation Using Remote Sensing Technologies: A Comprehensive Review

Forests, Год журнала: 2025, Номер 16(3), С. 449 - 449

Опубликована: Март 2, 2025

Forests play a key role in carbon sequestration and oxygen production. They significantly contribute to peaking neutrality goals. Accurate estimation of forest stocks is essential for precise understanding the capacity ecosystems. Remote sensing technology, with its wide observational coverage, strong timeliness, low cost, stock research. However, challenges data acquisition processing include variability, signal saturation dense forests, environmental limitations. These factors hinder accurate estimation. This review summarizes current state research on from two aspects, namely remote methods, highlighting both advantages limitations various sources models. It also explores technological innovations cutting-edge field, focusing deep learning techniques, optical vegetation thickness impact forest–climate interactions Finally, discusses including issues related quality, model adaptability, stand complexity, uncertainties process. Based these challenges, paper looks ahead future trends, proposing potential breakthroughs pathways. The aim this study provide theoretical support methodological guidance researchers fields.

Язык: Английский

---

Applications of Remote Sensing for Crop Residue Cover Mapping

Smart Agricultural Technology, Год журнала: 2025, Номер unknown, С. 100880 - 100880

Опубликована: Март 1, 2025

Язык: Английский

---

Toward the Optimal Spatial Resolution Ratio for Fusion of UAV and Sentinel-2 Satellite Imageries Using Metaheuristic Optimization

Advances in Space Research, Год журнала: 2025, Номер unknown

Опубликована: Фев. 1, 2025

Язык: Английский

---

Automating image classification by using multi-resolution spaceborne imagery, cloud-based machine learning algorithms and open-source software and data

Research Square (Research Square), Год журнала: 2025, Номер unknown

Опубликована: Май 7, 2025

This paper describes the extraction of critical terrestrial features from multispectral (MSS) spaceborne imagery. We present a methodology to extract spatiotemporal datasets by using machine learning algorithms, open-source software, and datasets. used Amazon Web Services sage-maker framework conduct full data life cycle (FDLC) projects automate steps acquisition mining visualization. The results demonstrate robust efficient model image analyses with spatio-temporal variables that may be useful for wide range applications. have significant implications analyze time-series imagery both near-real time other applications are driven archives.

Язык: Английский

---

SPIFFNet: A Statistical Prediction Interval-Guided Feature Fusion Network for SAR and Optical Image Classification

Remote Sensing, Год журнала: 2025, Номер 17(10), С. 1667 - 1667

Опубликована: Май 9, 2025

The problem of the feature extraction and fusion classification optical SAR data remains challenging due to differences in synthetic aperture radar (SAR) imaging mechanisms. To this end, a statistical prediction interval-guided network, SPIFFNet, is proposed for image classification. It consists two modules, propagation module (FPM) (FFM). Specifically, FPM imposes restrictions on scale factor batch normalization (BN) layer by means interval, features exceeding interval are considered redundant replaced from other modalities improve accuracy enhance information interaction. In stage, we combine channel attention (CA), spatial (SA), multiscale squeeze enhanced axial (MSEA) propose FFM fuse cross-modal cross-learning manner. counteract category imbalance, also implement weighted cross-entropy loss function. Extensive experiments three optical–SAR datasets show that SPIFFNet exhibits excellent performance.

Язык: Английский

---

Forested Swamp Classification Based on Multi-Source Remote Sensing Data: A Case Study of Changbai Mountain Ecological Function Protection Area

Forests, Год журнала: 2025, Номер 16(5), С. 794 - 794

Опубликована: Май 9, 2025

Forested wetlands in temperate mountain ecosystems play a critical role carbon sequestration and biodiversity maintenance, yet their accurate delineation remains challenging due to spectral similarity with forests anthropogenic interference. Here, we present an optimized two-stage Random Forest framework integrating 2019–2022 growing season datasets from Sentinel-1 C-SAR, ALOS-2 L-PALSAR, Sentinel-2 MSI, Landsat-8 TIRS environmental covariates. The methodology first applied NDBI thresholding (NDBI > 0.12) exclude 94% of urban/agricultural areas through masking, then implemented classifier (ntree = 1200, mtry 28) 10-fold cross-validation, leveraging 42 features including L-band HV backscatter (feature importance 47), SWIR (Band12; 57), land surface temperature gradients. This study pioneers 10 m resolution forest swamp map the Changbai Mountain wetlands, achieving 87.18% overall accuracy (Kappa 0.84) strong predictive performance (AUC 0.89). swamps showed robust classification metrics (PA 80.37%, UA 86.87%), driven by SAR’s superior discriminative power (p < 0.05). Quantitative assessment demonstrated that SAR increased canopy penetration scenarios 4.2% compared optical-only approaches, while thermal-IR reduced confusion forests. occupied 229.95 km2 (9% protected areas), predominantly transitional ecotones (720–850 elevation) between herbaceous forest. establishes multi-sensor fusion enables operational wetland monitoring topographically complex regions, providing transferable for ecosystems. dataset advances precision conservation strategies these climate-sensitive habitats, supporting sustainable development goals targets protection enhanced machine learning interpretability interference mitigation.

Язык: Английский

---

Remote sensing-based detection of brown spot needle blight: a comprehensive review, and future directions

PeerJ, Год журнала: 2025, Номер 13, С. e19407 - e19407

Опубликована: Май 22, 2025

Pine forests are increasingly threatened by needle diseases, including Brown Spot Needle Blight (BSNB), caused Lecanosticta acicola . BSNB leads to loss, reduced growth, significant tree mortality, and disruptions in global timber production. Due its severity, L. is designated as a quarantine pathogen several countries, requiring effective early detection control of spread. Remote sensing (RS) technologies provide scalable efficient solutions for broad-scale disease surveillance. This study systematically reviews RS-based methods detecting symptoms, assessing current research trends potential applications. A comprehensive bibliometric analysis using the Web Science database indicated that direct RS applications remain scarce. However, studies on other diseases demonstrated effectiveness multisource techniques symptom detection, spatial mapping, severity assessment. Advancements machine learning (ML) deep (DL) have further improved capabilities automated classification predictive modeling forest health monitoring. Climate-driven factors, such temperature precipitation, regulate distribution emerging pathogens. Geospatial analyses species (SDM) been successfully applied predict pathogen’s range expansion under changing climatic conditions. Integrating these models with monitoring enhances risk despite advancements, limited. review identifies key knowledge gaps highlights need optimize methodologies, refine models, develop warning systems management.

Язык: Английский

---

A Shdavit-Mca Block-Based Method for Remote Sensing Semantic Change Detection

Опубликована: Янв. 1, 2025

Язык: Английский

---